(PDF) Lecture Notes in Civil Engineering

Lecture Notes in Civil Engineering Eric Strauss Editor Proceedings of the 6th International Conference on Civil Engineering, ICOCE 2022, Singapore Innovations in Civil Engineering Lecture Notes in Civil Engineering Volume 276 Series Editors Marco di Prisco, Politecnico di Milano, Milano, Italy Sheng-Hong Chen, School of Water Resources and Hydropower Engineering, Wuhan University, Wuhan, China Ioannis Vayas, Institute of Steel Structures, National Technical University of Athens, Athens, Greece Sanjay Kumar Shukla, School of Engineering, Edith Cowan University, Joondalup, WA, Australia Anuj Sharma, Iowa State University, Ames, IA, USA Nagesh Kumar, Department of Civil Engineering, Indian Institute of Science Bangalore, Bengaluru, Karnataka, India Chien Ming Wang, School of Civil Engineering, The University of Queensland, Brisbane, QLD, Australia Lecture Notes in Civil Engineering (LNCE) publishes the latest developments in Civil Engineering - quickly, informally and in top quality. Though original research reported in proceedings and post-proceedings represents the core of LNCE, edited volumes of exceptionally high quality and interest may also be considered for publication. Volumes published in LNCE embrace all aspects and subfields of, as well as new challenges in, Civil Engineering. Topics in the series include: • Construction and Structural Mechanics • Building Materials • Concrete, Steel and Timber Structures • Geotechnical Engineering • Earthquake Engineering • Coastal Engineering • Ocean and Offshore Engineering; Ships and Floating Structures • Hydraulics, Hydrology and Water Resources Engineering • Environmental Engineering and Sustainability • Structural Health and Monitoring • Surveying and Geographical Information Systems • Indoor Environments • Transportation and Traffic • Risk Analysis • Safety and Security To submit a proposal or request further information, please contact the appropriate Springer Editor: - Pierpaolo Riva at

(Europe and Americas); - Swati Meherishi at

(Asia - except China, and Australia, New Zealand); - Wayne Hu at

(China). All books in the series now indexed by Scopus and EI Compendex database! More information about this series at https://link.springer.com/bookseries/15087 Eric Strauss Editor Proceedings of the 6th International Conference on Civil Engineering, ICOCE 2022, Singapore Innovations in Civil Engineering 123 Editor Eric Strauss Michigan State University Dimondale, MI, USA ISSN 2366-2557 ISSN 2366-2565 (electronic) Lecture Notes in Civil Engineering ISBN 978-981-19-3982-2 ISBN 978-981-19-3983-9 (eBook) https://doi.org/10.1007/978-981-19-3983-9 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd. The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore Preface 2022 6th International Conference on Civil Engineering (ICOCE 2022) was sup- posed to be held in Singapore during March 27–29, 2022. However, given the unstable situation of the epidemic, the conference was held as a fully virtual one in Zoom. We scheduled the equipment test for each presenter in advance to ensure successful delivery and avoid any unwanted technical glitches. For those who had Internet issues, a pre-recorded video was made as an alternative to a live presen- tation. The entire conference was archived for future references and streaming purpose. This conference is intended as a vehicle for the dissemination of research results on the latest advances made in the area of civil engineering, which gave an excellent opportunity for the scientists and researchers around the world to have a common platform to exchange their findings and to discuss the developments and make a base for starting collaborations in the national and international levels. It was also aimed at promoting contacts among researchers and research groups for the creation of multinational thematic and research networks, as well as promoting contacts for future collaborative joint projects within some of the international funding programs. The technical program committee put together a program, which included pre- sentations from four keynote speakers: Prof. Benjamin Sovacool (University of Sussex, UK), Prof. Prashant Kumar (University of Surrey, UK), Prof. Pen-Chi Chiang (National Taiwan University, Taiwan), and Dr. Christopher H. T. Lee (Nanyang Technological University, Singapore) and 28 papers that were accepted for publication in the ICOCE’22 conference proceedings after a rigorous review, which covers the following four specific areas: concrete structure design and analysis, structural mechanics and structural engineering, geological exploration and earthquake engineering, and urban planning and environmental assessment. On behalf of the organizing committee, we extend our heartfelt gratitude to the keynote speakers and the authors of the papers selected for presentation and pub- lication. We also extend our thanks to the international reviewers and the members of the program committee for their contribution and commitment to putting together v vi Preface a program of presentations that conference participants should find beneficial in their professional endeavor. ICOCE 2022 Conference Organizing Committees Organization Committees Conference Chairs Zongjin Li University of Macau, China Shane Snyder Nanyang Technological University, Singapore Conference Co-chairs Pen-Chi Chiang National Taiwan University, Taiwan Prashant Kumar University of Surrey, UK Program Chairs Roger Ruan University of Minnesota, USA Benjamin K. Sovacool University of Sussex, UK Publication Chair Eric Strauss Michigan State University, USA Technical Program Committees Akmal Abdelfatah American University of Sharjah, UAE Mubarak Al Alawi Sultan Qaboos University, Oman Godwin Akpeimeh University of Leeds, UK J. Amudhavel Bhopal–Indore Highway, India Irina Benedyk University at Buffalo, USA Klaas van Breugel Delft University of Technology, The Netherlands Ta-Peng Chang Civil and Construction Engineering, Taiwan Tech, Taiwan vii viii Organization Fadi Hage Chehade Lebanese University, Beirut, Lebanon Chow Ming Fai Monash University Malaysia, Malaysia Yasmin Fathy University of Cambridge, UK Hassan Hemida University of Birmingham, UK Yuner Huang The University of Edinburgh, UK Farhad Jazaei The University of Memphis, USA Fei Jin Cardiff University, Wales, UK Hue Thi Nguyen University of Wisconsin-Madison, Vietnam Kedsarin Pimraksa Chiang Mai University, Thailand Zhao Qin Syracuse University, USA Yongmin Kim University of Glasgow Singapore, Singapore Siti Fatin Mohd. Razali Universiti Kebangsaan Malaysia (UKM), Malaysia Korb Srinavin Khon Kaen University, Thailand Sudharshan N. Raman Monash University Malaysia, Malaysia Kong Fah Tee University of Greenwich, UK Linh Truong-hong Delft University of Technology, The Netherlands Chien Ming Wang The University of Queensland, Australia Yuandong Wang University of Utah, USA Yan Xiao Dalian University of Technology, China Jinhui Yan University of Illinois, USA Mijia Yang North Dakota State University, USA Xiong Yu Case Western Reserve University, Cleveland, USA Contents Concrete Structure Design and Analysis Linear Time Invariant Property of Modelling the Water Absorption Process within Cement-Based Materials . . . . . . . . . . . . . . . . . . . . . . . . . 3 Jinliang Liu, Li Song, Chenxing Cui, and Ran Liu Study of Failure Patterns in a Reinforced Concrete Beam Subjected to a Varying Range of Blast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Anita Bhatt and Sajad Ahmed Bhat Corroded Stirrups Effects on the Shear Behavior of Reinforced Concrete Slender Beams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Pier Paolo Rossi and Nino Spinella Response Control on Seismic Retrofit of Low-Rise RC Frame Using Viscous Damper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Panumas Saingam Strength and Durability Properties Empty Fruit Bunch Ash (EFBA) as Partial Replacement of Cement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 Gunalaan Vasudevan, Kribadharan Uthayasirppi, and Mohd Mawardi bin Hassim Experimental Study on the Performance of FRP Grid Reinforced ECC Composite Layer-Concrete Interface . . . . . . . . . . . . . . . . . . . . . . . 58 Weiwen Li, Jie Liu, Jiahao Zhang, Shiying Tang, Meizhong Wu, and Xu Yang Structural Mechanics and Structural Engineering Dynamic Strain Estimations of Beam Ends in Steel Moment-Resisting Frames Using Acceleration Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 Xiyang Yu and Xiaohua Li ix x Contents Studies on the Relationship Between Anchor Force of Prestressed Anchor Cable and Nonlinear Vibration of Anchor Head . . . . . . . . . . . . 87 Hao Li and Hui Cao Interrelationships of Load and Displacement of Barrette Piles for Various Interpretation Criteria Subjected to Uplift Loading . . . . . . . . . 98 Yit-Jin Chen, Anjerick Topacio, and Suneelkumar Laveti Research on the Influence of Water Horse on the Vortex Induced Resonance Response of Bridges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 Bifeng Liu and Changzhao Qian The Degradation of Avalanche Anchorage Systems . . . . . . . . . . . . . . . . 120 C. Paglia and C. Mosca Modeling and Analysis of Shanghai Central Tower . . . . . . . . . . . . . . . . 128 Yuchi Liu, Yiwen Chen, and Peitong Li From Digitized Systemization to New Era of Autonomous Materials Facilitation in Architectural Design and Actuation . . . . . . . . . . . . . . . . . 138 Mohammed Abbood and Ghada Al Slik Geological Exploration and Earthquake Engineering Analytical Investigation of the Influence of Seismicity on the Construction Cost of the R/C Load-Bearing Structure of a 15-Storey Building . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 Georgios Papadimitriou and Theodoros Chrysanidis The Influence of Buried Depth on the Stress Mode of Independent Foundation in Rock Foundation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163 Yunpeng Dong and Yin Ke Study on the Strength Characteristics of Metamorphic Slate Based on Point Load Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172 Wei Xingcan, Qi Shaoli, Yang Baowen, Wang Wenpo, and Zhong Zhibin Soıl Structure Interactıon Effects on the Seısmıc Response of Rc Structure wıth Rıgıd and Flexıble Foundatıons . . . . . . . . . . . . . . . 190 Gokaran Prasad Awadhya, Aayush Jha, Mrinal Thakur, and Sanidhya Sharma Study on Treatment of Ground Subsidence in Goaf . . . . . . . . . . . . . . . . 212 Hong Qiang Zhang, Ling Gao, and Ming Yue Feng Forward Modeling and Detection of the Potential Slip in Artificial Slope by GPR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221 Zhenjun Zhang, Luo Ye, and Gao Lv Contents xi Application of High Density Electrical Method in Karst Area . . . . . . . . 229 Hong Qiang Zhang, Zheng Bo Cao, and Wei Li Green Building Construction Implementation Barriers in the Philippines a Hierarchical Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234 Cris Edward Monjardin, Lovely Jasmin Dela Cruz, Ezekiel Esguerra, and Ondrea Nhika Dangzalan Urban Planning and Environmental Assessment Wind-Induced Dynamic Response Analysis of the Frame Lightning Rod . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 Ran Liu, Li Song, Chenxing Cui, and Jinliang Liu Research on Green Renewal Strategy Based on LID Adaptation Toolkit Selection–The Example of the North Bay Section of Macau Inner Harbour Terminal Neighborhood . . . . . . . . . . . . . . . . . . . . . . . . . 258 Linsheng Huang The Impact of Urbanisation on Catchment Discharge in Malaysia: A Case Study on Sungai Selangor Catchment . . . . . . . . . . . . . . . . . . . . 268 Mayuran Jayatharan, Andreas Aditya Hermawan, Amin Talei, and Izni Zahidi Can an Environmentally Sustainable Construction be Affordable? . . . . 281 Paulo Mendonça and Clara Vieira Climate Change and Sustainable Campus Planning: A Review of Michigan Universities’ Climate-Related Plans . . . . . . . . . . 289 AyseOzcan Buckley and Eric J. Strauss Synthesis of a Novel pH-Responsive Emulsifier Based on Dynamic Covalent Bond and Its Application in Reversible Oil-Based Drilling Fluids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313 Guoshuai Wang, Guancheng Jiang, Jun Yang, Yinbo He, and Yue Fu Removal of Aqueous Cu2+ by NaCl Modified 5A Zeolite . . . . . . . . . . . . 324 Jiantao Yang and Jianwen Wei Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 331 Concrete Structure Design and Analysis Linear Time Invariant Property of Modelling the Water Absorption Process within Cement-Based Materials Jinliang Liu1(B) , Li Song1,2 , Chenxing Cui1 , and Ran Liu1 1 School of Civil Engineering, Central South University, Shaoshan South Road 68, Changsha 410082, China

2 National Engineering Laboratory for High Speed Railway Construction, Changsha 410082, China Abstract. As a transmission medium of harmful ions, water has an important impact on the durability of concrete. The purpose of this study is to investigate the characteristics of water transfer in sound and cracked cement-based materi- als. Water absorption tests are performed considering initial saturation, damage crack, and water-cement ratio. Numerical models based on assumption of diffusion mechanism are proposed and solved by finite element method, and different consti- tutive models are employed and verified. The results show that water absorptivity is intensely affected by initial saturation and damage cracks. Besides, on the basis of diffusion equation, the linear relationship between cumulative water absorption and square root of time is found to be rigidly satisfied whatever the formula of diffusion coefficient. Keywords: Water absorption · Cement-based material · Modelling · Diffusion coefficient · Long term 1 Introduction As is known that ingression of water and harmful ions are two crucial issues that cause durability problem of cement-based materials. Freeze-thaw cycle damage caused by water will degrade the concrete strength, resulting in cracks and spalling [1–3]. On the other hand, water acts as the transport medium of aggressive ions such as chloride and sulfate ions that transfer through mechanism of diffusion or convection [4, 5]. Water itself and issues caused by it, account for most degradation problems concerning both durability and function. For concrete surface of nearly 0–20 mm, water capillary absorption in each wetting- drying cycle is regarded as the main mechanism of chloride ion transferring into a porous concrete because diffusion flux is quite small compared with capillary absorption. Hence, examining the capillary behavior in concrete is important to predict ion transfer coupling wetting-drying process. Weighing method is commonly employed to estimate water absorption characteristic of materials, and the water absorptivity can be deduced © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 3–14, 2023. https://doi.org/10.1007/978-981-19-3983-9_1 4 J. Liu et al. through water capillary test of a dried specimen [6–8]. However, the disadvantage of water absorption test is that we cannot obtain the time-varying water content distribution and wetting front within the specimen. To predict the concentration profile of chloride ion, water distribution in the process of capillary absorption is necessary. Methods of nuclear magnetic resonance, gamma rays and neutron radiography have been used to monitor the process of water transfer in real time [9–11]. The geometric quantitative information of pores such as pore size distribution was acquired from mer- cury intrusion porosimetry (micropore of smaller than 10 µm). In addition, numerical model based on diffusion mechanism is applied to simulate the water content profile with an assumed hydraulic diffusion coefficient [12, 13]. Mesoscale models considering aggregate, cement paste and interface zone were proposed to get a more detailed com- prehension instead of homogeneous medium assumption [14, 15]. Cracks and damage caused by loading provide fast paths for water absorption, and the influence on water absorptivity has been examined by tests [16–18]. In this paper, theoretical models based on continuum medium are reviewed including governing equation, initial and boundary conditions, and the saturation of concrete is chosen as unknown field variable. Then water absorption tests of dried mortar specimens, as well as cement paste cracked by temperature of 105 °C, were performed by weighing method. On the basis of test results, the values of hydraulic diffusion coefficients were estimated, and then employed in the numerical model to simulate saturation distribution within specimens. Finally, the disadvantages of known diffusion coefficient models, and of numerical model applied in long term absorption, were discussed, as well as the influence of initial saturation. 2 Models 2.1 Governing Equation As an extension of Darcy’s law, the unsaturated flow theory can be formulated as: u = −k(S)∇ψ (1) where u is the flow velocity (m/s); k(S) is the unsaturated permeability coefficient (m/s); S is the saturation of media (0–1); ψ is the height of water head (m). According to the law of mass conservation, that water flowing into media per unit time is equal to the mass increase, we can obtain ∂S + ∇ · u = 0. (2) ∂t Substituting Eq. (1) into Eq. (2), we can deduce [19] ∂S + ∇ · (−k(S)∇ψ) = 0. (3) ∂t In terms of the chain rule, Eq. (3) can be transformed into ∂S dψ + ∇ · (−k(S) ∇S) = 0 (4) ∂t dS Linear Time Invariant Property of Modelling the Water Absorption Process 5 from which we can extract the definition of hydraulic diffusion coefficient in the diffusion system as: dψ D(S) = k(S) . (5) dS Finally, the governing equation of water transferring in concrete can be inferred as: ∂S + ∇ · (−D(S)∇S) = 0. (6) ∂t 2.2 Hydraulic Diffusion Coefficient Known studies show that, in the initial time (roughly 20 h), water absorption mass is proportional to the square root of time, that is i = A · t 0.5 (7) where i is the equivalent height of water absorption (m); A is the water absorptivity (m/s0.5 ); t is the time (s). Through experimental results, we can obtain Δm i= (8) aρw where Δm is the cumulative absorption mass (kg); a is the area of face immersed into water (m2 ); ρw is the density of water (kg/m3 ). The hydraulic diffusion coefficient formulated in Eq. (5) still lacks of theoretical calculation method, and the empirical formula is often employed to estimate, that is [7] D(S) = Dd0 exp(nS) (9) where Dd0 is the diffusion coefficient of dry concrete (m2 /s); n is the material coefficient that is taken as 6 for concrete [20]. In addition, supposing that the initial saturation of concrete is zero and Eq. (9) is used in the water diffusion models, [21] deduces the theoretical relationship among A, n and Dd0 , that is A 2 1 1 1 ( )2 = Dd0 [exp(n)( − 2 ) − ( − 2 )] (10) φ n n n n where φ is the porosity of concrete. 2.3 Boundary and Initial Conditions Water contact surface of concrete can be regarded saturated because water is sufficient in the semi-infinite space. Hence, at the interface, we can define a Dirichlet boundary type as: S = 1.0 (11) 6 J. Liu et al. Concrete in service is hardly dried to the saturation of zero. However, in water absorption test, concrete is always completely dried at 105 °C so that the saturation becomes zero. Therefore, initial saturation is taken as: S = S0 (12) where S0 is equal to zero for the dry concrete, or a distribution function for the non-dry concrete. 3 Experiments 3.1 Materials and Specimens Cement type used in this test was P · I 42.5 (China standard GB175-2007), a reference cement with 100% cement clinker. Water was general tap water. The fine aggregate was natural river sand with the size of 0.08 mm to 20 mm. A series of mortar and cement paste cubic specimens with length of 100 mm, were cast with the water-cement ratio of 0.4 and 0.5, and the mix proportion was listed in Table 1. All specimens were demoulded after 24 h, and cured for 28 days in the condition of 20 °C and 90% RH. Table 1. Mix proportion and saturated water content of concrete (kg/m3 ). Specimens w/c Cement Water Sand Dry density Water loss Cement 0.4 495 198 1612 321 Cement 0.5 495 248 1503 346 Mortar 0.4 495 198 1114 2103 136 Mortar 0.5 495 248 1114 1993 172 3.2 Water Capillary Absorption Tests Three specimens of every type were used to measure the maximum water absorption. Before the capillary test, all specimens were immersed into water for seven days to reach saturated (msat ), and then dried at 105 °C for seven days (mdry ). Hence, the absorption saturation can be defined as: m − mdry S= (13) msat − mdry where m is the current mass of concrete (kg). The cement paste specimens were signif- icantly damaged by drying process, and many scaly cracks occurred on all surfaces because of dry shrinkage effect. However, for the mortar specimens, no significant damage was observed on all surfaces though certain damage exists. Linear Time Invariant Property of Modelling the Water Absorption Process 7 Accuracy of the weighing balance is 0.01 g, and indoor temperature is 26 °C during the tests. A demoulding side face was employed as the absorption surface that was immersed in water roughly 2 mm. Then the specimens were weighted after 10 min, 30 min, 2 h, 3 h, 4 h, 6 h, 1 days, 2 days, 3 days, and 4 days. The absorption surfaces should be wiped by a damp rag before weighting. After the capillary test, all specimens were again immersed into water to obtain the maximum absorption mass. For the test of non-dry specimens whose five faces were sealed by epoxy, they were place in a sealed container for 30 days after the water absorption reached the predetermined degree of saturation. The redistribution of water will contribute to a uniform initial state. 4 Results and Numerical Verification 4.1 Process of Cumulative Water Absorption Figure 1a shows the absorption process of initially dry mortar specimens. The relation- ship between cumulative absorption mass and square root of time presents two significant linear stages, and the time intersection is roughly between 6 h and 24 h. It can be seen that in the first stage of absorption, the absorption process is stable and basically the same for the specimens with same w/c. For the dry mortar specimens, the absorption mass within four hours is approximately 12–20 g; however, the cumulative absorption difference increased to approximately 20 g after four days. The absorptivity of the spec- imens with w/c of 0.5 is 1.65 times that of the specimens with w/c of 0.4, and the ratios of second slope to first are 0.2 and 0.24, respectively (see Table 2). (a) (b) Fig. 1. Water capillary absorption process of mortar specimens, (a) different water-cement ratio and (b) different saturation. 8 J. Liu et al. Table 2. Water absorptivity calculated through the test results. Specimen Saturation w/c A(m/s0.5 ) Ratio* Mortar Dry 0.4 1.01 × 10−5 0.240 Dry 0.5 1.66 × 10−5 0.200 0.65 0.4 2.00 × 10−6 0.450 0.30 0.4 4.40 × 10−6 0.580 Cement Dry 0.4 1.19 × 10−4 0.019 Dry 0.5 1.58 × 10−4 0.005 0.65 0.4 2.27 × 10−5 0.060 0.65 0.4 1.18 × 10−5 0.060 0.40 0.4 6.08 × 10−5 0.110 0.40 0.4 3.08 × 10−5 1.000 *Ratio of water absorptivity in the second stage to that in the first stage. Figure 1b shows the absorption processes of non-initially dry mortar specimens (w/c of 0.4) whose saturation were 0.65 and 0.30, respectively. We can see that the water absorptivity decreases with an increase in saturation, and the two-stage law of absorption still exists. The absorption mass within four hours drops to 2–6 g due to the initial water content. The water absorptivities decrease by 80% and 56%, respectively, for saturation of 0.65 and 0.30, and the ratios of second slope to first increase in comparison with that of dry specimens (see Table 2). On the basis of the measured results, if the water absorptivity and saturation are linearly fitted, the goodness of fit (R2 ) is 0.91. It suggests that this linear model can be applied to predict the water absorptivity under any saturation. Figure 2a shows the absorption process of cement paste specimens damaged by high temperature of 105 °C. It can be seen that the absorptivity is much greater than that of mortar specimens, reaching one order of magnitude (see Table 2). The absorption mass within four hours is 130–195 g, nearly half of the maximum water contents of 321 g and 346 g. The test results indicate that the development of micro and macro cracks will result in a significant variation in water absorptivity. Though the water absorption rate increases largely, the linear law of the first stage still remains undisturbed. Therefore, we suppose that in the first stage of water absorption the linear law should always exist regardless of the degree and method of damage. Linear Time Invariant Property of Modelling the Water Absorption Process 9 (a) (b) Fig. 2. Water capillary absorption process of cement specimens with crack damage, (a) different water-cement ratio and (b) different saturation. Figure 2b shows the water absorption of cement paste specimens (w/c of 0.4) whose initial saturation were 0.65 and 0.40, respectively. The absorptivities of the specimens is listed in Table 2, which are the same order of magnitude as that of dry mortar specimens. It can be seen that the linear law in the first stage is still valid in different saturation; however, the absorption characteristics show a great variation, for example, nearly two times difference between the parallel results, and an absorption curve with a single slope. Therefore, as both damage and saturation factors included, the water absorptivity and long-term absorption characteristic will grow more complex. For the cement paste tests, if the mean value of absorptivities is used to linearly fit the relationship between absorptivity and saturation, the goodness of fit (R2 ) is 0.92. 4.2 Verification of Numerical Results Through the numerical solution of water distribution, we can calculate the total water absorption as: Δm = wsat SdV − S0 (14) V where wsat is the saturated water content (kg/m3 ). The numerical model is valid in the first linear stage of water absorption because the hydraulic diffusion coefficient is defined according to the water absorptivity. Table 3 lists the simulated test results and the parameters employed in the numerical models. The partial differential governing equation Eq. (6) was solved by finite element method that will not be detailed in this paper. In addition, the constitutive equations and boundary conditions are presented in Sect. 2 as well, resulting in a highly nonlinear problem. 10 J. Liu et al. Table 3. Parameters used in the numerical models of different saturation. Specimen Porosity Saturation A(1e−5 m/s0.5 ) Dd0 (1e−11 m2 /s) Mortar 0.136 0.00 1.01 4.5 0.136 0.30 0.44 0.9 0.136 0.65 0.20 0.2 Cement 0.321 0.00 11.90 111.6 0.321 0.40 6.08 29.1 0.321 0.65 2.27 4.1 Figure 3 shows the verification of cumulative absorption mass, and numerical water content profiles of mortar specimens (w/c of 0.4) with initial saturation of 0, 0.3 and 0.65. It can be seen that all simulated absorption curves are completely linear; however, only the case of initially dry shows great agreement with the test results. The deviation of simulation results grows with the increase of initial saturation, which indicates that the suitability of Eq. (9) encounters challenge for the non-initially dry condition. Never- theless, it is interesting that the calculated absorption curve maintains a linear increase regardless of the initial saturation and the nonlinear hydraulic diffusion coefficient. Tech- nique of nuclear magnetic resonance (NMR) was reported to measure the water content profile in the process of water capillary [20]. However, precise water profile is not the focus of this paper; thus, only numerical results of water profiles are shown according to the mechanism models. After absorbing for an hour, the response depths of satura- tion are 6 mm, 4 mm and 3 mm, respectively, for initial saturation of 0, 0.3, and 0.65. Apparently, the maximum influence depth is nearly 12 mm even if a certainly imprecise calculation is taken into consideration for the non-initially dry conditions. In addition, we can see that the profiles will grow linear with the increase of initial saturation. Linear Time Invariant Property of Modelling the Water Absorption Process 11 Fig. 3. Verification and numerical results of mortar: (a) and (b) initial saturation of zero; (c) and (d) initial saturation of 0.3; (e) and (f) initial saturation of 0.65. 5 Discussion 5.1 Assumption of the Hydraulic Diffusion Coefficient As shown in Fig. 3, the numerical results of cumulative absorption always grow linearly with the square root of time though distribution curves are greatly different from each other. Therefore, a series of attempts were made to validate the linear law, and a constant diffusion coefficient was used instead of Eq. (9). As expected, we obtain a similar linear law, and the constant value of D is determined to be 4.3 × 10−9 m2 /s as the absorption curve coincides with the test results, shown in Fig. 4a. Consequently, for the proposed 12 J. Liu et al. numerical model including the Dirichlet boundary Eq. (11) and the governing equation Eq. (6), the relationship between total absorption mass and time must be Δm ∝ t 0.5 , regardless of the form of hydraulic diffusion coefficient. Figure 4b shows the concomitant water content profiles as the constant value of diffusion coefficient was employed. We can obtain a totally different saturation profile under the same volume integral of saturation; hence, a constitutive equation reflecting the actual mechanism is required to predict saturation profile. (a) (b) Fig. 4. Application of constant diffusion coefficient, (a) verification of cumulative absorption mass, and (b) numerical water content profiles. 5.2 Modelling the Long-Term Water Absorption Exponential and constant constitutive equations were respectively used to modelling the long-term water transfer in nearly 0–4 days, and the numerical results are illustrated in Fig. 5. It can be seen that the absorption curves of the both methods always coincide the whole time, and well agree with the test result in first stage, as shown in Fig. 5a. However, in the second stage of longer time, the numerical results show a growing deviation (roughly twofold) that cannot be neglected. Figure 5b shows the water content profiles of the second stage as a result of different assumptions of hydraulic diffusion coefficient. Because the cumulative absorption of simulation is greater than the test value, we speculate that the actual water distribution curve is slightly gentle (like the Constant ones) and the intrusion depth is smaller (like the Exponent ones). Apparently, a new theory is needed to eliminate the significant modelling deviation in the second stage if the long-term water absorption has to be taken into consideration. Linear Time Invariant Property of Modelling the Water Absorption Process 13 (a) (b) Fig. 5. Prediction of long-term water transfer, (a) cumulative water absorption, and (b) numerical water content profiles solved by exponential and constant diffusion coefficients. 6 Conclusions Considering water-cement ratio, initial saturation and damage cracks, water absorption tests of mortar and cement paste were carried out, and different numerical models were compared with the test results. The following conclusions were drawn. • Due to the initial saturation of not zero, such as 0.65 and 0.3, the water absorptivities nearly decrease by an order of magnitude for both mortar and cement paste. • The increased porosity in terms of water-cement ratio can improve water absorp- tivities, and the impact caused by large amounts of damage cracks is more significant. • The linear relationship between cumulative water absorption and square root of time is universal and firm in the first stage, regardless of initial saturation and damage degree. • For the diffusion model used in water transfer, linear relationship between cumulative water absorption and square root of time is rigidly satisfied whatever the formula of diffusion coefficient that rules the solution of saturation distribution. Acknowledgements. This research was supported by grants from the National Natural Science Foundation of China (Grant Nos. 52078492, 51778631 and U1934217), which are gratefully acknowledged. References 1. Sun, W., Zhang, Y., Yan, H., Mu, R.: Damage and damage resistance of high strength concrete under the action of load and freeze-thaw cycles. Cem. Concr. Res. 29(9), 1519–1523 (1999) 14 J. Liu et al. 2. Zhang, P., Wittmann, F.H., Vogel, M., Müller, H.S., Zhao, T.: Influence of freeze-thaw cycles on capillary absorption and chloride penetration into concrete. Cem. Concr. Res. 100, 60–67 (2017) 3. Bao, J., Xue, S., Zhang, P., Dai, Z., Cui, Y.: Coupled effects of sustained compressive loading and freeze–thaw cycles on water penetration into concrete. Struct. Concr. 22, E944–E954 (2021) 4. Polder, R.B., Peelen, W.H.: Characterisation of chloride transport and reinforcement corrosion in concrete under cyclic wetting and drying by electrical resistivity. Cem. Concr. Compos. 24(5), 427–435 (2002) 5. Chen, Y., Gao, J., Tang, L., Li, X.: Resistance of concrete against combined attack of chloride and sulfate under drying–wetting cycles. Constr. Build. Mater. 106, 650–658 (2016) 6. Hall, C., Yau, M.R.: Water movement in porous building materials—IX. The water absorption and sorptivity of concretes. Build. Environ. 22(1), 77–82 (1987) 7. Hall, C.: Water sorptivity of mortars and concretes: a review. Mag. Concr. Res. 41(147), 51–61 (1989) 8. Yang, L., Liu, G., Gao, D., Zhang, C.: Experimental study on water absorption of unsaturated concrete: w/c ratio, coarse aggregate and saturation degree. Constr. Build. Mater. 272, 121945 (2021) 9. Cid, J., Alquier, J., Crausse, P.: Study of moisture transfer in a deformable porous medium through attenuation of two different energy gamma rays. Rev. Sci. Instrum. 63(3), 2057–2064 (1992) 10. Maier, M., et al.: Mass transport in polymer electrolyte membrane water electrolyser liquid- gas diffusion layers: a combined neutron imaging and x-ray computed tomography study. J. Power Sources 455, 227968 (2020) 11. Lukić, B., Tengattini, A., Dufour, F., Briffaut, M.: Visualising water vapour condensation in cracked concrete with dynamic neutron radiography. Mater. Lett. 283, 128755 (2021) 12. Bažant, Z., Najjar, L.: Nonlinear water diffusion in nonsaturated concrete. Matériaux et Constr. 5(1), 3–20 (1972) 13. Yoo, J.-H., Lee, H.-S., Ismail, M.A.: An analytical study on the water penetration and diffusion into concrete under water pressure. Constr. Build. Mater. 25(1), 99–108 (2011) 14. Abyaneh, S.D., Wong, H., Buenfeld, N.: Computational investigation of capillary absorption in concrete using a three-dimensional mesoscale approach. Comput. Mater. Sci. 87, 54–64 (2014) 15. Sun, H.-M., Yang, W., Chen, R.-P., Kang, X.: A coarse-grained water model for mesoscale simulation of clay-water interaction. J. Mol. Liq. 318, 114085 (2020) 16. Smyl, D., Ghasemzadeh, F., Pour-Ghaz, M.: Modeling water absorption in concrete and mortar with distributed damage. Constr. Build. Mater. 125, 438–449 (2016) 17. Eghtesadi, S., Nokken, M.: Effect of cracking and improper consolidation as important con- crete defects on water absorption and electrical conductivity. J. Mater. Civ. Eng. 29(11), 04017201 (2017) 18. Li, X., Chen, S., Xu, Q., Xu, Y.: Modeling capillary water absorption in concrete with discrete crack network. J. Mater. Civ. Eng. 30(1), 04017263 (2018) 19. Richards, L.A.: Capillary conduction of liquids through porous mediums. J. Appl. Phys. 1(5), 318–333 (1931) 20. Leech, C., Lockington, D., Dux, P.: Unsaturated diffusivity functions for concrete derived from NMR images. Mater. Struct. 36(6), 413–418 (2003) 21. Lockington, D., Parlange, J.-Y., Dux, P.: Sorptivity and the estimation of water penetration into unsaturated concrete. Mater. Struct. 32(5), 342 (1999) Study of Failure Patterns in a Reinforced Concrete Beam Subjected to a Varying Range of Blast Anita Bhatt(B) and Sajad Ahmed Bhat Department of Civil Engineering, IIT Roorkee, Roorkee 247667, India

Abstract. Blast load is assumed to cause local failure, which causes progressive collapse of structure globally. A reinforced concrete beam may experience failure in various forms, i.e., shear, flexure, and combined failure. P-I (Pressure-impulse) curves based on a single degree of freedom analysis are used to describe structural response against blast load. The design codes also suggest preparing P-I curves based on SDOF to obtain structural response subjected to blast. The structural dynamic response of R.C. flexure members for local analysis and complete struc- tural frame for global analysis under a varying range of explosion scenarios was predicted by investigations. For distant blast, the model-based on flexure failure mode predicts approximately good results, while for near blast shear failure mode- based model understandably gives better results. A model based on one failure mode will not yield relatively accurate results for various blast ranges. The present study suggests that the model formulation that can capture both flexure and shear response yields an accurate response in the case of the near blast as well as distant blast, irrespective of the structure’s flexibility. Keywords: Distant blast · Near blast · Local failure · Global failure · Flexibility 1 Introduction The fundamentals of blast load and seismic design depend on dynamic behaviour and energy dissipation mechanisms [1]. Many kinds of researches were conducted in the past to assess a structure’s vulnerability subjected to blast loading. Investigations were con- ducted for R.C. structures to study their vulnerability when subjected to blast loads [2]. The natural period of vibration of structure and the higher modes play a vital role when the response of member/building against blast load is considered [3, 4]. Mindlin and Deresiewicz correctly articulated the fact that the shear stress distribution on a section depends on the mode shape of vibration [5]. They further stated that the maximum shear stress occurs at the neutral axis for the lower vibration modes, while for significantly higher modes of vibration, the shear stress is minimum at the neutral axis. Dynamic response of a structure subjected to blast loading involves non-linear inelastic material behaviour and high strain rate, thus increasing the complexity in analyzing the problem © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 15–29, 2023. https://doi.org/10.1007/978-981-19-3983-9_2 16 A. Bhatt and S. A. Bhat [2, 6]. To simplify the problem, sometimes blast loading can be represented by using a tri- angular pulse for a single degree of freedom system [7]. The traditional one-dimensional Euler-Bernoulli elastic beam theory for flexural elastic beam vibrations becomes an inad- equate model for analysis if higher modes are adopted [3]. The Euler-Bernoulli beam theory is not suitable for transverse impulsive-type loadings because the propagation velocity of the disturbance approaches infinity as its wave-length approaches zero [8]. The shearing deformations and rotary inertia become increasingly vital if higher modes are considered, thus indicating the Timoshenko beam theory formulation most suitable for analyzing structural elements subjected to blast loading. Since the structural elements experience considerable deformations under blast load- ing, the non-linear dynamic analysis must be performed, considering the high strain behaviour of materials and thus observing various modes of failure experienced by structural elements under blast loading. Another approach for analyzing the given prob- lem can be blast wave-structure interaction [9]. There are various technical manuals available for blast-resistant design, which can be adopted for guidelines [10, 11]. We can also simulate the problem via computer programming in software like ABAQUS, MATLAB, ANSYS, etc. [2, 12]. There are various mechanisms [13–15] in which R.C. structural elements can fail, namely flexural failure with formations of flexural plastic hinges [1, 7], flexure shear failure characterized by the formation of inclined tension cracks and flexural cracks [9, 16] and direct shear failure [8, 17, 18] near supports or joints of the elements that comprise the structure [19]. The present study analyzes an isolated reinforced concrete column and the reinforced concrete frame using SDOF modelling based on Euler-Bernoulli theory and continuum FEM modelling using ABAQUS software to observe local and global failure, respec- tively. Enhancing values of shear capacity and bending capacity can take care of high strain rate effects that arise during impulse loading and treating the Timoshenko beam as a viscoelastic beam. The research aims to study various numerical and analytical approaches involving highly dynamic events towards civil structure response subjected to blast loading. 2 Methodology 2.1 Analysis of Reinforced Concrete Column/Beam Subjected to Blast Loading External beams and columns are severely affected by blast loading as they are in close vicinity of the blast, thus making them primary load resisting members. Internal members are only affected by the load transmitted from the primary members. The analysis of individual reinforced concrete fixed columns subjected to blast loads is performed to observe the behaviour of individual members. The data regarding geometric/material properties of the column is summarized in Table 1 below: In the present study, a reinforced concrete column with fixed boundary conditions is adopted to observe local failure. ABAQUS was used for the modelling and analysis. The three-dimensional models of the reinforced concrete column were developed, and the reinforcement was modelled as solid circular bars, as shown in Fig. 1(a) and 1(b). Study of Failure Patterns in a Reinforced Concrete Beam Subjected 17 Table 1. Geometric/material properties of the column. Column section Material parameters Reinforcement L B H fck Ec fy Es No. of bars Dia (mm) (mm) (mm) (MPa) (GPa) (MPa) (GPa) (mm) 4100 450 400 48.5 34.8 415 200 10 25 (a) (b) Fig. 1. (a) Three-dimensional model of the column in ABAQUS. (b) ABAQUS modelling of reinforcement details. The Concrete Damage Plasticity (CDP) model was used as the material model for concrete, whereas reinforcing steel was modelled as elastic-perfectly plastic material. The grade of concrete was taken to be M48.5, and that of steel was Fe 415. The materials models are represented graphically below as Fig. 2(a) and 2(b). 18 A. Bhatt and S. A. Bhat (a) (b) Fig. 2. (a) Elastic-perfectly plastic stress-strain model for reinforcing steel of grade Fe 415. (b) Concrete Damage Plasticity (CDP) stress-strain model for M 48.5 concrete 2.2 Analysis of Reinforced Concrete Frame Subjected to Blast Loading Reinforced concrete frames may suffer significant damage when subjected to blast load- ings. A complete frame model should be analyzed under these loads to have a general idea about the damage caused by the blast at a global level. Analysis of a complete R.C. frame subjected to blast loading is considered. The frame chosen for this purpose was a three-storey plane frame with two bays, as shown in Fig. 3. Frame Details Number of storeys 3 (G +2) Number of bays 2 Clear storey height 4 m Clear bay width 5.4 m Beam 300 mm* 450 mm Three no. bars 20 mm dia. top Reinforcement Three no. bars 20 mm dia. bot- tom Column 450 mm*400 mm Reinforcement Ten no. bars 25 mm dia. Concrete M 48.5 Steel Fe 415 Fig. 3. 3D model of frame in ABAQUS. The three-dimensional model of the frame was developed in ABAQUS. The concrete elements of the frame were modelled as three-dimensional solid elements, while the Study of Failure Patterns in a Reinforced Concrete Beam Subjected 19 reinforcement steel was modelled as a truss element. The model was meshed with solid brick elements of size 100 mm. The element type used for analysis was C3D8R, which uses three degrees of freedom per node with a total of 8 nodes in the element. This element uses reduced integration for analysis with stiffness hourglass control. The section details of elements of the frame are described in Fig. 4. Fig. 4. Section details of beam and column used in the frame. The behaviour of steel reinforcement was modelled as a simple elastic-perfectly plastic stress-strain curve. As the grade of steel used is Fe 415, the yield stress is 415 MPa. The yield strain was obtained by using the elastic modulus (E) of 200 GPa. The stress- strain curve is presented in Fig. 2(a). The model used for concrete is the Concrete Damage Plasticity (CDP) model, which ensures the quasi-brittle behaviour of materials and is available in ABAQUS. The grade of the concrete used was M 48.5 (fck = 48.5 MPa). 2.3 Blast Load Calculation Numerous mathematical models were developed to estimate the blast loading approach- ing the structure. Brode introduced a formula based on the reduced distance to calculate the peak pressure) [2]. Later Newmark and Hensen established a relationship for calcu- lating maximum blast overpressure [20]. Krauthammer et al. [15] simplified the load- ing function to a simple triangular pulse. Kingery and Bulmash (1984) [21] developed complete charts based on the reduced distance for calculating all blast wave parameters. In the present study, blast pressure exerted on the column was calculated by the MATLAB program for three blast cases, as shown in Table 2 subsequently for comparing the FEM-based analysis with the SDOF-based analysis. 20 A. Bhatt and S. A. Bhat Table 2. Details of Blast load exerted on the column. Case Mass Standoff distance Peak Reflected Peak load for SDOF Duration (kg) (m) pressure (kN) (ms) (kPa) 1 100 10 830 980.1 1 2 300 10 2400 2833.04 6.04 3 600 10 4800 5667.8 10.35 The fixed boundary condition was simulated by arresting the degree of freedom for a particular set of nodes. The different nodes with arrested degrees of freedom were carefully selected to closely simulate the behaviour of the support. The loading on the column was applied as uniform pressure. The models were analyzed for loading mentioned in the aforementioned Table 2. The results corresponding to load case 2 (Peak pressure of 2400 kPa with a duration of 6.04 ms) are discussed in the subsequent section. The face of the column exposed to blast was selected, and a uniform pressure varying with time was applied to that face. The temporal loading variation is shown in Fig. 5(a) and 5(b). (a) (b) Fig. 5. (a) Blast load variation with time. (b) Blast loading as applied to the column. The load to the frame and column was applied using the CONWEP extension of ABAQUS. The load was applied to only the exposed surface of the frame. The CONWEP simulates the hemispherical blast wave and calculates the pressure function for each element separately. The program automatically calculates the arrival time for the pressure wave based on the geometry of the structure and the blast. The charge assumed for analysis was 300 kg of TNT at a distance of 10 m from the frame. The peak pressure at the middle of the ground floor column as generated by CONWEP is given in Fig. 6. Study of Failure Patterns in a Reinforced Concrete Beam Subjected 21 Pressure -Time History 1400000 Blast Pressure (Pa) 1200000 1000000 800000 600000 400000 200000 0 0 0.01 0.02 0.03 0.04 0.05 0.06 Time (s) Fig. 6. Blast pressure variation at the centre of the first column. 3 Results and Discussion Non-linear time history analysis was conducted under given loading (Table 2). The maximum displacement was recorded for the three cases. The comparison of the maxi- mum displacement from elastic-plastic SDOF idealization and continuum finite element model was done for all three load cases. The displacement time history plots based on SDOF based analysis and FEM-based analysis for case 2 are shown subsequently in Fig. 7. 0.03 FEM 0.025 SDOF Displacement (m) 0.02 0.015 0.01 0.005 0 0 1 2 3 4 5 6 7 8 t/td (Time as rao of blast duraon) Fig. 7. Displacement response of midpoint of a fixed column from elastic-plastic SDOF and FEM model. The behaviour of blast load on an isolated reinforced column is observed in detail using ABAQUS software based on FEM modelling. As the blast load approaches the structural element, it observes tension and compression damage. The column stiffness decreases as the damage in the concrete increases. The damage in the column is higher 22 A. Bhatt and S. A. Bhat at the maximum displacement. Figure 8(a) and 8(e) display the blast pressure at a time of 10 ms and 16 ms, respectively. Figure 8(b) and 8(f) depict the displacement corresponding to 10 ms and 16 ms, respectively. Figure 8(c) and 8(d) display the compression and tension damage plots for a time of 10 ms. Figure 8(g) and 8(h) display the compression and tension damage plots for a time of 16 ms. (a) (b) (c) (d) Fig. 8. (a) Blast pressure at time 10 ms. (b) Displacement at time 10 ms. (c) Damage compression at time 10 ms. (d) Damage tension at time 10 ms. (e) Blast pressure at time 16 ms. (f) Displacement at time 16 ms. (g) Damage compression at time 16 ms. (h) Damage tension at time 16 ms. Study of Failure Patterns in a Reinforced Concrete Beam Subjected 23 (e) (f) (g) (h) Fig. 8. continued 24 A. Bhatt and S. A. Bhat As the blast load approaches the structural element, both tension and compression damages were observed. Significant tension damages were observed at the outer face in the middle of the column and at the inner face of the support. 3.1 Progressive Damage The damage in the frame increases as the blast wave moves forward with time because more area is exposed to the blast pressure as time passes. The results are presented in the form of damage plots at different times obtained using FEM-based analysis. The compression damage response, tension damage response, and displacement time history were obtained corresponding to the blast pressure variation. The damage plots help identify or locate the zones where the damage is maximum for the structure. The located damage zones can be optimally reinforced to prevent collapse. The damage results can also prompt the designer to use different materials or structural components. The results of the analysis are described below. Figure 9(a) and 9(e) display the blast pressure at a time of 10 ms and 16 ms, respectively. Figure 9(b) and 9(f) depict the displacement corresponding to a time of 10 ms and 16 ms, respectively. Figure 9(c) and 9(d) display the compression and tension damage plots for a time of 10 ms. Figure 9(g) and 9(h) depict the compression and tension damage plots for a time of 16 ms. The analysis of the frame demonstrated the propagation of the blast pressure as well as the damage. Elements directly exposed to the blast were severely affected and suffered more damages. The damage started from the column exposed at the ground floor and propagated in the direction away from the blast. This analysis further emphasized the need for local analysis of exposed elements. Figure 10(a) and 10(b) display the displacement time histories for isolated fixed column and reinforced concrete frames at different floor levels, respectively. The max- imum displacement of 9.34 mm at time 0.013 s was observed for the isolated column. The maximum displacement of 9.12 mm at 0.013 s for the ground floor column, 15 mm at 0.015 s for the middle floor column, and 15.2 mm at 0.052 s for the top floor column were observed for framed structure. Figure 11(a) and 11(b) display the tension damage time history observed for an isolated fixed reinforced concrete column and reinforced concrete frame, respectively. The tension damage started at time 0.009 s, and correspondingly tensile stress 3.26e06 N/m2 was observed for the isolated column. The tension damage started at time 0.009 s, and correspondingly tensile stress 3.26e06 N/m2 was observed for the combined isolated column and frame. The analysis of the frame demonstrated the propagation of the blast pressure as well as the damage. Elements directly exposed to the blast were severely affected and suffered more damages. The damage started from the column exposed on the ground floor and propagated in the direction away from the blast. This analysis further emphasized the need for local analysis of exposed elements. Study of Failure Patterns in a Reinforced Concrete Beam Subjected 25 (a) (b) (c) (d) Fig. 9. (a) Blast pressure at time 10 ms. (b) Displacement at time 10 ms. (c) Damage compression at time 10 ms. (d) Damage tension at time 10 ms. (e) Blast pressure at time 16 ms. (f) Displacement at time 16 ms. (g) Damage compression at time 16 ms. (h) Damage tension at time 16 ms. 26 A. Bhatt and S. A. Bhat (e) (f) (g) (h) Fig. 9. continued Study of Failure Patterns in a Reinforced Concrete Beam Subjected 27 0.01 0.016 Ground Storey 0.009 0.014 Middle storey 0.008 Top storey Displacement (m) 0.012 Displacement (m) 0.007 0.006 0.01 0.005 0.008 0.004 0.006 0.003 0.004 0.002 0.001 0.002 0 0 0 0.02 0.04 0.06 0 0.02 0.04 0.06 Time (s) Time (s) (a) (b) Fig. 10. (a) Displacement time history for a fixed column. (b) Displacement time history for the frame on different floors. 1 1 0.9 0.9 0.8 0.8 0.7 0.7 Damage 0.6 0.6 Damage 0.5 0.5 0.4 0.4 Ground Storey 0.3 0.3 0.2 0.2 Middle Storey 0.1 0.1 Top Storey 0 0 0 0.02 0.04 0.06 0 0.02 0.04 0.06 Time (s) Time (s) (a) (b) Fig. 11. (a) Tension damage for a fixed column. (b) Tension damage for the frame on different floors. Figure 12(a) and 12(b) display the compression damage time history observed for an isolated fixed reinforced concrete column and reinforced concrete frame, respectively. The tension damage started at time 0.009 s, and correspondingly tensile stress 3.26e06 N/m2 was observed for the isolated column. The tension damage started at time 0.01 s, and correspondingly tensile stress 1.62e07 N/m2 was observed for the combined isolated column and frame. The compression damage in the column of the top storey is negligible. 28 A. Bhatt and S. A. Bhat 0.3 0.16 Ground 0.25 0.14 Storey 0.12 Middle 0.2 Storey Damage 0.1 Damage 0.15 0.08 0.1 0.06 0.04 0.05 0.02 0 0 0 0.02 0.04 0.06 0 0.02 0.04 0.06 Time (s) Time (s) (a) (b) Fig. 12. (a) Compression damage for a fixed column. (b) Compression damage for the frame on different floors. 4 Conclusion The dynamic response of reinforced concrete isolated column and frame subjected to blast loading was studied to observe the structure’s local and global behaviour. The response was observed using SDOF analysis (based on Euler-Bernoulli theory) and continuum finite element modelling software ABAQUS (based on Timoshenko theory). The propagation of damage for an isolated column and different elements of the frame was studied using ABAQUS. The following outcome can be derived from the research. The response of the reinforced concrete frame deviates from the traditional dynamic (seismic) response. The local element failure is predominant, and failure propagates from the elemental level to the global level. In a reinforced concrete frame, regions closest to the blast loading are most vulnerable to damage, and the damage propagation starts from this region. Additional care should be taken while designing such elements. The reinforced concrete frame and isolated reinforced concrete column observed both tension and compression damage, indicating both flexure and shear failure probability of structure. SDOF response based on a single mode of failure may not provide information regarding the actual behaviour of a structural element. Thus analysis models which can cover various modes of failure for a given instance should be preferred. This aspect still requires thorough validation. The behaviour and vulnerability of reinforced concrete frames have to be studied in detail to provide required strengthening design aspects for resisting blast loads, and will be undertaken by the authors. References 1. Smith, P.D., Hetherington, J.G.: Blast and Ballistic Loading of Structures (1994) 2. Jayasooriya, R.: Vulnerability and Damage Analysis of Reinforced Concrete Framed Buildings (2010) 3. Van Der Meer, L.J., Kerstens, J.G.M., Bakker, M.C.M.: P-I diagrams for linear-elastic cantilevered Timoshenko beams including higher modes of vibration. Heron 55, 51–83 (2010) Study of Failure Patterns in a Reinforced Concrete Beam Subjected 29 4. Bhatt, A., Bhargava, P., Maheshwari, P.: Contribution of higher modes in the dynamic response of reinforced concrete member subjected to blast. In: IOP Conference on Series Earth Envi- ronmental Science, vol. 871, p. 012003 (2021). https://doi.org/10.1088/1755-1315/871/1/ 012003 5. Mindlin, R.D., Deresiewicz, H.: Thickness-shear and flexural vibrations of a circular disk. J. Appl. Phys. 25, 1329–1332 (1954). https://doi.org/10.1063/1.1721554 6. Nassr, A.A., Razaqpur, A.G., Tait, M.J., Campidelli, M., Foo, S.: Strength and stability of steel beam columns under blast load. Int. J. Impact Eng. 55, 34–48 (2013). https://doi.org/10. 1016/j.ijimpeng.2012.11.010 7. Syrmakezis, C.A.: Introduction to Structural Dynamics (1996) 8. Ross, B.T.J.: Impulsive direct shear failure in RC slabs. J. Struct. Eng. 111, 1661–1677 (1985) 9. Ekström, J.: Concrete structures subjected to blast loading, p. 20 (2015) 10. Department of Defense of UFC 3-340-02. Structures to resist the effects of the accidental explosions. Unified Facil. Criteria. 1943 (2008) 11. IS:4991-1968: Criteria for Blast Resistant Design of Structures for Explosion Above Ground. Bur. Indian Stand, pp. 1–43 (1968) 12. Ibrahim, Y.E., Ismail, M.A., Nabil, M.: Response of reinforced concrete frame structures under blast loading. Procedia Eng. 171, 890–898 (2017). https://doi.org/10.1016/j.proeng. 2017.01.384 13. French, R., Maher, E., Smith, M., Stone, M., Kim, J., Krauthammer, T.: Direct shear behavior in concrete materials. Int. J. Impact Eng. 108, 89–100 (2017). https://doi.org/10.1016/j.iji mpeng.2017.03.027 14. Nassr, A.A., Ghani Razaqpur, A., Campidelli, M.: Effect of initial blast response on RC beams failure modes. Nucl. Eng. Des. 320, 437–451 (2017). https://doi.org/10.1016/j.nuceng des.2017.06.019 15. Krauthammer, T.: Analysis of impulsively loaded reinforced concrete structural elements-I. Theory 48, 851–860 (1993) 16. Yu, R., Zhang, D., Chen, L., Yan, H.: Non-dimensional pressure–impulse diagrams for blast- loaded reinforced concrete beam columns referred to different failure modes. Adv. Struct. Eng. 21, 2114–2129 (2018). https://doi.org/10.1177/1369433218768085 17. Wang, W., Zhang, D., Lu, F.Y., Tang, F.J., Wang, S.C.: Pressure-impulse diagram with multiple failure modes of one-way reinforced concrete slab under blast loading using SDOF method. J. Cent. South Univ. 20, 510–519 (2013). https://doi.org/10.1007/s11771-013-1513-z 18. Krauthammer, T., Astarlioglu, S.: Direct shear resistance models for simulating buried RC roof slabs under airblast-induced ground shock. Eng. Struct. 140, 308–316 (2017). https:// doi.org/10.1016/j.engstruct.2017.02.056 19. Ross, T.J.: Direct Shear Failure in Reinforced (1983) 20. Ngo, T., Mendis, P., Gupta, A., Ramsay, J.: Blast loading and blast effects on structures - an overview. Electron. J. Struct. Eng. 7, 76–91 (2007) 21. Kingery, C.N., Bulmash, G., Bulmash, G.: U.S. Army Ballistic Research Laboratory; U.S. Army Ballistic Research Laboratory. Airblast Parameters from Tnt Spherical Air Burst and Hemispherical Surface Burst; Technical Report Arbrl-Tr, 02555; U.S. Army Armament and Development Center, Ballistic Research Laboratory: Aberdeen Proving Ground, Maryland (1984) Corroded Stirrups Effects on the Shear Behavior of Reinforced Concrete Slender Beams Pier Paolo Rossi(B) and Nino Spinella Department of Civil Engineering and Architecture, University of Catania, 95123 Catania, Italy {pierpaolo.rossi,nino.spinella}@unict.it Abstract. Stirrups corrosion causes damage effects on the shear response of rein- forced concrete elements. Such damage mainly includes cracking and spalling of concrete in the cover, and reduction of the cross-sectional area of the reinforc- ing steel. The need for tools able to predict the reduction of the load-bearing capacity of reinforced concrete element has led to several formulations based on the numerical regression of available experimental results. This paper proposes a mechanical model for the prediction of the shear response of reinforced concrete slender beams with corroded reinforcement. The proposed procedure, based on the Modified Compression Field Theory, can reproduce the entire load-displacement curve. The corrosion effects on the effective beam width and rebar cross-section are taken into account. The accuracy of the proposed model is assessed using a database of sixty-two specimens. The proposed model achieves accurate results. Keywords: Shear · Corrosion · Stirrups · Beams 1 Introduction The lack of maintenance of Reinforced Concrete (RC) beams have led to deterioration of the mechanical properties of these structural elements, especially in existing bridges [1–3] and buildings [4]. Corrosion causes cracking and spalling of the concrete cover, and reduction of the cross-sectional area of the reinforcing steel [5]. These effects also depend on the type of corrosion (uniform or concentrated) of the reinforcement. Therefore, significant reduction of the ultimate strength and maximum deflection may occur. Moreover, concentrated corrosion (i.e. “pitting”) may also change the failure mode of the beam, from ductile for bending to brittle for shear [6–8]. Fer- nandez et al. [5] observed that pitting is significant for structures where the moment redistributions is achievable. Corrosion of the longitudinal reinforcing steel has largely been studied, and a large literature is available. By contrast, a few research studies are focused on corrosion of stirrups, and consequently on the reduction of the shear capacity of RC beams. Stirrups confine concrete, and then they reduce the critical shear crack width. Therefore, corrosion of stirrups significantly reduces the plasticity branch of the steel rebars and the shear strength capacity. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 30–37, 2023. https://doi.org/10.1007/978-981-19-3983-9_3 Corroded Stirrups Effects on the Shear Behavior 31 It follows that stirrups corrosion is a critical issue. It affects not only the shear strength but also the ductility of the beam. Therefore, the deformation capacity of RC beams needs to be assessed. However, most of the models proposed in the literature are able to predict only the load-bearing capacity at the Ultimate Limit State (ULS) of corroded RC beams, and, in many cases, the basic equations are empirical [9]. Recently, Cladera et al. [10] modified the Compression Chord Capacity Model (CCCM) to predict the shear strength of corrosion-damaged RC beams. Predictions were compared to the experimental results of 146 slender and deep beams failing in shear, where stirrups and/or longitudinal reinforcement were subjected to corrosion, and achieved satisfactory results. This study aims to propose a theoretical model to predict the entire load-deformation curve of corroded RC beams by extending a previously established procedure based on the Modified Compression Field Theory (MCFT) [11]. A crack element containing longitudinal and transverse smeared steel rebars is considered. Moreover, both reduced rebar cross-section of and concrete cover because of corrosion are taken into account. The proposed model allows the calculation of the strain and stress fields by respecting all the equilibrium and compatibility equations. A validation of the proposed model, against an experimental test database reported in the literature, is presented. 2 Stirrups Corrosion Structural Effects Steel rebars with uniform and/or concentrated corrosion show a reduction in the cross-sectional area (especially due to pitting) and deterioration of the geometric and mechanical properties of the concrete cover due to cracking and spalling. 2.1 Cross-Section Reduction of Steel Rebar The cross-sectional loss ratio, ηa , can be evaluated as: As0 − As φ2 − φ2 ηa = × 100 = 0 2 × 100 (1) As0 φ0 where As0 and φ 0 are the cross-sectional area and diameter, respectively, of the steel rebar before corrosion; As and φ are the cross-sectional area and diameter after corrosion. The maximum value (ηam ) along a length of the steel rebar is an estimate of pitting. It is safe to use ηam for predicting the shear strength of RC members. When only the weight loss ratio is reported, ηw = 100 × (m0 – m)/m0 (m0 = mass of a length of the steel rebar before corrosion; and m = mass after corrosion), it should been converted into cross-section loss ratios, ηam . With this aim, Cladera et al. [10] recently simplified a discontinued equation between ηam and ηw , originally introduced by Lu et al. [9], in the following form: ηam = 1.36 ηw (2) 32 P. P. Rossi and N. Spinella 2.2 Effective Beam Width Higgins et al. [12] observed that the volumetric expansion of corrosion products of steel stirrups causes spalling of the concrete cover. Thus, the external part of the beam is cracked and not able to carry stress. This corrosion-structural effect depends on stirrup spacing (sv ) and concrete cover (c). Higgins and co-authors proposed two equations, based on empirical data and theoretical computations, to estimate the effective beam width (bw,eff ). In the following, these equations are reported in the form proposed by Cladera et al. [10], presenting continuity for the case sv = 5.5(c + φ v ), where φ v is the stirrup diameter: sv bw,eff = bw − 2(c + φv ) + (3) 5.5 5.5 bw,eff = bw − (c + φv )2 (4) sv Equations (3) and (4) are for sv ≤ 5.5(c + φ v ) and sv > 5.5(c + φ v ), respectively. Moreover, the same authors suggested to reduce the beam width (bw ) to the effective value (bw,eff ) only when stirrups exhibit at least 10% of cross-section loss. 3 Proposed Model for the Shear Response of Corroded RC Beams The proposed model predicts the shear-displacement response of slender RC beams with corroded rebars. It is based on the MCFT equations, and, as assumed by other researchers in their formulations [13, 14], the flexural model is solved separately by the shear one. 3.1 Flexural Model A top axial strain (εct ) value is assumed, then the flexural cross-section strain is estimated by adjusting the longitudinal strain of bottom reinforcement (εsb ) until the internal axial load equals the external one [15]. Flexural analysis provides the bending moment (M), the curvature (χ ), and the average axial strain from flexure (εxf ) at the centroid level of the cross-section (see Fig. 1). Corroded Stirrups Effects on the Shear Behavior 33 Fig. 1. a) Lateral view and b) cross-section of the RC beam. c) Flexural and d) Shear model. 3.2 Shear Model The equilibrium conditions of MCFT require that: σcv = σc1 − τ/cotθ (5) σc2 = σc1 − τ (1/cotθ + cotθ ) (6) with σ cv = concrete stress along the vertical direction; σ c1 and σ c2 = concrete principal tensile and compressive stresses, respectively; τ = shear stress; and θ = strut angle (see Fig. 1). Clamping stress (σ v ) can be obtained as σ v = σ cv + ρ sv σ sv ; where ρ sv = Asv /(bw sv ) = geometric ratio of the transverse reinforcement; Asv is the total area and sv spacing of stirrups, respectively; and σ sv is the stirrup stress. In slender beams, the clamping stress along a cross-section enough far from the point load can be neglected [16], and then Eq. (6) allows calculation of the shear stress as follows: τ = (σc1 + ρsv σsv )cotθ = (σc1 − σc2 )cotθ/ 1 + cot 2 θ (7) Hence, the strut angle inclination θ is obtained: cot 2 θ = (−σc2 − ρsv σsv )/(σc1 + ρsv σsv ) (8) At the same time, the compatibility equations of the MCFT require that: εx + εv = ε1 + ε2 (9) cot 2 θ = (εv − ε2 )/(εx − ε2 ) (10) 34 P. P. Rossi and N. Spinella with εx and εv = horizontal and vertical strains, respectively; and ε1 and ε2 = principal tensile and compressive strains, respectively. Equations (8) and (10) return the same value. However, the two equations of the strut angle give the vertical strain (εv ), which is independent of the crack angle (θ ). In any case, elastic or plastic behavior of transverse reinforcement must be distinguished. If stirrups are elastic, the steel stress is σ sv = E s εv , and a 2nd-order equation in εv is obtained. The close-solution for εv is the follow: εv = −Be + Be − 4Ae Ce /(2Ae ) 2 (11) where coefficients Ae , Be , and C e are: Ae = ρsv Es (12) Be = σc1 +Aεv (εx − 2ε2 ) (13) Ce = σc2 (εx − ε2 ) − σc1 ε2 (14) If stirrups yield, εv is the solution of a linear equation: εv = −Cy /By (15) where coefficients By and C y are: By = σc1 + ρsv fyv (16) Cy = σc2 (εx − ε2 ) − σc1 ε2 + ρsv fyv (εx − 2ε2 ) (17) The compression strength of concrete in a diagonally cracked web is reduced as a function of the principal strain. Therefore, the following equations are used [6, 11]: ε2 ε2 2 σc2 = fce 2 − (18) εc0 εc0 fce 1 η= = ≤1 (19) fc 0.8 + 0.34(ε1 /εc0 ) where f ce = effective compressive strength of concrete; and εc0 = strain at peak stress in concrete. Cracked concrete is assumed to carry tensile stress by tension stiffening [11]: σc1 = fct / 1 + 500 ε1 (20) The tensile stress (σ c1 ) contribution to shear strength [Eq. (7)] has an upper limit (σ c1,max ) due to the capacity of cracked concrete to bridge forces across the crack sides. The maximum admissible value of the tensile stress is the minimum of the two Corroded Stirrups Effects on the Shear Behavior 35 expressions derived from equilibrium of forces across the crack in the longitudinal and transverse directions: σc1,max = min τi cotθ + ρsl fyl − σsl ; τi /cotθ + ρsv fyv − σsv (21) 0.18 fc τi = ≤ ρsv fyv − σsv − ρsl fyl − σsl sinθ cosθ (22) 0.31 + 24w dg +16 where f yl and f yv = yielding strength of the longitudinal and vertical rebars, respectively; τ i = local shear stress; d g = maximum coarse aggregate size (in mm); w = ε1 S mθ is the average crack width (in mm); and f c is in MPa. In Eqs. (21) and (22), the residual reinforcement stresses at the crack are taken as zero when the yield stress is reached. To estimate the average diagonal crack spacing, S mθ = 1/(sinθ /S ml + cosθ /S mv ), the average crack spacing along the two orthogonal directions S ml and S mv are assumed equal to the stirrups spacing (sv ) and the effective depth (d) of the cross-section, respectively. 3.3 Analytical Procedure For a generic top-strain value (εct ), a flexural analysis is performed, and values of bending moment (M), curvature (χ ) and flexural axial strain (εxf ) are obtained. For a simply supported beam, the shear force is constant along the shear span (a); thus, the applied shear stress at the critical section - away av from the support - is taken as τ flex = (M/av )/(bw d). The shear stress capacity [Eq. (7)] must be ≥ τ flex . If this is not the case, shear failure is achieved. Summarizing, the following steps are involved in the shear model: 1. Set a value of ε2 . 2. Set a value of ε1 , and then calculate η from Eq. (19) and σ c2 from Eq. (18). 3. Use the stored values of θ and εv to calculate w and σ c1 . 4. Calculate εv from Eq. (11) or (15). 5. Calculate σ sv = min{E s εv ; f yv }. 6. Calculate the new θ by Eq. (10) and σ c1,max from Eqs. (21) and (22). If σ c1 > σ c1,max , then set σ c1 = σ c1,max and return to Step 3. 7. Check if the value of ε2 obtained by Eq. (9) is equal to the assumed one. In case not, return to Step 2 and adjust ε1 . 8. Calculate τ from Eq. (7). If τ = τ flex return to Step 1 and adjust ε2 . 4 Validation of the Proposed Model A database recently collected by Cladera et al. [10] is used here to validate the proposed model. It includes 62 slender beams with corroded stirrups, which failed in shear. All the specimens have rectangular cross-sections, and the height of cross-section is less than 350 mm. Moreover, the RC beams are over-reinforced in bending (ρ l = 0.99 ÷ 3.01%). The concrete compressive strength (f cm ) is ranging between 22.5 and 36 P. P. Rossi and N. Spinella 50 MPa. The geometric percentage of stirrups (ρ v ) is between 0.14 and 0.52%. The corrosion degree of stirrups reaches the maximum value of 97.2%. The experimental versus numerical results at failure are reported in Fig. 2. They highlight that the proposed model accurately predicts the shear strength of RC beams with corroded stirrups. A mean value equal to 1.01 and a Coefficient of Variation (CoV) equal to 0.29 have been achieved for the ratio of the experimental to numerical shear strength (See Fig. 2). Fig. 2. Comparison between the experimental and numerical shear strength. 5 Conclusions This research study presents an analytical model to predict the shear behavior of RC beams with corroded stirrups. The proposed procedure take into account the structural effects of corrosion: rebar cross-section loss and effective beam width reduction due to spalling of cover. The numerical results reported in the paper show that the proposed model provides good performance in terms of shear strength prediction (mean = 1.01 and CoV = 0.29). Further investigations, also with the help of sophisticated tools like the Non-Linear Finite Elements Methods, are needed to better figure out the relationship between the inclination angle of strut and the rate of corrosion of stirrups. References 1. Colajanni, P., Recupero, A., Ricciardi, G., Spinella, N.: Failure by corrosion in PC bridges: a case history of a viaduct in Italy. Int. J. Struct. Integrity 7(2), 181–193 (2016). https://doi. org/10.1108/IJSI-09-2014-0046 2. Recupero, A., Spinella, N.: Experimental tests on corroded prestressed concrete beams sub- jected to transverse load. Struct. Concr. 20, 2220–2229 (2019). https://doi.org/10.1002/suco. 201900242 3. Recupero, A., Spinella, N.: Preliminary results of flexural tests on corroded prestressed concrete beams. Proc. Fib. Symp. (2019). Krakow, Poland Corroded Stirrups Effects on the Shear Behavior 37 4. Lo Presti, A., Recupero, A., Spinella, N.: Influence of rebar corrosion on RC frame push-over response. In: Hordijk, D.A., Luković, M. (eds.) High Tech Concrete: Where Technology and Engineering Meet, pp. 2118–2126. Springer, Cham (2018). https://doi.org/10.1007/978-3- 319-59471-2_242 5. Fernandez, I., Herrador, M.F., Marí, A.R., Bairán, J.M.: Structural effects of steel rein- forcement corrosion on statically indeterminate reinforced concrete members. Mater. Struct. 49(12), 4959–4973 (2016). https://doi.org/10.1617/s11527-016-0836-2 6. Spinella, N., Colajanni, P., Recupero, A., Tondolo, F.: Ultimate shear of RC beams with corroded stirrups and strengthened with FRP. Buildings 9(2), 34 (2019). https://doi.org/10. 3390/buildings9020034 7. Campione, G., Cannella, F.: Engineering failure analysis of corroded R.C. beams in flexure and shear. Eng. Fail. Anal. 86, 100–114 (2018). https://doi.org/10.1016/J.ENGFAILANAL. 2017.12.015 8. Recupero, A., Spinella, N., Tondolo, F.: Failure analysis of corroded RC beams subjected to shear-flexural actions. Eng. Failure Anal. 93, 26–37 (2018). https://doi.org/10.1016/j.engfai lanal.2018.06.025 9. Lu, Z.H., Li, H., Li, W., Zhao, Y.G., Dong, W.: An empirical model for the shear strength of corroded reinforced concrete beam. Constr. Build. Mater. 188, 1234–1248 (2018). https:// doi.org/10.1016/J.CONBUILDMAT.2018.08.123 10. Cladera, A., Marí, A., Ribas, C.: Mechanical model for the shear strength prediction of corrosion-damaged reinforced concrete slender and non slender beams. Eng. Struct. 247, 113163 (2021). https://doi.org/10.1016/J.ENGSTRUCT.2021.113163 11. Vecchio, F.J., Collins, M.P.: The modified compression-field theory for reinforced concrete elements subjected to shear. ACI J. Proc. 83(2), 219–231 (1986). https://doi.org/10.14359/ 10416 12. Higgins, C., et al.: Shear capacity assessment of corrosion-damaged reinforced concrete beams. Corvallis, Oregon, USA (2003) 13. Colalillo, M.A., Sheikh, S.A.: Behavior of shear-critical reinforced concrete beams strength- ened with fiber-reinforced polymer—analytical method. ACI Struct. J. 111(6), 1385–1396 (2014). https://doi.org/10.14359/51687036 14. Mostafaei, H., Vecchio, F.J.: Uniaxial shear-flexure model for reinforced concrete elements. J. Struct. Eng. 134(9), 1538–1547 (2008). https://doi.org/10.1061/(ASCE)0733-9445(200 8)134:9(1538) 15. Spinella, N.: N-M-χ interaction for arbitrary cross section under biaxial bending and axial load. Pollack Periodica 8(3), 87–100 (2013). https://doi.org/10.1556/Pollack.8.2013.3.9 16. Acevedo, A.B., Bentz, E.C., Collins, M.P.: Influence of clamping stresses in the shear strength of concrete slabs under uniform loads. J. Earthq. Eng. 13(1), 1–17 (2009). https://doi.org/10. 1080/13632460902813190 Response Control on Seismic Retrofit of Low-Rise RC Frame Using Viscous Damper Panumas Saingam(B) Department of Civil Engineering, School of Engineering, King Mongkut’s Institute of Technology Ladkrabang, Bangkok 10520, Thailand

Abstract. In recent years, seismic design and detailing requirements for buildings have considerably improved worldwide. For example, in Thailand, a new seismic design code was published in 2021, but many existing buildings do not satisfy the new code and require retrofit. The seismic retrofit is required to improve the seis- mic performance of the existing building. However, the response control method to control the target story drift ratio of the retrofitted RC buildings using the viscous damper is lack introduction. This study proposes a response control retrofit strat- egy using viscous dampers, designed using an equivalent linearization approach. A constant stiffness method is introduced to efficiently distribute the dampers along with the building height. The stiffness of the damper is equally distributed for all stories. A design example is introduced of a low-rise reinforced concrete school building in Thailand, which was damaged in the 2014 Mae Lao earthquake. Nonlinear response history analysis is used to validate the introduced method. The results indicate that the average peak story drifts ratios can be controlled within the target story drift ratio of 0.67% rad. Keywords: Response control · Seismic retrofit · Low-rise RC building · Viscous dampers 1 Introduction Thailand has historically been considered to have a low seismic hazard, and the most current existing buildings were designed to resist only gravity load. However, the seismic resistance was not considered in the designs. In recent years, the earthquake has damaged several buildings around the world. Therefore, in 2009, the Department of Public Works and Town & Country Planning of Thailand published a seismic design specification for new buildings [1], followed by a specification for seismic retrofit [2]. Shortly after the May 15, 2014 Mae Lao earthquake struck, causing extensive damage to older buildings that were constructed before the seis- mic specifications were implemented. Much of the damage was observed in reinforced concrete (RC) structures, as reported in [3, 4] including some school buildings. Figure 1a shows a typical 2-story RC school building in Thailand, which was constructed in many places in the country. The buildings are non-ductile RC moment frames with vertical © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 38–48, 2023. https://doi.org/10.1007/978-981-19-3983-9_4 Response Control on Seismic Retrofit of Low-Rise RC Frame 39 irregularities due to infill masonry walls. The 2-story building, as shown in Fig. 1b, received significant structural damage during the Mae Lao earthquake, as indicated by the severe damage to the beam-column joint at the top of the ground story columns. According to the severe damage from the Mae Lao in 2014, the Ministry of Interior of Thailand published a new seismic design code, which is a regulation in 2021 [5]. This affects that all buildings in the seismic region in Thailand are required to improve their seismic performance by strengthening or retrofitting. A conventional retrofit solution for seismically deficient reinforce concrete (RC) frames is to install a stiff shear wall [6, 7], which limits drift and ensures that the mainframe remains elastic but imposes large floor accelerations. This implies extensive nonstructural damage, as building contents and nonstructural components are unlikely to be detailed for seismic resistance in Thailand. An alternative retrofit solution is to employ energy dissipation devices to control both drift and accelerations while protecting the existing structure. Energy dissipation devices have been reported to be an effective seismic retrofit solu- tion for RC frame buildings and have been applied in practice. A retrofit design method for RC frame structures, where the buckling-restrained braces (BRBs), are installed in parallel with a supplementary elastic steel frame was introduced in [8, 9]. In addi- tion, a retrofit RC building was proposed in the study [10]. The results from the study [8–12] indicated that energy-dissipating devices can improve efficiently the seismic performance of the retrofitted RC buildings. The challenge in Thailand is that the seis- mic hazard is relatively small, with response control retrofits requiring smaller dampers installed at fewer stories than in a typical Japanese application. While still a potentially effective retrofit solution, the low demands introduce unique challenges in determining an efficient number, size and distribution of dampers, as the optimal damper type, distri- bution and design approach may be different from those countries. Although the study [9] proposed installing viscous dampers into the target building, the controlled response on seismic retrofit RC building method with a viscous damper is rarely proposed. This study introduces a response control method to retrofit the RC buildings, which is named a constant stiffness method. The method is used to design and retrofit the example 2-story building as shown in Fig. 1. The seismic performance of the existing RC building and the retrofitted RC building is investigated and compared. The design is verified through nonlinear response history analysis. Fig. 1. School building: (a) Elevation of 2-story, (b) Observed damage. 40 P. Saingam 2 Constant Stiffness Method The inelastic story force-displacement response of the bare RC frame is first obtained through pushover analysis. While the example building as shown in Fig. 1 was subjected to large drift and strength degradation due to column bending failure in Mae Lao 2014 earthquake, only the response up to the target story drift is needed for this analysis, which is set as θ tar = 1/150 (0.67% rad) to validate the proposed design method. A tri-linear degrading Takeda model is adopted to represent the existing RC frame [13] and is calibrated to match the area under the pushover curve at each story ith story. The post-yield response is assumed perfectly plastic (α 2 = 0), the yield story drift θ fy is limited to 1/100–1/300 rad, and the crack (δ fci ) to yield (δ fyi ) displacement ratio is initially set as μc = 0.1 [13], but permitted up to μc = 0.2. The yield shear force Qfyi and displacement δ fyi are then estimated, and the cracking shear force Qfci and displacement δ fci are adjusted to produce the same shear force ratio N = Qfyi /Qfci and cracked stiffness ratio α 1 = [(Qfyi − Qfci )/(δ fyi − δ fci )]/K fi at all stories, where the initial story stiffness K f0i = Qfci /δ fci . This treatment reduces the multi-story frame to a simplified representation, with the same pre-yield stiffness ratio α 1 , crack-to-yield drift ratio μc , ductility μf = δ tar /δ fyi , and secant stiffness K fμ = pK f 0 at each story. The multi-degree of freedom (MDOF) model is then reduced to an equivalent single-degree- of-freedom (SDOF) system using the equivalent height (H eq ), mass (M eq ), and stiffness (K f ) [8]. The cyclic hysteretic response of the SDOF system is shown in Fig. 2 for the cracked and yielding stages. pKf0=Kfμ Q Kf0 Q RC frame α1Kf0 RC frame Kf0 α1Kf0 pKf0=Kfμ Qfy Qfy (μc<μf<1) (μf >1) Qf K ul Kul Qfc Efe Qfc Efe Ef δfc=μcδfy μfδfy δfy δ δfc=μcδfy δfy μfδfy δ Ef Backbone (a) (b) Hysteresis loop Fig. 2. Hysteresis loops for RC frame: (a) Cracked (μc μf > 1, μf < 1) and (b) Yielding (μf > 1). The hysteretic energy dissipated by the RC frame (E f ) depends on the unloading stiffness (K ul ), with the unloading stiffness degradation parameter λ assumed as 0.4 [13]. The equivalent hysteretic damping for a constant cyclic displacement (hf μ = hf 0 + Ef /4πEfe ) is then determined from the hysteretic energy E f , strain energy E fe and intrinsic damping hf 0 is assumed to be 0.03 for RC structures. Response Control on Seismic Retrofit of Low-Rise RC Frame 41 As displacement ductility in each cycle varies when subjected to earthquake excitation, the study [14] introduced the average damping concept (Eq. 1) is employed. μtar 1 hf μ = hf 0 + hf μ − hf 0 d μ (1) μtar 1 However, for simplicity, the average equivalent damping (hfμ ) (Eq. 2) may be esti- mated from the equivalent damping of the maximum cycle (hf μ ) and a calibrated damping reduction factor (Rfμ ). The average hfμ and peak hf μ equivalent damping are shown in Fig. 3a and the corresponding reduction factors Rfμ is shown in Fig. 3b. hf μ = hf 0 + Rf μ hf μ − hf 0 (2) 0.15 1 h`fμ 0.12 hfμ 0.8 0.09 0.6 h`fμ ,hfμ Rfμ=0.4 Rfμ 0.06 0.4 0.03 0.2 0 0 0 0.4 0.8 μf 1.2 1.6 2 0 0.4 0.8 μf 1.2 1.6 2 Fig. 3. Equivalent damping reduction factor: (a) Equivalent damping and (b) Rfμ . The spectral displacement S d (T fμ ,hfμ ) of the bare RC frame is estimated from the design elastic displacement response spectrum at the secant period (T fμ ), reduced from the 5% damped spectrum using equivalent damping (hfμ ) and reduction factor proposed which was introduced as a = 25 in the study [15]. The secant period is given by Eq. 3 and uses the secant stiffness (K fu ) of the bare RC frame at the target drift. The roof drift of the bare RC frame (θ fμ ) is estimated from Eq. 4 and dampers are required if θ fμ exceeds the target story drift θ tar . Meq Tf μ = 2π (3) Kf μ Sd (Tf μ , hf μ ) θf μ = (4) Heq 42 P. Saingam The viscous (VS) dampers are velocity-dependent devices, which are effective in controlling drifts and enhancing the system energy dissipation. The VS are typically installed in series with an elastic brace element, with the assembly acting in parallel to the RC and supplemental steel frames, as indicated by Fig. 4(a). The component force-displacement relationships are shown in Fig. 4(b), where E dVS is the equivalent damping of the viscous damper, Ka the loss stiffness Eq. 8, C d the damping coefficient, K b the brace stiffness, ω the circular frequency, E e the equivalent potential energy of a total system, Ka the storage stiffness Eq. 5, and ηa the brace-damper subassembly’s loss factor Eq. 7. Cd2 Kb ω2 Ka = (5) Kb2 + Cd2 ω2 2 θf μ Ka p θtar Dh2 − p = (6) Kf γs + 1 ηa Ka Kb ηa = = (7) Ka Cd ω (a) Viscous damper Q Steel Frame EdVS Qdy Q Ka ” QSFμ + KSF μfyδfy δ μfyδfy δ (b) Fig. 4. (a) Viscous damper with the brace, RC frame and elastic steel frame model. (b) Viscous damper and elastic steel frame force-displacement model. Response Control on Seismic Retrofit of Low-Rise RC Frame 43 Cd Kb2 ω Ka = (8) Kb2 + Cd2 ω2 The required loss stiffness ratio (rdVS ) = Ka /Kf of the brace-damper subassembly is referred to as the added component and is given by for the RC frame cracking and yielding stages as shown in Eqs. 9(a) and 9(b), respectively. θf μ 2 μ (1−p) p θ −11 + 25 hf 0 + π1 · pμc +μc · Rf μ K f rdVS = a = (μc μf > 1, μf < 1) (9a) Kf 1 + 25hf 0 (γs + 1/ηa ) + (0.5x25ReqVS ) ⎛ ⎛ λ ⎞⎞ θf μ 2 pμf +μc −p μf (1+μc ) p θ − 1 ⎝1 + 25⎝hf 0 + π1 · pμf +μc · Rf μ ⎠ ⎠ Ka rdVS = Kf = (μf > 1) (9b) 1 + 25hf 0 (γs + 1/ηa ) + (0.5x25ReqVS ) Though the supplemental damping provided by the viscous dampers is velocity, rather than displacement dependent, the hysteretic damping of the RC frame still con- tributes to equivalent damping of the system, which consequently varies cycle by cycle. The reduction factor ReqVS relating the average (heq ) and peak cycle (heq ) equivalent damping is shown in Fig. 5. 1 ReqVS =0.50 3.0 0.8 2.0 1.0 0.5 0.6 Kd /K f= 0.1 ReqVS 0.4 0.2 0 0 0.4 0.8 μf 1.2 1.6 2 Fig. 5. Damping reduction factor for the system with VS dampers (ReqVS ). 3 Design Example and Validation 3.1 Design Example This section applies the constant drift method procedure to the 2-story RC school build- ings depicted in Fig. 1, which requires seismic retrofit. Thailand Seismic Design Code [5]. The newest seismic Thai code has been written based on ASCE 7-05 [16], and the design level spectral response acceleration parameters for these structures are S DS = 0.56(g) and S D1 = 0.24(g) (site class D, Phan, Chiang Rai), approximately half of the seismic demands in Japan. 44 P. Saingam The story masses of the 2-story building are 266 and 172 tons at the first and roof stories, respectively, and the fundamental period of the bare RC frame is 0.59 s in both the longitudinal and transverse directions. Pushover curves and calibrated tri-linear Takeda models for the 2-story building are shown in Fig. 6 for the first story and second story, respectively. Figure 6a and 6b illustrate the story shear to story displacement of the 1st story for longitudinal and transverse directions, respectively. Figure 6c and 6d show the story shear to story displacement of the 2nd story for longitudinal and transverse directions, respectively. Structural properties of the bare SDOFRC structures are summarized in Table 1. The ratios of the area under the pushover curves and tri-linear model (Apushover /Atri ) are close to 1.0 at each story, indicating a good fit. Table 2 shows damper distributions for 2-story building model. 500 500 H1 = 3450 mm H1 = 3450 mm Qfy=442 Qfy=459 400 1F Pushover curve 400 1F Pushover Curve Story shear (kN) Story shear (kN) 1F Takeda Model 1F Takeda Modal 300 300 θtar=1/150 rad. θtar=1/150 rad. (0.67%) 200 α1Kf1 = 0.6Kf1 200 α1Kf1 = 0.60Kf1 Qfc=130 θfy=1/275 rad. (0.36%) Qfc=135 100 θfy =1/275 rad. (0.36%) 100 Kf1 =51.90 kN/mm Kf1 = 53.76 kN/mm 0 δfc=2.51 δ =12.55 δtar=23 0 0 10 fy 20 30 40 50 60 70 δfc=2.51 δfy=12.55 δtar=23 0 10 20 30 40 50 60 70 Stroy displacement (mm) Story displacement (mm) (a) Longitudinal direction for 1st story (b) Transverse direction for 1st story 500 500 H2 = 3550 mm H2 = 3550 mm Qfy=442 400 400 2F Pushover Curv Qfy= 357 2F Takeda Modal Story shear (kN) Story shear (kN) 300 2F Pushover cur 300 2F Takeda Mode θtar =1/150 rad. (0.67%) 200 θtar=1/150 rad. (0.67%) 200 α1Kf2 = 0.60Kf2 Qfc =105 α1Kf2 = 0.60Kf2 Qfc=130 100 θfy=1/275 rad. (0.36%) θfy =1/275 rad (0.36%) 100 Kf2 =40.67 kN/mm Kf2 =50.41 kN/mm 0 δfc=2.58 δfy=12.91 δtar=23.67 0 δfc=2.58 δfy=12.91 δtar=23.67 0 10 20 30 40 50 60 0 10 20 30 40 50 60 7 Story displacement (mm) Story displacement (mm) (c) Longitudinal direction for 2nd story (d) Transverse direction for 2nd story Fig. 6. Pushover curve and tri-linear model of the 2-story building (a) Longitudinal direction for the first story, (b) Transverse direction for the first story, (c) Longitudinal direction for the second story, (d) Transverse direction for the second story. Table 1. Characteristic of bare RC frame Direction μf K f0 H eq M eq μc α1 T fμ K fμ kN/mm mm ton sec kN/mm Rfμ = 0.6 and θ tar = 1/150 rad Longitudinal 1.83 46.6 4510 412 0.20 0.60 0.97 17.3 Transverse 1.83 49.4 4469 415 0.20 0.60 0.95 18.3 Response Control on Seismic Retrofit of Low-Rise RC Frame 45 Table 2. Damper distributions for 2-story building model Direction Story Ka /Kf K fi Kai Kb heq Cd ni kN/mm kN kN/mm kN · s/mm Longitudinal 2 0.11 40.7 4.4 68.2 0.104 0.36 2 1 51.9 5.7 68.7 0.46 2 Transverse 2 0.09 50.4 4.5 68.2 0.097 0.36 1 1 53.8 4.8 68.7 0.39 1 3.2 Validation of Effectiveness of Constant Stiffness Method To validate the retrofit designs, nonlinear response history analyses were performed, targeting the design acceleration response spectrum described earlier (S DS = 0.56(g) and S D1 = 0.24(g)). Two suites of ground motions were used, reflecting common practice in Japan and the US, which the Thai code is based upon. First, a suite of four earthquake ground motions were spectrally matched, consisting of El Centro NS (1940), JMA Kobe NS (1995), TAFT EW (1925), and Hachinohe NS (1968). The duration of four observed waves was 30 s for each wave and compared to the design spectrum in Fig. 7(a). Additionally, a suite of 11 scaled single component records were selected from the PEER NGA2 ground motion database 2 (Fig. 7(b)). Scaling was conducted over a target period range of 0.2T 1,min and 1.5T 1,max following ASCE 7-16 [17], where T 1,min and T 1,max are the minimum and maximum fundamental periods from the two models, resulting in a target period range of 0.1 to 2 s. Records were limited to strike-slip events with magnitudes of 6 ≤ M w ≤ 7.5 within 20 km and on soil class D (180 ≤ V s,30 ≤ 360 m/s), consistent with the dominant seismic hazard in the Chiang Rai province and local site conditions. Scale factors varied from 0.5 to 2.0, and the average spectrum matches or exceeds the target spectrum over the range of interest. While the average acceleration response spectra are similar for both suites, the average displacement spectra exceed the design spectra by a relatively large margin for the scaled suite at periods greater than 1 s, as shown in Fig. 7 while the displacement spectra are shown in Fig. 7(c). The peak story drift ratios of the existing RC frame and the retrofitted models using viscous dampers are shown in Fig. 8 for the 2-story building. Only the longitudinal direction is shown here as the response is similar in the two orthogonal directions. Drift is concentrated at the first story, exceeding the target story drift angle and matching the observed damage experienced during the Mae Lao earthquake. Figure 8(a) and Fig. 8(b) show the peak story drift ratios of the existing RC frame and the retrofitted with viscous dampers, respectively. Using the spectrally matched suite, adding dampers in proportion to the RC frame stiffness using the constant stiffness method improves the seismic performance of the retrofitted building. The second story drift under all ground motions is 0.2% for the design using the constant stiffness method (Fig. 8(b)) but increases to 0.67% to 0.78% at the first story. The scaled ground motions produce a similar average drift distribution for the 2-story building (Fig. 8) but exhibit greater record-to-record variability. 46 P. Saingam 1000 1000 800 800 Acceleration (cm/s 2 ) Acceleration (cm/s2 ) 600 600 400 400 200 200 0 0 0 0.5 1 1.5 2 0 0.5 1 1.5 2 Period (s) Period (s) (a) Matched suite (b) Scaled suite 25 h=5 Average(Spectral spectrum ,Sd (cm) 20 Average matched suite) Displacement (Scaled suite) 15 Pseudo disp 10 design spectra 5 0 0 0.5 1 1.5 2 Period (s) (c) Displacement spectra Fig. 7. 5% damped response spectra: (a) Matched suite, (b) Scaled suite and (c) Displacement spectra. 2 2 θtar = 1/150 (0.67%) θtar = 1/150 (0.67%) Average (Matched suite) Average (Matched suite) Average (Scaled suite) Story Story Average (Scaled suite) 1 1 0.0 0.5 1.0 1.5 0.0 0.5 1.0 1.5 Peak story drift (%) Peak story drift (%) (a) Existing RC frame (b) Retrofitted building Fig. 8. Peak story drifts of the 2-story building in longitudinal direction: (a) Bare RC frame, (b) Retrofitted building. Response Control on Seismic Retrofit of Low-Rise RC Frame 47 4 Conclusions A response control retrofit based on equivalent linearization, which is called a constant stiffness method, was introduced to assign an efficient damper distribution. The peak story drift ratios of the introduced method were compared to the existing RC frame on the example RC building. The results indicated that assigning viscous damper in proportion to the bare RC frame stiffness using the constant stiffness method improved the seismic performance of the retrofitted building. In addition, the average peak story drift ratios from both matched and scaled suites can be controlled within the target story drift ratio of 0.67% rad. Further study should apply and investigate the performance of the proposed retrofit method to taller RC buildings. References 1. Department of Public Works and Town & Country Planning (DPT). Thailand Seismic Design Specification (2009) 2. Department of Public Works and Town & Country Planning (DPT). Strengthening Buildings Recommendation Specification (2014) 3. Lukkunaprasit, P., et al.: Performance of structures in the Mw 6.1 Mae Lao earthquake in Thailand on May 5, 2014 and implications for future construction. J. Earthq. Eng. 20, 219–242 (2015) 4. Ornthammarath, T., Warnitchai, P.: 5 May 2014 MW 6.1 Mae Lao (Northern Thailand) earth- quake: interpretations of recorded ground motion and structural damage. Earthq. Spectra 32, 1209–1238 (2016) 5. Ministry of Interior. Thailand Seismic Design Code (2021) 6. Canbay, E., Ersoy, U., Ozcebe, G.: Contribution of reinforced concrete infills to seismic behavior of structural systems. ACI Struct. J. 100, 637–643 (2003) 7. Foutch, D.A., Hjelmstad, K.D., Calderon, E.D.V., Gutierrez, E.F., Downs, R.E.: The Mexico earthquake of September 19, 1985: case studies of seismic strengthening for two buildings in Mexico City. Earthq. Spectra 5, 153–174 (1989) 8. Sutcu, F., Takeuchi, T., Matsui, R.: Seismic retrofit design method for RC buildings using buckling-restrained braces and steel frames. J. Constr. Steel Res. 101, 304–313 (2014) 9. Lee, D., Taylor, D.P.: Viscous damper development and future trends. Struct. Des. Tall Build. 10, 311–322 (2001) 10. Saingam, P., et al.: Composite behavior in RC buildings retrofitted using buckling-restrained braces with elastic steel frames. Eng. Struct. 219, 110896 (2020) 11. Di Sarno, L., Manfredi, G.: Experimental tests on full-scale RC unretrofitted frame and retrofitted with buckling-restrained braces. Earthq. Eng. Struct. Dyn. 41, 315–333 (2012) 12. Saingam, P., Matsuzaki, R., Nishikawa, K., Sitler, B., Terazawa, Y., Takeuchi, T.: Experimental dynamic characterization of friction brace dampers and application to the seismic retrofit of RC buildings. Eng. Struct. 242, 112545 (2021) 13. Takeda, T., Sozen, M.A., Norby Nielsen, N.: Reinforced concrete response to simulated earthquakes. J. Struct. Div. 96(12), 2557–2573 (1970). https://doi.org/10.1061/JSDEAG.000 2765 14. Newmark, N.M., Rousenblueth, E.: Fundamentals of Earthquake Engineering. Prentice-Hall Inc. (1971) 15. Kasai, K., Ito, H.: Passive control design method based on tuning of stiffness, yield strength, and ductility of elasto-plastic damper. J. Struct. Constr. Eng. AIJ 595, 45–55 (2005) 48 P. Saingam 16. American Society of Civil Engineers (ASCE). Minimum Design Loads for Buildings and Other Structures 2005 (ASCE/SEI 7-05) 17. American Society of Civil Engineers (ASCE). Minimum Design Loads for Buildings and Other Structures 2016 (ASCE/SEI 7-16) Strength and Durability Properties Empty Fruit Bunch Ash (EFBA) as Partial Replacement of Cement Gunalaan Vasudevan1(B) , Kribadharan Uthayasirppi1 , and Mohd Mawardi bin Hassim2 1 Department of Construction Management, Faculty of Built Environment, Tunku Abdul Rahman University College Malaysia, Kuala Lumpur, Malaysia

2 Jurutera Awam, Cawangan Jalan, Ibu Pejabat JKR Malaysia, Kuala Lumpur, Malaysia Abstract. This research showed the results of experiments evaluating the use of empty fruit bunch ash (EFBA) from the oil palm industry as partial replacement for ordinary Portland cement. Research on the reuse of waste materials in the concrete industry has been quite intensive in the past decade. The objective of this research is to identify the performance of EFBA as a partial cement replacement in the production of concrete. EFBA of various amounts, namely 5%, 10%, 15% and 20% by volume, was added as a replacement for ordinary Portland cement. The results showed that EFBA concrete greatly improved the compressive and flexural strength of concrete. The rate of water absorption of EFBA concrete was reduced as EFBA filled up the existing voids, making it more impermeable. However, the compressive strength of the EFBA concrete decreases gradually when the amount of EFBA increased. It can be concluded that the optimum percentage of EFBA as a partial cement replacement is 15%. In this direction, an experimental investigation of ultrasonic pulse velocity, carbonation test, compressive strength, flexural strength and water absorption was undertaken to test the performance of EFBA and admixtures as partial replacement for cement in concrete. Keywords: Empty fruit bunch ash · Admixture · Strength and durability concerete properties 1 Introduction Oil palm trunks and oil palm fronds represent about 75% of those wastes and that they are left rotten within the plantations for mulching and nutrient exercise. The remaining 25% consisting of palm kernel shells, monocarp fibre and empty fruit bunches are dried and utilized palm oil mills or power plants to generate electricity through combustion at 800–1000 °C [1]. Burning of oil palm by-products results in the formation of oil palm ash (or palm oil fuel ash) that is disposed to open fields because it is useless as fertilizer © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 49–57, 2023. https://doi.org/10.1007/978-981-19-3983-9_5 50 G. Vasudevan et al. because of its less nutrient worth. The ash will simply get carried by wind and end in air pollution on a moist day inflicting traffic hazards conjointly potential of health hazards resulting in bronchi and respiratory organ diseases. Therefore, a way of using Empty Fruit Bunch Ash (EFBA) as a possible replacement for cement should be projected to cut back the environmental result of cement and therefore the value to provide cement [2]. Concrete is one of the highly used constructional material in the world. Annually almost 2.9 billion tons of cement are produced and at this pace, it is expected to reach 4 billion tons by 2020. Clinker is the main ingredient in cement which is obtained by heating up limestone which involves burning of fossil fuel that accounts for 40–50% of emission. Additionally, decomposing of limestone in this process accounting for the remaining 50– 60% which in combined accounts for 5–7% of global CO2 . The combustion process in producing cement mainly contributes to the emission of CO2 which is a major greenhouse gas (GHG). Build-up of large amount of GHG is the primary cause of global warming. Consequently, results in severe climate changes such as sea-level rise from thermal expansion and ocean acidification from increased absorption of CO2 which then pours down as acid rain affecting the integrity of modern structures [3]. Such issues have raised concerns on environment where it has become a necessity to come up with alternative cementitious material for concrete construction. 2 Methodology The Fig. 1 show conceptual framework to determine the performance of the concrete that contain the Empty Fruit Bunch Ash (EFBA) by preparing concrete cube sample and tested to obtain some of the basic engineering properties. The concrete mix design is done by systematic analysis and chooses the proportion of the ingredient to use the concrete mix to produce an economical concrete and also with strength that desired when the cube is hardened. The variables which can be controlled are water cement ratio, maximum aggregate size, aggregate grading and use of admixture. Strength and Durability Properties Empty Fruit Bunch Ash (EFBA) 51 Fig. 1. Conceptual framework. 3 Result and Discussion 3.1 Sieve Analysis Based on Fig. 2, show that as for Empty Fruit Bunch Ash (EFBA), the dried EFBA was sieved through a 150 um sieve to remove coarse and undesired particles which were incompletely burnt during combustion. The EFBA used had a specific gravity and Blaine’s surface area of 1.88 and 506 m2 /kg, respectively. Normal mining sand with sizes between 0.3 and 5 mm was used as fine aggregate in this investigation whereas the water absorption, specific gravity and fineness modulus of the sand were found as 0.81%, 2.79 and 2.88, respectively [4]. As for the coarse aggregate, an aggregate size of between 4.7 mm–10 mm were used which had a water absorption, specific gravity and bulk density of 0.74%, 2.65 and 1455 kg/m3 , respectively. 52 G. Vasudevan et al. Fig. 2. Sieve analysis. 3.2 Compressive Strength Test Based on Fig. 3, the graph indicates a linier increase in compressive strength from 0% of EFBA to 15% of EFBA for all 3 curing days with 15% EFBA cured for 28 days sample obtaining the highest reading of 34.85 MPa.Upon curing for 28 days, the compressive increased even more even for the controlled sample which obtained 26.22 MPa. As for 5%, 10% and 15%, the reading also increased gradually to 28.88 MPa, 32.98 MPa and 34.85 MPa, respectively. The increase of compressive strength to that of the controlled sample for 5%, 10% and 15% is about 10.1%, 25.8% and 32.9%, respectively. In short, the increasing trend of the compressive strength linier to the increasing percentage of EFBA obtaining up to twice the compressive strength compared to controlled sample Fig. 3. Compressive strength test. Strength and Durability Properties Empty Fruit Bunch Ash (EFBA) 53 Fig. 4. Results of UPV indicates that EFBA is equivalent to cement in acting as a binder. EFBA forms a stronger bond when added with cement up to a certain percentage. 3.3 Ultrasonic Pulse Velocity (UPV) Based on Fig. 4 the UPV values are increasing gradually with the addition of EFBA up to 15% for all 7, 14 and 28 days of curing. Beyond 15% which is at 20% of EFBA replacement, the UPV value reduced but not lower than the controlled sample which indicates that the pozzolanic material still makes a significant difference. At 28 days of curing, the UPV value recorded 30.52 m/s for the controlled sample which was the highest amongst the 3 curing days sample. The UPV value for 5%, 10% and 15% gradually increased to 36.95 m/s, 40.52 m/s and 45.6 m/s or about 21.1%, 32.8% and 49.4% of that of the controlled sample, respective but then reduces to 41.44 m/s for 20% EFBA yet still higher than the controlled sample value. Based on the result, it indicates that the test run for 7 days of cured sample shows the highest average UPV value amongst the other curing days. This could be due to the ongoing pozzolanic reaction which forms pores within the sample which affects the velocity of pulse travelling through the sample. 3.4 Flexural Strength Test Based on the Fig. 5, the flexural strength of the samples increased with curing time but decreases with an increase in EFBA replacement beyond 15%. When EFBA is ground to a reasonably high fineness, the rate of flexural strength gain of the samples were significantly improved due to the hydration reaction, nucleation effect, packing effect and pozzolanic reaction [4]. These results conclude that EFBA with high fineness can be used as a good pozzolan in cement-based materials and can be used to replace OPC up to 15%. At 28 days, the controlled sample obtained 2.8 MPa flexural strength or about 21.7% increase compared to the controlled sample of 7 days. The flexural strength for 5%, 10% and 15% of EFBA increased to 2.9 MPa, 3.1 MPa and 3.2 MPa or about 3.6%, 10.7% and 14.3%, respectively. At 20% EFBA, the pattern is still the same as other curing days. 54 G. Vasudevan et al. Fig. 5. Flexural strength test. 3.5 Rebound Hammer Concrete Test The Fig. 6, show at the 28 days of curing, again the trend is the same as 7 days and 14 days of curing but record the best or highest average result compared to other samples. The controlled sample obtained a total of 21 MPa of force which is the highest amount recorded amongst the controlled samples. As the EFBA replacement are increased to 5%, 10% and 15%, the value obtained increased to 25 MPa, 26 MPa and 28 MPa or about 19%, 23.8% and 33.3% of that of the controlled sample. Fig. 6. Rebound hammer test. 3.6 Microstructures of Empty Fruit Bunch Ash (EFBA) The apparent morphologies of the EFBA examined by SEM are shown in Fig. 7. Scanning Electron Microscope (SEM) test was conducted on the control mix sample and on the Strength and Durability Properties Empty Fruit Bunch Ash (EFBA) 55 15% EFBA, 28 days cured sample [5]. Adding EFBA to the concrete had provided additional calcium oxide which is an important component in producing secondary calcium-silicate-hydrate (C-S-H) gel. It is found that by adding EFBA into the sample, small pores are formed in the structure and C-S-H covers almost the entire fractured surface. The majority large spaces were filled with C-S-H gel because of the increased pozzolanic reactivity and hydration rate thus creating a denser structure compared to the controlled sample [6]. With increasing curing time, production of Ca(OH)2 decreases leading to an increased formation of C-S-H. Hence, EFBA can still be used as a partial replacement for cement without affecting the water absorption rate. Fig. 7. (a) SEM images of control mix, (b) SEM images of EFBA. 3.7 Carbonation Test Carbonation is a consequence of the transformation of calcium hydro-oxide [Ca(OH)2 ] to calcium carbonate [CaCO3 ] altering the microstructure of the cement paste by decreasing porosity. Figure 8 shows that about 50% relative humidity (RH) and about 5% CO2 for 28 days can cause the pore voids being filled and also likely for C-S-H attack which is what defects reinforcements.Oxygen and moisture are the components required for corrosion of reinforcement steel. Thru this test, the depth of carbonation is determined. The test is carried out by spraying Phenolphthalein indicator on freshly exposed surface of concrete sample. Carbonation depth is accessed when the indicator turns pink in contact with alkaline in sample with pH value greater than 9. At lower lever pH, the 56 G. Vasudevan et al. indicator is colourless as the sample is carbonated and the protective layer gets destroyed exposing the reinforcement to corrosion. With enhanced dispersion and filler effect, the sample shows a fairly low carbonation which indicates that EFBA can serve as a favourable pozzolanic material to replace OPC in producing a low carbonation depth and high strength mortar. Fig. 8. Carbonation test. 4 Conclusion and Recommendation The overall objective of the work was to investigate the feasibility of EFBA as a partial replacement of cement to concrete mix. This study included the preparation of concrete mixes containing and the evaluation of EFBA concrete properties in fresh and hardened states. The studying properties involved mix workability, compressive strength and flex- ural strength. According to experimental results, the usage of in EFBA concrete mixes as an alternative of disposal for EFBA is possible. The current type of EFBA accumulated in dumping sites and the expected future type were used in making concrete mixes. The influence of both types on concrete properties was studied. In all cases there was an optimum quantity of EFBA which can be used without introducing any change in mix preparation and acceptable properties were still be produced [7]. The study showed that the EFBA can be used in production of cement without changing the normal industrial process. Recommendation for further research The following recommendations are pro- posed for further research and study in order to from a complete picture of using empty fruit bunch ash in concrete mixes: 1) Investigation is needed on the different superplasticizer used together with empty fruit bunch ash in the concrete mix design. 2) Study is required for palm kernel shell powder mix with other by-products in the concrete mix design. 3) Further investigation of the higher percentage of replacement of the cement by the empty fruit bunch ash in concrete mix design. Strength and Durability Properties Empty Fruit Bunch Ash (EFBA) 57 4) Longer curing time is required in order to understand the behaviour of the concrete, and it might be giving the different result for empty fruit bunch ash. 5) Durability test such as sulphate attack test must be further study in order to understand percentage loss in weight of the empty fruit bunch ash concrete by sulphate attack. Acknowledgment. This study was supported by the Faculty of Built Environment, Tunku Abdul Rahman University College under the Centre for Construction Research. References 1. Vasudevan, G.: Performance of Alum Sludge as partial replacement for cement adding superplasticizer. In: IOP Conference Series: Materials Science and Engineering (2019) 2. Thomas, S., Kumar, S., Arel, H.S.: Sustainable concrete containing palm oil fuel ash as a supplementary cementitious material – a review. Renewable and Sustainable Energy Reviews 80, 550–561 (2017) 3. Kroehong, W.: Effect of palm oil fuel ash fineness on the microstructure of blended cement paste. Constr. Build. Mater. 25, 4095–4104 (2011) 4. Islam, M.M.U., Mo, K.H., Alengaram, U.J., Jumaat, M.Z.: Mechanical and fresh properties of sustainable oil palm shell lightweight concrete incorporating palm oil fuel ash. J. Clean. Prod. 115, 307–314 (2016) 5. Hamada, H.M., Yahaya, F.M., Muthusamy, K., Jokhio, G.A., Humada, A.M.: Fresh and hard- ened properties of palm oil clinker lightweight aggregate concrete incorporating Nano-palm oil fuel ash. Constr. Build. Mater. 214, 344–354 (2019) 6. Khankhaje, E., Hussin, M.W., Mirza, J., Rafieizonooz, M., et al.: On blended cement and geopolymer concretes containing palm oil fuel ash. Mater. Design 89, 385–398 (2016) 7. Vasudevan, G.: Performance of steel slag as a partial replacement fine aggregate incorporating superplasticizer. In: Awang, M., Meor M Fared, M. (eds.) ICACE 2019. Lecture Notes in Civil Engineering, vol. 59, pp. 151–160. Springer, Singapore (2020). https://doi.org/10.1007/978- 981-15-1193-6_17 Experimental Study on the Performance of FRP Grid Reinforced ECC Composite Layer-Concrete Interface Weiwen Li1 , Jie Liu1 , Jiahao Zhang1 , Shiying Tang1 , Meizhong Wu1 , and Xu Yang2(B) 1 Guangdong Provincial Key Laboratory of Durability for Marine Civil Engineering, College of Civil and Transportation Engineering, Shenzhen University, Shenzhen 518060, China 2 College of Civil and Environmental Engineering, Harbin Institute of Technology, Shenzhen 518055, China

Abstract. In this paper, the bond slip between the composite reinforcement layer and concrete is studied by single shear test. The composite reinforcement layer contains the CFRP (Carbon Fiber Reinforced Polymer) grid embedded within ECC (Engineering cementitious composites) matrix. Considering the influence of bond length, matrix type and ECC strength, the failure mode of the interface and the effect of various factors are clarified through the discussion of the experimental phenomenon, load slip relationship and bond slip relationship. The result indicated that the effective bond length increases with the decrease of ECC matrix strength, when the effective bond length exceeds 550 mm or the ECC matrix strength is greater than E40, the failure load can be increased. In comparison with mortar composite layer, ECC composite layer tends to be relatively ductile. Finally, two of existing FRP-concrete interface models were selected to fit the experimental curve, and the fitting effect of existing FRP-concrete interface models is acceptable. Keywords: Interface · ECC · FRP · Concrete · Bond behavior · Single shear test 1 Introduction As a kind of material for reinforcement, carbon fiber reinforced polymer (CFRP) has a lots of advantages, such as light weight, high strength, and corrosion resistance. Never- theless, what can’t be ignored is the disadvantage of its poor bonding performance with concrete. So many researchers use inorganic cement-based materials to replace epoxy resin, which is commonly used as a binder for FRP. Therefore, many studies on the FRP-concrete interface using inorganic cementitious materials as binders have begun to emerge, among which fabric-reinforced cementitious matrix (FRCM), textile-reinforced mortar (TRM) [1–3] and other reinforcements have emerged. This research direction produced a lots of achievement, including the interface of FRP strip-concrete [4], the interface of FRP woven mesh-concrete [5], etc. However, these materials tend to have © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 58–70, 2023. https://doi.org/10.1007/978-981-19-3983-9_6 Experimental Study on the Performance of FRP Grid 59 low stiffness and poor bonding performance, which leads to unsatisfactory reinforcement effects on the structure. A lot of researches have shown that CFRP grid has the characteristics of high rigidity and good stability, and has obvious advantages in the reinforcement of concrete structures [6]. At the same time, engineering cementitious composites (ECC) have the advantages of multiple cracks and strong adhesion to concrete [7–9]. According to the above summary, we can know that the composite layer formed by the CFRP grid embedded with ECC matrix (ECC composite layer) can provide a new idea for the reinforcement of structures in engineering. Therefore, this paper researched the bonding performance of the interface between ECC composite layer and concrete. The failure form and failure load are discussed firstly, and then the load-slip curve and bond-slip curve are analyzed, the experimental parameters including the bond length and the types of matrix as well as the strength of ECC matrix. Finally, the constitutive model of the interface between FRP and concrete was used to fit the experimental bond-slip curve. The results show that the model used in this paper is great, and provides a good empirical theory and research direction for the subsequent study of the bond performance of the ECC composite layer-concrete interface. 2 Experimental Study In this single shear test, the composite layer, bonded on the concrete block, as shown in Fig. 1. A total of 20 specimens were tested, and the research variables are shown in Table 1. 2.1 Materials The average cube strength and elastic modulus of the concrete is 25.5 MPa, 26.8 GPa. The average cube strength of mortar is 34.5 MPa, which is close to ECC within 60% Fig. 1. Specimen dimension (units: mm). 60 W. Li et al. fly ash. And configure ECC with different strengths by adjusting the content of fly ash [10], the details of ECC (unit in g/L) are listed in Table 2 as follows [11]. Fig. 2. CFRP grid. The size of the CFRP grid is 50 mm × 50 mm, and the cross-sectional area of a single grid is 26.4 mm2 (See in Fig. 2). The test results show that the tensile strength of the CFRP grid is 1515 MPa and the elastic modulus is 114.8 GPa. 2.2 Specimen Preparation The dimensions of each part of the test specimens are shown in Fig. 1. The layout of the strain gauge can be seen from Fig. 3. It should be note that the pouring sequence is to place a layer of 10 mm thick matrix firstly, and them the grid is placed on the surface of the matrix, finally a layer of 10 mm thick matrix is covered. Fig. 3. Details of strain gauge (units: mm). 2.3 Test Methods This single shear test adopts displacement-controlled loading, and the loading rate is 0.2 mm/min. As shown in Fig. 4. The machine applied load to the composite layer by clamping the end reinforcement sheet of grid. Paste three metal sheets (MS) and 5 displacement meters (DM) are used to measure the displacement, and several strain gauges (SG) can collect data of strain of grid. Experimental Study on the Performance of FRP Grid 61 Fig. 4. Schematic of single shear test. Table 1. Details of characterizations of specimens. Specimens Matrix types Bond length (mm) Numbers of specimens M-250 Mortar 250 5 M-350 350 M-450 450 M-550 550 M-650 650 E40-250 ECC containing 40% FA 250 5 E40-350 350 E40-450 450 E40-550 550 E40-650 650 E60-250 ECC containing 60% FA 250 5 E60-350 350 E60-450 450 E60-550 550 E60-650 650 E80-250 ECC containing 80% FA 250 5 E80-350 350 E80-450 450 E80-550 550 E60-650 650 62 W. Li et al. Table 2. Mixture proportions and properties of ECC. Types Cement Quartz Fly ash Water Water PE Ultimate Ultimate sand reducer fiber compressive tensile strength/MPa strength/MPa E40 800 266.6 533.3 400 5 15 51.5 6.489 E60 533.3 266.6 800 400 6 15 34.9 5.821 E80 266.6 266.6 1066.6 400 7.5 15 22.1 2.878 3 Result 3.1 Failure Mode The test results show that, regardless of ECC composite layer or the mortar composite layer, most of the failure modes of the specimens show the interfacial debonding (DB), and only the failure mode of specimen E80-650 shows grid fracture and interface is not debonded (FGB). By comparison, the mortar composite layer is brittle failure and irregular to follow, while the ECC composite layer is relatively ductile and its debonding process shows two different forms of sudden peeling and gradual debonding as the bond length increases. The load-slip curve is shown in Fig. 6. In a word, sudden peeling mainly occurs in the (a) (b) (c) (d) Fig. 5. Details of cracks (a, b and c are for ECC, d is for mortar). Experimental Study on the Performance of FRP Grid 63 35 30 25 Load(kN) 20 15 E40-250 10 E40-350 E40-450 5 E40-550 E40-650 0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 S (mm) (a)E40 35 30 25 Load(kN) 20 15 E60-250 10 E60-350 E60-450 5 E60-550 E60-650 0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 S (mm) (b) E60 45 40 35 30 Load(kN) 25 20 15 E80-250 E80-350 10 E80-450 E80-550 5 E80-650 0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 S (mm) (c) E80 35 30 25 Load(kN) 20 15 M-250 10 M-350 M-450 5 M-550 M-650 0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 S (mm) (d) M Fig. 6. Load-slip curve of specimens. 64 W. Li et al. specimens with bond lengths of 250 mm and 350 mm. There are only a few cracks before the specimens are destroyed, and finally the interface suddenly deboned. Nevertheless, gradually debonding occurs in the specimens with bond lengths of 450 mm, 550 mm, and 650 mm. Its failure characteristics are mainly termed (1) the specimen is in the elastic stage and there is no obvious damage when the load is small in the early stage of the test; (2) As the load increases to larger, the upper surface of the composite layer loading end began to bring fine cracks (weft cracks) into being, which are perpendicular to the loading direction; (3) When the load continued to increase in the middle of the test, the longitudinal cracks will appear, the number of cracks gradually increases and develops toward the free end, and the composite layer peels off at the loading end partially so that the amount of interface slip increases; (4) In the latest stage, The width and number of cracks in the composite layer further expand and become stable. The range of local debonding extends to the free end, and the amount of interface slip is about to reach its peak. Finally, the entire composite layer is completely peeled and destroyed. Cracks can be seen in Fig. 5. 3.2 Failure Load The failure load of each specimen is summarized in Table 3, and the influence of factors such as the bond length, the strength of ECC matrix and the type of matrix are discussed. In view of the effect of bonding length, the failure load increases with the bond length of the composite layer increases. When the bond length of the composite layer exceeds the effective bond length, the failure load will not continue increase. Therefore, the effective bond length range of different matrix composite layers can be roughly determined. The effective bond length ranges of E40 and E60 specimens is 550 mm– 650 mm, and the effective bonding length of E80 specimens is greater than 650 mm, as is shown in Table 3. Among them, the composite layer of the matrix of E40, E60 and Mortar showed many decreases after the bond length exceeded 550 mm. The author believes that this is mainly due to the inevitable shrinkage cracks formed during curing [12]. When the strength of ECC matrix is different, it is found that the interface failure load does not change linearly with the decrease in the strength of the ECC matrix when the bond length exceeds 450 mm. In general, the greater strength of the ECC matrix results in the greater ultimate bond stress of the interface, so that the interface failure load will also increase. However, the test results show that the failure load of E60 group of specimens has decreased, while the failure load of E80 group of specimens has increased significantly. The reason is that the effective bond length of E80 specimens is affected by the strength of the ECC matrix, and it also can be said that the smaller the strength of ECC matrix, the longer effective bond length. So under the coupling effect of the maximum bond stress of interface and the effective bond length, the failure load of E80 specimens increased. The type of matrix is different, and the bonding performance is also different. As shown in Table 3, the failure load and the slip of ECC composite layer are significantly greater than that of the mortar composite layer. It can be seen that ECC composite layer matrix has significant advantages of high ductility and large bearing capacity of interface. Experimental Study on the Performance of FRP Grid 65 2.5 2.0 Stress(MPa) 1.5 1.0 0.5 E40 E60 E80 0.0 0.00 0.25 0.50 0.75 1.00 1.25 S(mm) (a) Bond-slip curve of E40, E60 and E80 (b) Each stage of bond -slip curve Fig. 7. Bond-slip curve. Table 3. Test result. Specimens Global slip/mm Failure load/kN Failure mode Prediction of effective bond length E40-250 1.24 17.4 DB 550 mm–650 mm E40-350 1.78 24.8 DB E40-450 2.46 17.0 DB E40-550 3.37 33.0 DB E40-650 3.99 30.5 DB E60-250 1.79 19.1 DB 550 mm–650 mm (continued) 66 W. Li et al. Table 3. (continued) Specimens Global slip/mm Failure load/kN Failure mode Prediction of effective bond length E60-350 1.27 18.1 DB E60-450 2.43 18.4 DB E60-550 3.09 21.19 DB E60-650 4.03 22.5 DB E80-250 1.59 22.8 DB ≥650 mm E80-350 1.36 17.4 DB E80-450 2.7 28.1 DB E80-550 4.5 40.7 DB E80-650 5.5 43.3 FGB M-250 0.5 11.4 DB 450 mm–550 mm M-350 1.7 14.6 DB M-450 2.2 23.3 DB M-550 3.0 20.5 DB M-650 3.5 17.5 DB 3.3 Bond-Slip Curve According to the results, the displacement value of the composite reinforcement layer at the free end is almost zero, so it can be ignored. The value of the strain gauge is calculated by differential interpolation to obtain the bond-slip curve. According to the shape of the curves, the bond-slip curves of E40, E60, and E80 specimens can be divided into three stages: elastic rising stage, plastic rise stage, and plastic fall stage, as shown in Fig. 7. Among them, the elastic rise stage curve is roughly linear, the bond stress of interfacial increases rapidly with the increase of slip, and the bonding stiffness of the interface is larger; The interface shear stress of the plastic rise stage is greater than the chemical bonding force between concrete and ECC, the growth rate of interface shear stress decreases gradually with the increase of slip, and the bonding stiffness of the interface gradually becomes smaller; When the shear stress peak is reached, the curve enters the plastic fall stage. At this time, the interface is damaged, it is also means that the aggregate on the interface between ECC and concrete wears, the shear stress of the interface decreases and the amount of slippage increases rapidly. The overall shape of the bond slip curve, meanwhile, is not affected by the strength of the ECC matrix, and before the composite layer is completely peeled off, a residual stress section similar to the FRCM bond slip curve appears at the end of the bond slip curve [4] This is mainly due to the mechanical occlusion of the ECC and concrete interface and the bridging effect of the fibers at the ECC matrix interface, and the bond-slip curve shows a certain degree of ductility. Experimental Study on the Performance of FRP Grid 67 3.4 Comparision of Models Considering that there are few studies on prediction models related to the interface between ECC and concrete, but the prediction models related to the interface between FRP and concrete have a wide range of applications. Among them, the models proposed by He [13] and Wu [14] have great prediction effects, as seen Eq. 1 and Eq. 2. These two models are fitted to the experimental results, and the fitted bond-slip curve is consistent with the experimental results, as shown in Fig. 8. In the rising section of the curve, both the He model and the Wu model are in good agreement with the experimental values; In the falling section, the experimental data are basically distributed on both sides of the fitting curve of Wu model, while the fitting curve of He model has relatively poor accuracy. Table 4 lists the relevant fitting parameters and the fitting coefficients of each model. The average values of the fitting coefficients R2 of He model and Wu model are 0.818 and 0.965, respectively. It can be seen that using Wu model as the constitutive Fig. 8. Fitting curve of bond-slip relationship. 68 W. Li et al. model of bond-slip relationship between the ECC composite layer and concrete will be more accurate. He model: s n τ (s) = τmax ( ) (1) s0 (n − 1) + (s/s0 )n Wu model: τmax (ω + eω − 1) −ωs C + 2e−ω − 1 ωs −ωs τ (s) = 2 e s0 + C − ω · × 1 − e s0 1 − e−ω (ω + Ceω + 1) ω+e −1 s0 (2) where n is a parameter that mainly controls the softening branch; ω and C are infinite constants; τmax is the average maximum bond stress; s0 are the slip at the maximum bond stress. Table 4. Fitting result. Specimens τ max /MPa s0 /mm n R2 (He model) C ω R2 (Zhou model) E40-250 2.1 0.33 1.701 0.945 1.152 0.441 0.983 E40-450 2.39 0.21 1.412 0.976 1.135 0.384 0.993 E40-550 2.78 0.27 1.575 0.767 1.103 0.362 0.954 E60-250 1.26 0.13 1.453 0.836 1.155 0.650 0.923 E60-350 1.21 0.10 1.386 0.917 1.198 0.681 0.995 E60-450 1.74 0.11 1.539 0.680 1.155 0.639 0.945 E80-350 0.80 0.10 2.955 0.933 1.112 0.944 0.995 E80-450 0.93 0.31 1.219 0.377 1.115 0.962 0.906 E80-550 1.10 0.12 2.166 0.938 1.303 0.874 0.994 Average value 0.818 0.965 4 Summary In this study, the bonding behavior of the interface between the FRP grid-reinforced ECC composite layer and concrete was studied by single shear test. The conclusions are as follows: (1) The debonding process of the interface between ECC composite layer and concrete is mainly divided into sudden peeling and gradual debonding as the bond length increases. Furthermore, the surface crack morphology of the ECC composite layer at the time of failure is not affected by factors such as the strength of ECC matrix, the types of ECC matrix, and the bond length. Experimental Study on the Performance of FRP Grid 69 (2) The failure load of the interface between ECC composite layer and concrete is affected by the coupling of the effective bond length and the strength of ECC matrix. Under the condition that the effective bond length increases with the decrease of ECC matrix strength, when the effective bond length exceeds 550 mm or the ECC matrix strength is greater than E40, the failure load can be increased; (3) The test results show that the bond-slip curve of the interface between the FRP grid-ECC composite layer and the concrete is basically the same. It can be divided into three stages: elastic rising stage, plastic rising stage, and plastic falling stage. And fitting it with the bond-slip model of the interface between FRP and concrete can get a good fitting work. Acknowledgment. The authors would like to appreciate the financial support provided by the National Natural Science Foundation of China (NSFC Grant Numbers: 51878415, 51908373). Technical support is acknowledged from Guangdong Provincial Key Laboratory of Durability for Marine Civil Engineering (SZU), No. 2020B1212060074. References 1. Bournas, D.A., Lontou, P.V., Papanicolaou, C.G., Triantafillou, T.C.: Textile-reinforced mor- tar versus fiber-reinforced polymer confinement in reinforced concrete columns. ACI Struct. J. 104(6), 740 (2007) 2. ACI Committee 549: Guide to design and construction of externally bonded fabric-reinforced cementitious matrix (FRCM) systems for repair and strengthening concrete and masonry structures (2013). ACI 549.4R-13 3. Focacci, F., D’Antino, T., Carloni, C.: The role of the fiber–matrix interfacial properties on the tensile behavior of FRCM coupons. Constr. Build. Mater. 265, 120263 (2020) 4. Nelson, L.A., Al-Allaf, M., Weekes, L.: Analytical modelling of bond-slip failure between epoxy bonded FRP and concrete substrate. Compos. Struct. 251, 112596 (2020) 5. Jiang, J., Jiang, C., Li, B., Feng, P.: Bond behavior of basalt textile meshes in ultra-high ductility cementitious composites. Compos. B Eng. 174, 107022 (2019) 6. Liu, Z., Yue, Q., Li, R., Ding, Y.: FRP grid in civil engineering applications. In: The 9th National Symposium on FRP application in Construction Engineering (2015) 7. Li, V.C., Leung, C.K.Y.: Steady-state and multiple cracking of short random fiber composites. J. Eng. Mech. 118(11), 2246–2264 (1992) 8. Yang, X., Gao, W.Y., Dai, J.G., Lu, Z.D., Yu, K.Q.: Flexural strengthening of RC beams with CFRP grid-reinforced ECC matrix. Compos. Struct. 189, 9–26 (2018) 9. Zheng, Y.: Experiment and calculation method research on reinforced concrete (RC) beams strengthened with the composite of FRP grid and ECC. A dissertation submitted to Southeast University (2018) 10. Liu, C., Bi, Y., Hua, Y.: Mechanical properties of PVA-ECC with high volume fly ash and mechanism analysis of fly ash. Bull. Chin. Ceram. Soc. 36(11), 3739–3744 (2017) 11. Fiber reinforced polymer composite grids for civil engineering. GB/T 36262—2018 12. Li, M., Li, V.C.: Behavior of ECC/concrete layered repair system under drying shrinkage conditions/Das Verhalten eines geschichteten Instandsetzungssystems aus ECC und Beton unter der Einwirkung von Trocknungsschwinden. Restorat. Build. Monuments 12(2), 143– 160 (2006) 70 W. Li et al. 13. He, W., Wang, X., Ding, L., Wu, Z.: Experimental study on bond behavior of interface between fiber-reinforced polymer grids and concrete substrate. Compos. Struct. 257, 113164 (2021) 14. Zhou, Y.W., Wu, Y.F.: General model for constitutive relationships of concrete and its composite structures. Compos. Struct. 94(2), 580–592 (2012) Structural Mechanics and Structural Engineering Dynamic Strain Estimations of Beam Ends in Steel Moment-Resisting Frames Using Acceleration Data Xiyang Yu and Xiaohua Li(B) School of Civil Engineering, Chongqing University, Chongqing 400045, China

Abstract. Dynamic-strain-based damage detection methods for fractured beams in steel moment-resisting frames have been proposed previously. However, it is impractical to install a dense-array strain sensing system in a real steel frame, while using acceleration measurements to estimate strain responses of the steel frame can make the collection of strain data more convenient. In this study, a linear-model- based state observer (LMBO) was used to estimate dynamic strain responses of beam ends in a steel moment-resisting frame subjected to base excitations, and required measurements were only story accelerations. In the state observer, a steel frame experiencing base motions is equivalent to the frame with corrective forces and added grounded dampers. The effectiveness of this method was numerically investigated through a finite element model of a three-story steel moment-resisting frame in SAP 2000 software. The results show that when the input excitation was white noise or an earthquake motion recorded in the 1989 Loma Prieta earthquake, estimated strain responses were close to the corresponding simulated results, and the values of the Time Response Assurance Criterion (TRAC) between the both strain responses exceeded 0.8. Keywords: Dynamic strain estimation · State observer · Steel moment-resisting frame 1 Introduction After a major earthquake, it is essential to evaluate the post-earthquake safety of build- ings, which will be helpful to decision making and re-occupancy [1]. For the steel moment-resisting frame, the earthquake-induced damages are more likely to occur at welded beam ends due to the strong-column and weak-beam design philosophy. For example, reconnaissance reports of the 1994 Northridge earthquake and the 1995 Kobe earthquake pointed out that a large number of steel moment-resisting frames suffered damages at welded beam ends [2–4]. Furthermore, this kind of structural damages will threaten the safety of steel buildings [5, 6]. Thus, accurate information of local damages (i.e., fractures near beam ends) of the steel moment-resisting frame will improve the reliability of the post-earthquake performance evaluation. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 73–86, 2023. https://doi.org/10.1007/978-981-19-3983-9_7 74 X. Yu and X. Li For reasons such as limitations of manual inspections, many structural damage detec- tion theories have been proposed in the past decades [7], such as model parameter-based methods [8, 9] and signal processing-based methods [10, 11]. For example, Li et al. pro- posed an approach based on time-frequency analysis and fractal dimension, which was successfully used to determinate locations of seismic damages in a ten-story shear-type building model [11]. Many of these methods are capable of determining variations in structural global vibration characteristics or further identifying damage locations, but it is challenging to quantify the damage severity. In recent years, Li (the author) et al. have conducted a series of studies to deal with the problem [12–15]. By comparing strain responses on steel beams under small-amplitude dynamic excitations before and after a major earthquake, Li et al. first proposed a dynamic-strain-based damage index [12] and developed simplified derivation of a damage curve for the seismically induced beam fractures [13], and then Li et al. presented a decoupling algorithm for evaluating multiple beam damages in a steel frame [14]. In the experimental investigations, the above methods determined the locations of damages at beam ends in a five-story steel moment-resisting frame and quantified the corresponding damage extents. It was worth noting that the strain data used in the methods proposed by Li et al. were measured under small-amplitude loads (e.g., ambient vibrations), and the steel frame was in the elastic stage. However, a dense-array strain sensing system is necessary for the dynamic-strain- based methods, which is difficult to meet in reality. With the progress of strong-motion instrumentation programs (SMIP) in seismically active regions, using acceleration data to estimate strain responses of the steel moment-resisting frame can make the collection of strain data more convenient, which is the main motivation of this study. In the field of strain estimation, Hernandez et al. proposed a linear-model-based state observer (LMBO) [16, 17], where a finite element model was utilized to estimate local strain responses of a structure in the elastic stage by using limited acceleration measurements. This method has been successfully applied in a seven-story full-scale reinforced concrete cantilever shear wall tested by a shaking table [18], and compared with the measured strain data, the accuracy of the estimated results of the method is significantly higher than that of the results from directly applying the base excitation to the finite element model of the shear wall. However, the application of this method in strain estimations of the steel moment-resisting frame has not been studied before. In this paper, the effectiveness of the method was numerically studied, thus laying the foundation for the subsequent experimental verification and dynamic-strain-based damage detection. Firstly, Sect. 2 presents the simplified derivation of the LMBO which is a feedback system combining a mathematical model with measured data [19]. Then in Sect. 3.1 and 3.2, a finite element model was established in SAP 2000 software [20] according to a three-story steel moment-resisting frame in the structural laboratory of the Disaster Prevention Research Institute, Kyoto University, Japan. In Sect. 3.3 and 3.4, the estimated strain responses at beam ends from the LMBO were compared with the simulated results from the original finite element model, when the input excitation was a stationary random process (banded white noise) or a nonstationary random process (an earthquake motion recorded in the 1989 Loma Prieta earthquake). Dynamic Strain Estimations of Beam Ends in Steel Moment-Resisting Frames 75 2 Linear-Model-Based State Observer The detailed derivation of the LMBO method proposed in [16–18] is briefly summarized here for the sake of completeness. The building structure in the elastic stage is linear and time-invariant. In the state observer (i.e., LMBO method), a structure subjected to external excitations is equivalent to a structure with corrective forces and added grounded dampers. The motion equation of a building structure with n degrees of freedom is defined as: M ẍ(t) + C D ẋ(t) + Kx(t) = b1 u(t) (1) where M Rn×n is a diagonal mass matrix; x(t) Rn , ẋ(t) Rn and ẍ(t) Rn are the displacement, velocity and acceleration relative to the ground respectively; K Rn×n is a stiffness matrix of the structure; u(t) Rz is a matrix of excitations and z is the number of external excitations. The excitations are the equivalent loads related to seismic accelerations when the structure is subjected to earthquakes; b1 Rn×z is the spatial distribution of the excitations. In the LMBO method, assuming that the velocity measurements are available (in this paper, the velocity measurements are obtained from the acceleration measurements), the estimated displacement x(t) Rn can be given by the following equation (state observer): ¨ ˙ M x(t) + (C D + λT βλ) x(t) + K x(t) = λT β q̇k (t) (2) ˙ ¨ where x(t) Rn and x(t) Rn are the estimated velocity and acceleration relative to the ground; λ R r×n is a matrix which is decided by the locations of measurement points, and r is the number of measurement points; β Rr×r is a symmetric matrix which is free to select but affects the accuracy of the estimated results; q̇k (t) Rr is a matrix consisting of the velocity measurements. After comparing Eq. (1) with Eq. (2), it can be found that a structure subjected to a seismic load can be equivalent to the structure with corrective forces and added grounded dampers by using the LMBO. Figure 1 shows a shear model with corrective forces F = λT β q̇k (t) and added grounded dampers C = λT βλ. Such a physical interpretation allows the state observer to be implemented in a finite element software and solved quickly, which is one of its significant advantages [16]. In order to obtain accurate estimation results, it is necessary to find the appropriate β. The trace of the covariance matrix of the displacement estimation error is set as the objective function L: L = tr E (x(t) − x(t))(x(t) − x(t))T +∞ = tr( 0 b1 Suu (w)bT1 ∗0 + Γ 0 λT βSvv (w)β T λΓ ∗0 )dw (3) −∞ and Γ 0 = (−Mw2 + C D iw + λT βλiw + K)−1 (4) 76 X. Yu and X. Li where Suu (w) and Svv (w) are the power spectral density (PSD) matrices of the external excitation and measurement noise, respectively. Then the matrix β corresponding to the minimum value of L is the optimal solution. To simplify the process of solving for the optimal solution of β, let β = θ Ir , where I is an identity matrix. Thus, the selection of β can be expressed concisely as: min(L) = L(θoptimal ∈ R+ ) (5) Fig. 1. Demonstration of the implementation of the LMBO method: (a) a shear model subjected to a seismic load, (b) a shear model with the LMBO 3 Numerical Example 3.1 Basic Information A finite element model of a three-story steel frame was established in SAP 2000 according to a three-story steel moment-resisting frame tested on a shaking table in the structural laboratory of the Disaster Prevention Research Institute, Kyoto University, Japan. Then the simulated acceleration measurements from the finite element model were used to estimate the strain responses of the bottom flanges at the beam ends, so as to numerically verify the effectiveness of the LMBO method in strain prediction of the steel frame in the elastic phase. The three-story frame specimen is shown in Fig. 2a. The steel material of the frame was SS400. It had two bays in X direction, 4 m in total, and one bay in Y direction, 1 m in total. The heights of the first to the third stories were 915 mm, 880 mm and 880 mm, with mass of 1700 kg, 1600 kg and 1600 kg respectively. Special design of the damage link at the beam end is shown in Fig. 2b. More design details about this steel frame can be found in [21, 22]. Figure 3 shows sensor deployment. In the numerical example, the locations of simulated acceleration measurements were the same as the Dynamic Strain Estimations of Beam Ends in Steel Moment-Resisting Frames 77 locations of the real accelerometers, and the locations of simulated and estimated strain responses were at the beam ends (as shown in Fig. 3). Moreover, sampling frequencies of all measurements were 100 Hz. Due to the symmetry of the steel frame, the strain responses at the 1st to 12th measurement points (as shown in Fig. 3) were estimated in this paper. It was worth noting that the input excitations were all in the X direction and only the measurement data in the X direction were considered because the frame was rigid (moment-resisting) in the X direction. Fig. 2. Shaking table test of the three-story steel moment-resisting frame: (a) overview, (b) beam– column joint Fig. 3. Deployment of sensors and measurement points 3.2 Model Verification Figure 4 shows the finite element model of the three-story steel frame simulated by using SAP 2000. The beams and columns consist of linear elastic frame elements. The beams 78 X. Yu and X. Li and columns are rigidly connected. The material properties were selected as follows: Young modulus E = 200.5 GPa and Poisson ratio v = 0.3. Due to the special design of the damage link, it was replaced by I-shaped cross section with the same section modulus in bending, and the section modulus of an undamaged link is 8667 mm3 [21]. In addition, all structural damping ratios were classically taken to be 5%. The simulated results of the finite element model and the measured results of the shaking table test were compared from the aspects of frequency domain and time domain to verify the feasibility of finite element model. In the frequency domain, the first three structural frequencies in the X direction identified from three types of data: (1) Measured acceleration responses. They were measured from the real steel frame subjected to a white noise base excitation (with vibration frequency band of 0–50 Hz and effective value of 7 gal), and then structural frequencies in the X direction were obtained by using a frequency domain decomposition method. The frequencies from the first to third orders were 4.364 Hz, 10.240 Hz, and 11.550 Hz respectively; (2) Results of modal analysis. In SAP 2000, the modal analysis of the finite element model was carried out to obtain the three natural frequency in the X direction, and the corresponding results were 4.509 Hz, 9.915 Hz, 11.636 Hz; (3) Simulated acceleration responses. The same white noise excitation as the vibration test was applied to the finite element model in SAP 2000, and then the simulated acceleration of the third floor was decomposed in frequency domain to obtain the structural frequency. The corresponding results were 4.382 Hz, 9.192 Hz, 12.570 Hz. Based on the frequencies obtained from the measured acceleration (i.e., 4.364 Hz, 10.240 Hz, and 11.550 Hz), the differences of the same order frequencies in the three types of data were less than 3.5%, 10.3% and 8.9% respectively. In the time domain, Fig. 5 compares the measured and simulated absolute acceler- ations of the third story when the Kobe-30% seismic wave (i.e., the amplitude of the input wave is 30% of the original wave recorded in the Kobe earthquake, and the peak acceleration is 2.53 m/s2 ) was input to the real specimen and the finite element model. It could be found that the both time histories were relatively consistent. Thus, the finite element model is verified in linear range with measurement results. Fig. 4. Finite element model of three-story steel frame Dynamic Strain Estimations of Beam Ends in Steel Moment-Resisting Frames 79 Fig. 5. Absolute acceleration response of 3rd story under Kobe-30% 3.3 White Noise Base Excitation A banded white noise base excitation (with vibration frequency band of 0.2–30 Hz and a sampling frequency of 100 Hz) was input to the finite element model. This kind of load could be regarded as a stationary random process. A white noise process with a 10% root mean square (RMS) noise-to-signal value was superimposed on the three simulated acceleration responses of the third story (three accelerometers at the third story), and then the processed acceleration data were regarded as measurement data. In order to obtain a suitable feedback matrix β, the three-story frame was simplified as a time-invariant shear model with lumped mass. And structural parameters of this shear model were the same as these of the finite element model. The inter-story stiffnesses of the first to third stories were 15900 kN/m, 22300 kN/m and 10700 kN/m respectively. Mass matrix M = diag (1700 kg, 1600 kg,1600 kg). Meanwhile, it was assumed that structural damping matrix CD was formulated using the Rayleigh damping model: 2ξ ω1 ω2 2ξ CD = M+ K (6) ω1 + ω2 ω1 + ω2 where ω1 and ω2 = first two natural frequencies of the structure; ξ = 5%. In this study, the velocity signals were integrated from the accelerations and then applied to a high- pass filter with a 0.2 Hz cutoff frequency. The power spectral density (PSD) matrices of the external excitation and measurement noise were given by Suu (w) = diag (35.08, 31.05, 31.05) and Svv (w) = 8.02 × 10−7 I1 . In this case, an optimal value was θ optimal = 2550 Ns/m with L = 2.11 × 10–9 . Figure 6 shows the relationships between L and θ when the PSD of measurement noise was Svv , 2Svv or 3Svv . A suboptimal value of θ was 1350 Ns/m when the PSD of measurement noise was 2Svv , and it yielded a value of L = 2.21 × 10–9 which was closed to the optimal. More importantly, as can be seen from Fig. 6, this suboptimal θ was still valid under stronger measurement noise. Thus, θ = 1350 Ns/m was selected to implement the LMBO method. In addition, the damper value (i.e., θ ) was divided by three and assigned to the added grounded dampers at the locations of accelerometers of the third story in the X direction. The dynamic responses at the beam ends, including the bending moment and axial force, could be obtained from the finite element model through SAP 2000 software. The strain responses of the bottom flange at the beam ends were calculated by using material mechanics and then applied to a band-pass filter (0.4–15 Hz). Due to limited space, Figs. 7, 8, 9 and 10 compare the simulated and estimated strain responses at the first to 80 X. Yu and X. Li Fig. 6. Relationships between L and θ Fig. 7. Simulated and estimated strains at 1st measurement point: (a) time history, (b) detail time history, (c) frequency domain fourth measurement points, and it could be found that they were relatively consistent in both time and frequency domains. Figure 11 compares the values of the Time Response Assurance Criterion (TRAC) [23, 24] between the simulated and estimated strains at all twelve measurement points. [{Strains (t)}T {Straine (t)}]2 TRAC = (7) [{Strains (t)} {Strains (t)}][{Straine (t)}T {Straine (t)}] T where {Strains (t)} = time history of the simulated strain signal; {Straine (t)} = time history of the estimated strain signal. The value of TRAC is between zero and unity, and the closer it is to unity, the higher the agreement between the two signals. It could be found from Fig. 11 that the values of TRAC at all twelve measurement points exceeded 0.8, which indicted a high agreement between the simulated and estimated strains. Thus, in linear elastic range, the LMBO method was capable of estimating strain responses at beam ends in the steel moment-resisting frame when the input excitation was a stationary random process. 3.4 Earthquake Excitation An earthquake motion recorded in the 1989 Loma Prieta earthquake was input to the finite element model. This kind of load could be regarded as a nonstationary random Dynamic Strain Estimations of Beam Ends in Steel Moment-Resisting Frames 81 Fig. 8. Simulated and estimated strains at 2nd measurement point: (a) time history, (b) detail time history, (c) frequency domain Fig. 9. Simulated and estimated strains at 3rd measurement point: (a) time history, (b) detail time history, (c) frequency domain 82 X. Yu and X. Li Fig. 10. Simulated and estimated strains at 4th measurement point: (a) time history, (b) detail time history, (c) frequency domain Fig. 11. Values of TRAC at all twelve measurement points process. A white noise process with a 10% root mean square (RMS) noise-to-signal value was superimposed on the three simulated acceleration responses of the third story, and then the processed acceleration data were regarded as the measurement data. The velocity signals were obtained by using the same data processing method as Sect. 3.3. Through the time-invariant shear model in Sect. 3.3, a value of θ = 1100 Ns/m was selected to implement the LMBO method. It was worth noting that the strain responses were calculated by using material mechanics and then applied to a band-pass filter (2.5–15 Hz). Figures 12, 13, 14 and 15 show the simulated and estimated strain responses at the first to fourth measurement points, and two kinds of signals were also relatively consistent in both time and frequency domains. Figure 16 shows the values of TRAC, and they all exceeded 0.83, which indicted a high agreement between the simulated and estimated strains. Thus, in linear elastic range, the LMBO method was also capable of estimating strains at beam ends in the steel moment-resisting frame when the input excitation was a nonstationary random process. Dynamic Strain Estimations of Beam Ends in Steel Moment-Resisting Frames 83 Fig. 12. Simulated and estimated strains at 1st measurement point: (a) time history, (b) detail time history, (c) frequency domain Fig. 13. Simulated and estimated strains at 2nd measurement point: (a) time history, (b) detail time history, (c) frequency domain 84 X. Yu and X. Li Fig. 14. Simulated and estimated strains at 3rd measurement point: (a) time history, (b) detail time history, (c) frequency domain Fig. 15. Simulated and estimated strains at 4th measurement point: (a) time history, (b) detail time history, (c) frequency domain Fig. 16. Values of TRAC at all twelve measurement points Dynamic Strain Estimations of Beam Ends in Steel Moment-Resisting Frames 85 4 Conclusion This paper presents the use of a linear-model-based state observer (LMBO) to estimate the strain responses at beam ends in a steel moment-resisting frame. Although this method has been proposed before, it was the first time to use it in estimating strain responses of the steel moment-resisting frame. The effectiveness of the LMBO method was numerically verified through a finite element model of a steel frame in the SAP 2000 software. The results show that when the input excitation was a stationary random process (banded white noise) or a nonstationary random process (an earthquake motion recorded in the 1989 Loma Prieta earthquake), estimated strain responses were close to the corresponding simulated results, and the values of Time Response Assurance Criterion (TRAC) between the both strain responses exceeded 0.8. 5 Future Work Further research is required for the strain estimation and damage detection of beam ends in steel moment-resisting frames. The future work is as follows: (1) It is necessary to find a better method to obtain the suitable value of θ; (2) The effectiveness of the LMBO method in strain estimation will be experimentally verified; (3) The LMBO method will be combined with the dynamic-strain-based damage detection method proposed by the author Li in order to identify and quantify seismic damages at the beam ends of steel frames. Acknowledgments. This work was supported by the National Natural Science Foundation of China (Grant No. 52178454) and Natural Science Foundation of Chongqing (Grant No. cstc2019jcyj-msxmX0254). References 1. Celebi, M., et al.: Real-time seismic monitoring needs of a building owner—and the solution: a cooperative effort. Earthq. Spectra 20(2), 333–346 (2004) 2. Nakashima, M.: Reconnaissance report on damage to steel buildings structures observed from the 1995 Hyogoken-Nanbu (Hanshin/Awaji) earthquake, Abridged English edition. Steel Committee of Kinki Branch, the Architectural Institute of Japan (AIJ) (1995) 3. Youssef, N.F.G., Bonowitz, D., Gross, J.L.: A survey of steel moment-resisting frame build- ings affected by the 1994 Northridge earthquake. US National Institute of Standards and Technology (1995) 4. Mahin, S.A.: Lessons from damage to steel buildings during the Northridge earthquake. Eng. Struct. 20(4–6), 261–270 (1998) 5. Rodgers, J.E., Mahin, S.A.: Effects of connection fractures on global behavior of steel moment frames subjected to earthquakes. J. Struct. Eng. 132(1), 78–88 (2006) 6. Lignos, D.G., et al.: Numerical and experimental evaluation of seismic capacity of high-rise steel buildings subjected to long duration earthquakes. Comput. Struct. 89(11–12), 959–967 (2011) 7. Hou, R., Xia, Y.: Review on the new development of vibration-based damage identification for civil engineering structures: 2010–2019. J. Sound Vib. 491, 115741 (2021) 86 X. Yu and X. Li 8. Yoon, M.K., Heider, D., Gillespie, J.W., et al.: Local damage detection with the global fitting method using operating deflection shape data. J. Nondestr. Eval. 29(1), 25–37 (2010) 9. Caddemi, S., Caliò, I.: The exact explicit dynamic stiffness matrix of multi-cracked Euler- Bernoulli beam and applications to damaged frame structures. J. Sound Vib. 332(12), 3049– 3063 (2013) 10. Huang, N.E.: The Hilbert-Huang Transform in Engineering. Taylor and Francis Group, New York (2005) 11. Li, H., Tao, D., Huang, Y., et al.: A data-driven approach for seismic damage detection of shear-type building structures using the fractal dimension of time–frequency features. Struct. Control. Health Monit. 20(9), 1191–1210 (2013) 12. Li, X., Kurata, M., Nakashima, M.: Evaluating damage extent of fractured beams in steel moment-resisting frames using dynamic strain responses. Earthq. Eng. Struct. Dyn. 44(4), 563–581 (2015) 13. Li, X., Kurata, M., Nakashima, M.: Simplified derivation of a damage curve for seismically induced beam fractures in steel moment-resisting frames. J. Struct. Eng. 142(6), 04016019 (2016) 14. Li, X., Kurata, M., Suzuki, A.: Decoupling algorithm for evaluating multiple beam damages in steel moment-resisting frames. Earthq. Eng. Struct. Dyn. 46(7), 1045–1064 (2017) 15. Suzuki, A., Kurata, M., Li, X., et al.: Residual structural capacity evaluation of steel moment- resisting frames with dynamic-strain-based model updating method. Earthq. Eng. Struct. Dyn. 46(11), 1791–1810 (2017) 16. Hernandez, E.M.: A natural observer for optimal state estimation in second order linear structural systems. Mech. Syst. Signal Process. 25(8), 2938–2947 (2011) 17. Erazo, K., Hernandez, E.M.: A model-based observer for state and stress estimation in struc- tural and mechanical systems: experimental validation. Mech. Syst. Signal Process. 43(1–2), 141–152 (2014) 18. Erazo, K., Hernandez, E.M.: High-resolution seismic monitoring of instrumented buildings using a model-based state observer. Earthq. Eng. Struct. Dyn. 45(15), 2513–2531 (2016) 19. Roohi, M., Hernandez, E.M., Rosowsky, D.: Nonlinear seismic response reconstruction and performance assessment of instrumented wood-frame buildings—validation using NEESWood Capstone full-scale tests. Struct. Control. Health Monit. 26(9), e2373 (2019) 20. Wilson, E.L., Habibullah, A.: SAP 2000 Software, Version 21. Computer and Structures, Inc. (CSI), Berkeley (2019) 21. Nishino, H.: Evaluation of residual seismic performance of steel buildings using local damage identification sensing and model updating. Kyoto University, Kyoto (2014) 22. Matarazzo, T.J., Kurata, M., Nishino, H., et al.: Postearthquake strength assessment of steel moment-resisting frame with multiple beam-column fractures using local monitoring data. J. Struct. Eng. 144(2), 04017217 (2018) 23. Avitabile, P., Pingle, P.: Prediction of full field dynamic strain from limited sets of measured data. Shock. Vib. 19(5), 765–785 (2012) 24. Maes, K., Iliopoulos, A., Weijtjens, W., et al.: Dynamic strain estimation for fatigue assessment of an offshore monopile wind turbine using filtering and modal expansion algorithms. Mech. Syst. Signal Process. 76, 592–611 (2016) Studies on the Relationship Between Anchor Force of Prestressed Anchor Cable and Nonlinear Vibration of Anchor Head Hao Li1(B) and Hui Cao1,2 1 School of Civil Engineering, Chongqing University, Chongqing 400045, China

2 MOE Key Lab of New Technology for Construction of Cities in Mountain Area, Chongqing University, Chongqing 400045, China Abstract. In order to ensure the normal operation of prestressed anchor cable, the detection of anchor force is a very important link in the health detection of prestressed anchor cable. By analyzing the relationship between nonlinear vibra- tion characteristics and anchoring force, this paper explores the nondestructive testing method of anchoring force. The concrete workbench was used to simulate the soil anchored by the anchor cable. Under the condition of maintaining the tension at all levels by the pull-out apparatus, the lateral excitation was carried out at the anchor head of the anchor cable, and the free vibration acceleration signal of this part was recorded. The first-order frequency signal was decomposed by AMD, and Hilbert transform was performed on it to obtain the relationship between the frequency, amplitude and time of the signal, and then the frequency- amplitude curve reflecting the nonlinear vibration characteristics was obtained. The same amplitude segment is selected on the frequency-amplitude curve, and the corresponding frequency difference is used as the nonlinear index to analyze the relationship between the tension (anchoring force) and the nonlinear vibration degree of each level. The test and analysis results show that the degree of nonlinear vibration decreases with the decrease of anchorage force. Therefore, the nonlinear index can be used to judge the anchoring force of anchor cable. Keywords: Prestressed anchor cable · Nonlinear vibration characteristics · Anchoring force · Nondestructive testing · AMD decomposition 1 Introduction Prestressed anchor cable is widely used in practical engineering. In the long-term work of the anchor cable, the cable relaxation, soil creep and other unfavorable factors are easy to reduce the anchorage force of the anchor cable, which leads to anchorage failure. Anchoring force is an important index to evaluate the supporting quality of anchorage system, so the anchorage force detection of prestressed anchor cable is a very important subject. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 87–97, 2023. https://doi.org/10.1007/978-981-19-3983-9_8 88 H. Li and H. Cao At present, the detection of anchorage force of prestressed anchor cable mainly adopts mechanical or drawing method. Nondestructive testing methods, such as vibration signal analysis of anchorage force research is relatively less. Dang et al. [1] proposed and verified that the PCA-based algorithm can filter the influence of ambient temperature on the impedance monitoring of anchorage zone of anchor cable, and found that the impedance sensor arranged near the anchor head had better effect on the damage monitoring of anchorage zone. Huang et al. [2] verified the reliability of the nondestructive testing instrument based on the stress wave method by testing the anchorage quality of the bolt and anchor cable in the project. Based on the understanding that the stiffness of the structure changes with the ampli- tude before and after the damage, many scholars use nonlinear vibration characteristics to study the nondestructive testing of the structure. Spalvier [3] proposed torsional- vibration testing technology to characterize the nonlinear of concrete structure to deter- mine the stress state of the structure. Alnuaimi [4] compared the linear and nonlinear ultrasonic methods to analyze the internal damage of concrete structures, and proved that the nonlinear ultrasonic parameters were more sensitive to the changes in the inter- nal structure of materials. Wan et al. [5] used nonlinear characteristics to detect gas on nano-beam surface. Cao et al. [6, 7] studied the identification of debonding by analyzing the nonlinear characteristics of CFST. At the anchor head of the anchor cable, the anchoring force makes the anchor plate closely contact with the concrete anchor pier. Under transverse excitation, the interface effect similar to the crack of concrete beam exists on the interface between the anchor plate and the concrete pier, which makes the vibration of the anchor head present non- linear vibration characteristics. In the case of a certain roughness of the interface, the interface effect is mainly affected by the size of the anchorage force. The larger the anchorage force, the closer the interface contact, the more obvious the interface effect and the stronger the nonlinear vibration characteristics. Therefore, it is worth studying to analyze the nonlinear vibration characteristics of the anchor head so as to realize the anchorage force detection. In this paper, the dynamic detection of prestressed anchor cable specimens under different working conditions is carried out, and the relationship between the nonlinear vibration characteristics of prestressed anchor cable and the anchoring force is analyzed to explore the feasibility of non-destructive testing of anchor cable anchoring force based on nonlinear vibration characteristics. 2 Principle and Method 2.1 Relationship Between Nonlinear Vibration Characteristics and Anchoring Force The so-called interface effect refers to the micro-viscosity, micro-collision and micro-slip on the interface during vibration. Heller et al. [8] analyzed the interface effect between the metal sheets tightened by two bolts. With the increase of bolt loosening, the interface effect between the metal sheets decreased, and the nonlinear vibration characteristics caused by the interface effect also weakened. Studies on the Relationship Between Anchor Force 89 Similar to the relationship between the bolt tightness and the interface effect of the metal sheet, the interface effect between the anchor plate and the concrete anchor pier of the anchor cable will also be affected by the anchoring force. It is expected that with the decrease of the anchoring force, the interface effect decreases and the nonlinear characteristics weaken. 2.2 AMD and Hilbert Transform The free vibration signal of the anchor head of prestressed anchor cable under exter- nal excitation has multiple frequencies. Firstly, improved AMD [9] (analytical mode decomposition) is used to decompose the first-order frequency component from the free vibration signal, and then Hilbert transform is performed on the decomposed component to obtain the frequency-amplitude curve, and then the nonlinear degree of the signal is analyzed. AMD constructs the orthogonal function of specific frequency by setting the binary cutoff frequency parameter, and conducts Hilbert transform on the product of it and signal, so as to filter out the components less than specific frequency [10]. AMD sets different binary cutoff frequency parameters to decompose each single-component signal in the original signal. For the mixed signal x(t), it is composed of n single-component signals n xi (t)(i = 1, 2, · · · , n), whose expression is: x(t) = d xid (t). i=1 The frequency of each single component ωi (t)(i = 1, 2, · · · , n) satisfies the follow- ing: |ω1 (t)| < ωb1 (t) < |ω2 (t)| < · · · < ωb(n−2) (t) < |ωn−1 (t)| < ωb(n−1) (t) < |ωn (t)| (1) where ωbi (t) ∈ (ωi (t), ωi+1 (t))(i = 1, 2, · · · , n − 1) is the binary cut-off frequency parameter set. By setting this parameter, each signal component of the original signal can be decomposed, as shown in Formula (2): d xi (t) = si (t), i = 1 (2) xid (t) = si (t) − si−1 (t), i = 2, · · · , n For non-stationary signals with time-varying frequency, si (t) can be calculated by Formula (3): si (t) = sin θbi (t) H x(t) cos θbi (t) − cos θbi (t) H x(t) sin θbi (t) (i = 1, 2, · · · , n − 1) (3) where θ bi (t) is the phase angle; H {·} is Hilbert transform. 90 H. Li and H. Cao For real signal x(t), the Hilbert transformation is as follows: 1 1 x(τ ) H [x(t)] = x(t) ∗ = dτ (4) πt π R t−τ Hilbert transform can get the amplitude and frequency of x(t), which can be calculated by the following formula: ⎧ ⎪ ⎨ A(t) = x2 (t) + H 2 [x(t)] ⎪ 1 d θ (t) 1 d H [x(t)] (5) ⎩ f (t) = = arctan 2π dt 2π dt x(t) The relationship between the amplitude, frequency and time of the signal can be obtained by Formula (5). 2.3 Calculation of Nonlinear Vibration Characteristics Compared with the high-order frequency component, the first-order frequency compo- nent is easily excited under external excitation, and is less affected by environmental noise. Therefore, according to the measured acceleration signal, the spectrum map is used to identify the binary cut-off frequency for AMD decomposition, and the first-order fre- quency component is separated. The first-order frequency component is integrated twice to obtain the displacement signal. The relationship between the amplitude, frequency and time of the signal is obtained by Hilbert transform. Using the frequency varia- tion corresponding to the same amplitude segment to quantify the nonlinear vibration characteristics of the specimen, the nonlinear index of the specimen can be obtained. 3 Dynamic Test of Anchor Cable 3.1 Design and Fabrication of Specimens A 700 mm × 700 mm × 300 mm concrete block (hereinafter referred to as the work- bench) is used to simulate the soil working with the anchor cable, and the PVC pipe is used in the center to retain a 30 mm diameter through the hole along the length direction. The steel strand with a nominal diameter of 15.2 mm is used as the anchor cable through the channel of the center of the workbench. The section of the specimen is shown in Fig. 1, the specific parameters are shown in Table 1, and the pouring and working state of the workbench are shown in Fig. 2. Considering two different anchor cable conditions, two specimens are designed: specimen 1 does not deal with, simulate the anchor head in good condition of prestressed anchor cable; the anchor plate of specimen 2 was corroded, and the prestressed anchor cable with serious corrosion of anchor head was simulated. The other parameters of the two specimens were the same. Studies on the Relationship Between Anchor Force 91 The workbench is firmly connected to the laboratory floor through four anchor bolts, as shown in Fig. 2(b). The concrete workbench has large quality and is closely connected with the ground anchorage. By knocking on the specimen, the measured vibration signal is analyzed, and it is found that the vibration amplitude of the workbench can be ignored. Fig. 1. The concrete workbench design Table 1. Specimen parameters Specimen Parameters Size of workbench (Length × width × height /mm) 700 × 700 × 300 Concrete grade C40 Nominal diameter of steel strand (mm) 15.2 Penetration channel diameter (mm) 30 Length of steel strand (mm) 200 + 700 + 600 Standard value of tensile strength of steel strand (Mpa) 1860 Design value of tensile strength of steel strand (MPa) 1320 Size of anchor plate (Length × width × thickness /mm) 120 × 120 × 20 Measuring range of acceleration sensor (g) ±5 Frequency response range of acceleration sensor (Hz) 0.2–1600 Axial sensitivity of acceleration sensor (mV/g) 1006 (a)Acceleration sensor under anchor plate (b)Pull-out apparatus in work Fig. 2. Drawing of specimen making and working state 92 H. Li and H. Cao 3.2 Experimental Process The maximum working tension of the steel strand in this test was calculated according to the domestic and foreign specifications, and the maximum tension was set to 100 kN considering the limitation of the tonnage of the drawing instrument for the test. The specific test steps are as follows: (1) The pull-out apparatus is used to load the load step by step, 10 kN is the first level, until 100 kN, each load is maintained for 5 min; (2) Under each level of load, the hammer imposes transverse impact load on the upper part of the anchor plate at the anchorage end. The free vibration acceleration signal is collected by the acceleration sensor attached to the bottom of the anchor plate, and the sampling frequency is 5000 Hz. (3) When the load reaches 100 kN, the load is unloaded step by step, and each step is kept for 5 min. The force hammer is used to excite the anchor plate, and the acceleration signal is collected. The sampling frequency is 5000 Hz. The test steps of the two specimens are the same. 4 Analysis of Relationship Between Nonlinear Vibration Characteristics and Tension 4.1 Nonlinear Index Taking the signal of a specimen1 under 50 KN tension as an example, the calculation of signal nonlinear index is introduced. For the measured signal, according to the method described in Sect. 2.3, the first- order frequency component is separated from the signal and the displacement signal is obtained by integration. The frequency-time curve and amplitude-time curve are obtained by Hilbert transform. The least square method is used to fit the frequency- amplitude curve according to the time correspondence. The curves were normalized, that is, the values of each ordinate of the curve were subtracted from their maximum values, and the frequency-amplitude curves after normalization at all levels of tension in the loading stage of specimen 1 were plotted as shown in Fig. 3. It can be seen in the figure that the first-order frequency decreases with the increase of amplitude at all levels of tension, showing nonlinear vibration characteristics. But the amplitude of the first- order frequency reduction given by these curves is different. In order to quantitatively compare the nonlinear vibration characteristics of specimens under different tension levels, the absolute value of frequency variation corresponding to the same amplitude section on the frequency-amplitude curve can be taken as a nonlinear index. The larger the nonlinear index is, the stronger the nonlinear characteristics are. Studies on the Relationship Between Anchor Force 93 It can be seen from Fig. 3 that when the tension increases from 30 kN to 40 kN, the amplitude-frequency curve decreases significantly. The analysis shows that when the tension is less than a certain value (30 kN), the asperity between the contact surface has a certain extrusion, showing interface effect, resulting in nonlinear vibration. When the tensile force exceeds this value, the contact surface is compacted, and the untouched asperities also turn to contact state. The interface effect is significantly enhanced, and the nonlinearity is significantly increased. Fig. 3. The specimen 1 frequency-amplitude curve 4.2 Relationship Between Nonlinear Index and Tension The frequency-amplitude curve of the first-order vibration component of the two spec- imens under different tensile forces is taken as the same amplitude segment (0.1–4) × 10–7 m, and the corresponding nonlinear index is calculated. The relationship between the nonlinear indexes of specimen 1 and specimen 2 and the tensile force is shown in Fig. 4 and Fig. 5. It can be seen that the tensile force of the two specimens is positively correlated with the nonlinear index, and the nonlinear index of specimen 2 is generally greater than that of specimen 1. 94 H. Li and H. Cao Fig. 4. The specimen 1 trend graph of nonlinear index Fig. 5. The specimen 2 trend graph of nonlinear index In order to investigate the variation of nonlinear index with tension in detail, the state of tension at 10kN is taken as the initial state, and the nonlinear index f 1 is the reference value. Define the percentage of nonlinear index corresponding to different tension: δ i = f i /f 1 , reflect the change degree of nonlinear index and initial state of different tension; define nonlinear index percentage difference: i = δ i − δ i-1 , reflecting the speed of nonlinear index change; similarly, the percentage of non-linear indicators under unloading condition is defined: δ u = f u /f 1 , and the percentage difference of non- linear indicators is defined: u = δ u − δ u-1 . The specific calculation results are shown in Table 2 and Table 3. Studies on the Relationship Between Anchor Force 95 Table 2. The specimen 1 comparison table of nonlinear index Tensioning Loading stage Unloading stage force P/kN Nonlinear Nonlinear Percentage Nonlinear Nonlinear Nonlinear index f i index difference index f u index index percentage of nonlinear percentage percentage δi index i δu difference u 10 0.408 100.0% 0.328 100.0% 20 0.482 118.1% 18.1% 0.513 156.4% 56.4% 30 0.658 161.3% 43.1% 0.587 179.0% 22.6% 40 1.132 277.5% 116.2% 0.764 232.9% 54.0% 50 1.249 306.1% 28.7% 0.936 285.4% 52.4% 60 1.475 361.5% 55.4% 1.371 418.0% 132.6% 70 1.513 370.8% 9.3% 1.626 495.7% 77.7% 80 1.626 398.5% 27.7% 1.768 539.0% 43.3% 90 1.783 436.5% 38.0% 1.781 543.6% 4.6% 100 1.855 454.7% 18.1% 1.855 565.5% 22.0% Table 3. The specimen 2 comparison table of nonlinear index P/kN Loading stage Unloading stage fi δi i fu δu u 10 0.541 100.0% 0.513 100.0% 20 0.788 145.6% 45.6% 0.853 166.3% 66.3% 30 1.195 220.9% 75.3% 0.980 191.1% 24.8% 40 1.371 253.4% 32.5% 1.175 229.2% 38.0% 50 1.508 278.6% 25.3% 1.356 264.5% 35.4% 60 1.582 292.3% 13.7% 1.456 283.9% 19.4% 70 1.731 319.9% 27.5% 1.711 333.7% 49.8% 80 1.755 324.4% 4.5% 1.700 331.4% −2.2% 90 1.793 331.3% 7.0% 1.814 353.7% 22.3% 100 1.828 337.9% 6.5% 1.828 356.5% 2.8% The following conclusions can be drawn from the above chart: (1) The change trends of nonlinear indexes of specimens 1 and 2 are similar, and the tension is positively correlated with the nonlinear indexes. This is because the 96 H. Li and H. Cao greater the tensile force, the closer the contact between the anchor plate and the concrete workbench, the stronger the interface effect; (2) With the increase of tensile force, the growth rate of nonlinear index slows down. This is because when the tension is large, the increase of contact tightness slows down; (3) The nonlinear index of specimen 2 is generally greater than that of specimen 1. This is because the corrosion surface roughness of the specimen 2 anchor plate is greater than that of the anchor plate without corrosion, which enhances the interface effect and increases the degree of nonlinear vibration. In summary, the nonlinear index of steel strand (prestressed anchor cable) always keeps the same direction as the tensile force (anchorage force), that is, the nonlinear index increases with the increase of tensile force (loading stage), and decreases with the decrease of tensile force(unloading stage). Therefore, the nonlinear index can be used as an index to determine the anchoring force of anchor cable. The corrosion of the anchor head has an increasing effect on the nonlinear index, which can be appropriately considered when estimating the anchorage force. 5 Conclusion By analyzing the relationship between anchorage force and nonlinear vibration charac- teristics, this paper explores the nondestructive testing method of anchorage force. The prestressed anchor cable is simulated by the steel strand through the concrete worktable. The free vibration acceleration signal of the anchor head of the anchor cable is collected under different tensile forces. The first-order frequency component is obtained by AMD decomposition, and the nonlinear vibration index of the anchor cable is obtained by Hilbert transform. The relationship between the nonlinear index of the anchor cable and the tensile force is studied. The following conclusions are obtained. (1) On the interface between the anchor plate of the prestressed anchor cable and the concrete workbench, the interface effect is obvious, and the lateral free vibration signal of the anchor plate has obvious nonlinear characteristics. Using AMD and Hilbert transform, the index reflecting the degree of nonlinearity can be successfully obtained. (2) The nonlinear index of prestressed anchor cable is positively correlated with the tensile force, and the nonlinear index is sensitive to the contact surface state. When the anchor head corrosion is large, the nonlinear index is larger. (3) If only considering the contribution of the interface effect of the anchor head to the nonlinear vibration, the size of the anchorage force of the anchor cable can be judged by using the nonlinear index as the basis and considering the corrosion condition of the anchor head. Nondestructive testing of anchorage force by nonlinear index is a feasible idea. Considering the corrosion degree of anchor head and the influence of field measured noise, further work is needed to study the non-destructive testing method of anchor cable anchorage force based on the practical application of this paper. Studies on the Relationship Between Anchor Force 97 References 1. Thanh, C.H., Dang, N.L., Kim, J.T.: PCA-based filtering of temperature effect on impedance monitoring in prestressed tendon anchorage. Smart Struct. Syst. 22(1), 57–70 (2018) 2. Kejun, H., Guanghui, W., Xueming, W., et al.: Testing technology and application of bolting and quality inspection of roadway based on principle of stress wave. Coal Technol. 36(04), 76–77 (2017) 3. Spalvier, A., et al.: Torsional vibration technique for the acoustoelastic characterization of concrete. Mater. Struct. 53(1), 67–78 (2020) 4. Alnuaimi, H.N., et al.: Monitoring concrete curing by linear and nonlinear ultrasonic methods. ACI Mater. J. 118(3), 61–69 (2021) 5. Lei, W., Canchang, L., Weixu, K., et al.: Nonlinear vibration analysis of a nano-beam considering gas diffusion surface stresses. J. Vibr. Shock 40(10), 50–56 (2021) 6. Jun, W., Hui, C.: Discriminant analysis of concrete debonding of CFST based on nonlinear vibration characteristics. J. Civ. Archit. Environ. Eng. 40(01), 48–54 (2018) 7. Hui, C., Yaxiang, L.: Nonlinear vibration characteristics and de-bonding recognition of concrete-filled steel tube columns. J. Vibr. Shock 39(01), 260–265 (2020) 8. Heller, L., Foltête, E., Piranda, J.: Experimental identification of nonlinear dynamic properties of built-up structures. J. Sound Vib. 327(1), 183–196 (2009) 9. Zhixiang, H., Zuocai, W., Yixin, R., et al.: On the analytical mode decomposition theory and algorithm for discrete vibration signal processing. J. Vibr. Eng. 29(02), 348–355 (2016) 10. Genda, C., Zuocai, W.: A signal decomposition theorem with Hilbert transform and its appli- cation to narrowband time series with closely spaced frequency components. Mech. Syst. Signal Process. 28, 258–279 (2012) Interrelationships of Load and Displacement of Barrette Piles for Various Interpretation Criteria Subjected to Uplift Loading Yit-Jin Chen(B) , Anjerick Topacio, and Suneelkumar Laveti Chung Yuan Christian University, Taoyuan, Taiwan

Abstract. This paper evaluates various interpretation criteria for barrette piles subjected to uplift loading conditions. Eight load test results were gathered and employed for the analysis in order to determine the application of these interpreta- tion criteria to barrette piles. The database was divided into drained and undrained soil conditions. Analysis of each of the interpretation criteria was performed in relation to the displacement ranges of each of the interpreted capacities. It was found out that the interpretation criteria L1 provided the initial linear elastic stage or the serviceability design at mean displacements of 4.1 mm and 7.3 mm, respec- tively, for drained and undrained soil conditions. On the other hand, the interpreta- tion criteria of DeBeer, van der Veen, Terzaghi and Peck, Davisson, L2 , and slope tangent fell on the same ranges of interpreted capacities with mean displacements ranging from 15 to 25 mm for drained and from 21 to 34 mm for undrained soil conditions. Finally, the interpretation criteria of DIN4026, Fuller and Hoy, and Chin all over-estimate the capacity with mean displacement exceeding 40 mm for drained and 53 mm for undrained soil conditions. In addition, the interrelation- ships of the load and the displacement for each of the interpretation criteria were further analyzed. A normalized load-displacement curve was determined in order to assess the corresponding mean displacements at which each of these interpreta- tion criterion’s loads are mobilizing along the curve. Statistical analysis was also applied to determine the consistency and reliability of each of the interpretation criteria. Normalized load-displacement equations for barrette piles subjected to uplift loading condition were also calculated for both drained and undrained soil conditions to be utilized and recommended for engineering practice and design of barrette piles for uplift loading. Keywords: Uplift loading · Barrette piles · Displacement · Interpretation criteria · Load test 1 Introduction Various conditions of a pile foundations can lead into various load (Q) – displacement (ρ) curve types that is gathered from axial load tests on such foundations. These varieties may exhibit any one of three shapes, A, B, or C, as shown in Fig. 1. But due to the © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 98–107, 2023. https://doi.org/10.1007/978-981-19-3983-9_9 Interrelationships of Load and Displacement of Barrette 99 requirements of structures that can only withstand a range of displacements, most of the load-displacement curves that are gathered from load test results resemble that of curve C. This may pose a dilemma as the capacity of the pile is not clearly visible on such condition of the load-displacement curve. Therefore, the capacity almost always needs to be interpreted from the load test results. Interpretation criteria (e.g., [1–10]) have been proposed over the years for interpreting such failure load. Table 1 defines nine representative criteria for the interpreted failure load based on a variety of assumptions, individual judgments, extrapolations, and others from the measured load–displacement curve. As found in practice, these interpretation criteria will give different results that can vary substantially. Fig. 1. Typical load–displacement curves for pile foundations With these uncertainties in the interpretation of the capacity of a foundation, it is of utmost importance to analyze the application of these interpretation on various condi- tions and pile types. These load test data may provide vital information in determining the effects of different loading conditions to various soil and pile properties. Various researchers have also compiled relational databases of axial load test on different types of piles [11–16]. And since the 1980s, Kulhawy and co-workers have examined this issue in detail for drilled foundations. Their research [9, 10] and [17–20] mainly focused on the L1 (elastic limit) – L2 (failure threshold) method. Later, Chen and co-authors ([13, 16] and [21–24]) performed a more extensive evaluation to cover the existing represen- tative uplift and compression interpretation criteria for various soil and pile types. What lacked in these analyses is a detailed comparison of various interpretation criteria when they are applied to barrette piles under uplift loading conditions. Therefore, in this paper, nine representative uplift interpretation methods are exam- ined in detail to assess their relative merits and their interrelationships. A database consisting of axial uplift load tests for barrette piles under drained and undrained soil conditions is used for this purpose. The results are compared statistically and graphically, and conclusions are reached for consistent use in practice. 100 Y.-J. Chen et al. Table 1. Definitions of representative uplift interpretation criteria for pile foundations Method Classification Definition of interpreted capacity, Q Value of Qult which is the ultimate load that gives a Mathematical van der Veen (1953) straight line when log (1-Q/Qult) is plotted versus total modeling settlement. Mathematical Load is equal to inverse slope, , of line Chin (1970) modeling with Q = load and s = total settlement. Terzaghi and Peck Settlement Load occurs at 1.0 in (25.4 mm) total settlement. (1967) limitation Settlement Load occurs at which change in slope on log-log total DeBeer (1970) limitation settlement curve. Settlement Minimum load occurs at a rate of plastic settlement of Fuller and Hoy (1970) limitation 0.05 in per ton (0.14 mm/kN). Settlement DIN4026 (1975) Load corresponds to displacement at 2.5% B. limitation Load occurs at a displacement equal to the pile elastic Graphical compression line, , plus 0.15 in (3.8 mm) + B (in Davisson (1972) construction inch or mm)/120, in which Q = load, D = depth, A = area, E = Young’s modulus, B = pile diameter. Load occurs at a displacement equal to the initial slope Graphical slope tangent (1985) of the load-displacement curve plus 0.15 in (3.8 mm) + construction B (in inch or mm)/120. L1 and L2 designate the elastic limit and failure thresh- Graphical old, respectively. Failure is defined qualitatively as the L1 - L2 (1989, 2002) construction load beyond which a small increase in load produces a significant increase in displacement. 2 Database The database that was utilized in this study consisted of eight (8) load test results of barrette piles under uplift loading conditions. These load tests were done both in drained and undrained soil conditions, thus, the database was further divided into the said soil conditions, respectively. Division of the database into drained and undrained groups is governed by the prominent soil type along the pile length of each load test. Table 2 shows the soil and pile parameters that have been utilized in the study for its analysis. It can also be calculated in the table that the average equivalent diameter of the database is at 1.98 m while average pile length is at 41.7 m ranging from 3.5 to 57.5 m. Interrelationships of Load and Displacement of Barrette 101 3 Interpreted Axial Uplift Capacity As discussed, nine different criteria were used to analyze the interpreted capacity, as given in Table 1. These criteria were selected because they represent various displacement ranges and may represent the distribution of the interpreted results from the lower, middle and higher ranges as seen in past researches. Table 3 shows the results of each of the interpreted capacities (Q) based on each of the methods and represents different ranges of the capacities. However, during extrapolation, some load tests were terminated before achieving the available interpolated values. Following the conclusions of Phoon and Tang [25], bias is deemed inside a reasonable range for extrapolation from a load test terminated at 75% or higher of the actual Davisson capacity which is around 133% or lower of the final terminated load from any load test. Thus, these interpreted results were denoted as greater than (>) the value of 133% of the terminated load. In addition to the results of the interpreted capacities, the relative displacements (ρ) are also determined in order to assess the location of each of the interpreted capacities Table 2. Soil and pile information for barrette piles Shaft Test location Soil layer Soil Width, Side, Equivalent Pile L/B No. description condition W (m) S (m) diameter B length (m) L (m) TPU1 Taipei, Clay, sand Undrained 1.20 2.70 2.03 34.50 17.0 Taiwan and rock TPU2 Taipei, Clay, sand Undrained 1.20 2.70 2.03 50.30 24.8 Taiwan and rock TPU3 Taipei, Clay and Drained 1.30 2.70 2.11 52.00 24.6 Taiwan silty gravels TPU4 Taipei, Silt sand, Drained 1.20 2.70 2.03 51.40 25.3 Taiwan silt clay TPU5 Bangkok, Soft and Undrained 1.50 3.00 2.40 57.50 24.0 Thailand hard clay and dense to very dense sand TPU6 Taipei, Silty clay, Drained 0.8 2.7 1.66 45.3 27.3 Taiwan sandy silt, and gravel TPU7 Taipei, Silty clay, Drained 0.80 2.60 1.63 39.50 24.2 Taiwan sandy silt, and gravel TPU8 Zurich, Loose to Drained 1.00 3.00 1.95 3.50 1.8 Switzerland dense moraine 102 Y.-J. Chen et al. Table 3. Interpreted uplift capacities utilizing various interpretation criteria Shaft Soil Interpreted capacity, Q(kN) no. condition L1 L2 DAV* ST* T&P* DeBeer DIN* F&H* VDV* Chin TPU3 Drained 12500 20000 22700 22400 22482 20400 24297 24600 21000 26259 TPU4 8000 15000 >15656 15400 >15656 13200 >15656 >15656 13000 >15656 TPU6 12500 20800 21000 20300 20549 20000 22300 23000 17000 25771 TPU7 8000 12000 12200 12100 12473 12500 13304 13600 11000 >14352 TPU8 9800 14000 15700 15600 15798 12700 >15960 13800 12000 >15656 TPU1 Undrained 20000 30000 32400 31200 31093 25000 33699 34000 26000 36785 TPU2 15000 29700 26400 26400 25302 25100 30549 32000 28000 38550 TPU5 35250 61500 62100 66700 57773 60500 69525 69000 60000 >71820 Note: *: DAV – Davisson, ST – Slope-tangent, T&P – Terzaghi and Peck, DIN – DIN4026, F&H – Fuller and Hoy, VDV – van der Veen and their distribution along the load-displacement curve as seen in Table 4. Furthermore, comparison between each of the interpreted capacities was done in order to assess where each of the interpretation methods are distributed along the normalized load displacement curve. In order to assess this, a normalizing interpretation method must be determined in order to check each of the other methods’ location in the curve in relation to the normalizing method. the L2 method was used as the normalizing criterion. This graphical method interprets the capacity as the start of the load–displacement curve’s final linear region. This method is effective for load–displacement curves resembling that of curves B and C with the application of a hyperbolic extension. After normalizing the interpreted capacities and calculating the mean values for both drained and undrained soil conditions, it can be found that the interpreted load of the L1 provided the initial linear elastic stage of the developed normalized load–displacement curve. It may be used to predict the serviceability load that can be resisted by barrette piles or designs that require displacements that do not exceed mean displacements of 4.1 mm and 7.3 mm, respectively for drained and undrained soil conditions. Interrelationships of Load and Displacement of Barrette 103 Table 4. Relative displacements utilizing various interpretation criteria Shaft Soil Relative displacement, ρ (mm) no. condition L1 L2 DAV* ST* T&P* DeBeer DIN* F&H* VDV* Chin TPU3 Drained 3.9 13.6 27.2 24.8 25.4 14.9 52.9 63.3 17.0 >63.3 TPU4 2.1 16.5 24.6 20.7 25.4 8.0 24.6 24.6 7.5 >24.6 TPU6 6.1 27.0 28.4 23.9 25.4 22.4 41.5 53.6 12.5 >53.6 TPU7 5.8 20.5 22.4 21.4 25.4 25.7 40.7 50.6 14.1 >50.6 TPU8 2.4 8.5 23.1 21.2 25.4 5.3 30.1 7.8 4.3 >30.1 TPU1 Undrained 5.5 20.6 34.4 26.0 25.4 9.9 50.8 56.8 11.2 >56.8 TPU2 8.5 44.6 28.9 28.9 25.4 24.8 50.8 65.0 35.3 >65.0 TPU5 8.0 32.0 33.3 46.7 25.4 30.0 59.8 57.0 29.1 >76.0 Note: *: DAV – Davisson, ST – slope-tangent, T&P – Terzaghi and Peck, DIN – DIN4026, F&H – Fuller and Hoy, VDV – van der Veen Most of the interpretation criteria fell on the transition region of the normalized load– displacement curve with the methods of L2 , Davisson, slope-tangent, Terzaghi and Peck, van der Veen, and DeBeer. These interpretation criteria provided good estimates of the capacity of barrette piles and are effective for designs that require mean displacements that do not exceed a range from 15 to 25 mm for drained and from 21 to 34 mm for undrained soil conditions. Lastly, the methods of Fuller and Hoy, DIN4026, and Chin had overestimated capacities and thus were un-conservative in interpreting the capacity of barrette piles for both drained and undrained soil conditions with mean displacements exceeding 40 mm for drained and 53 mm for undrained soil conditions. Graphical representation of the location for these methods on the normalized load- displacement curve can be seen in Figs. 2 and 3 for drained and undrained soil conditions, respectively. 104 Y.-J. Chen et al. Fig. 2. Normalized load displacement curve for barrette piles in drained soils Fig. 3. Normalized load displacement curve for barrette piles in undrained soils It can also be seen in the calculated results that the drained soils mobilize capacity at a lower displacement values in comparison to the undrained soil conditions. This means that lower capacities can be expected from sandy soils at lower displacements in comparison to clayey soils at slightly higher displacements. Normalized load-displacement curve equations are also computed based on the data that were interpreted for both drained and undrained soil conditions. These equations may help in simplifying the analysis of each interpreted capacities in relation to that of the normalizing method that is L2 . The equations are listed below for drained and undrained soil conditions, respectively. Q ρ = for drained soils r2 = 0.99 (1) QL2 3.21 + 0.82ρ Interrelationships of Load and Displacement of Barrette 105 Q ρ = for undrained soils r2 = 0.99 (2) QL2 7.09 + 0.78ρ Furthermore, the results of this preliminary analysis may be able to shed light on the behaviour of each of the interpretation criteria when applied to barrette piles under uplift loading conditions. In order to increase the reliability and decrease the uncertainty of the results of the analysis, additional load tests should be employed to the database. It is therefore recommended for future expansion of the study the addition of load test data, in order to increase the range of pile and soil properties included in the analysis. Also, analysis of the behaviour of the interpretation criteria to the side and tip resistances is advised in order to present a more robust comparison between the interpretation methods that are being studied. 4 Summary and Conclusions Axial uplift load test data were used to evaluate the capacity of barrette piles in various soil conditions. The database included 8 field uplift load tests, including 5 drained and 3 undrained soil conditions. Nine representative interpretation criteria were utilized to evaluate the available data. From these analyses, the following results were drawn: 1. L1 method provided the initial linear elastic stage or the serviceability region of the developed normalized load–displacement curve with mean displacements that do not exceed 4.1 mm and 7.3 mm, respectively, for drained and undrained soil conditions. 2. The methods of L2 , Davisson, slope-tangent, Terzaghi and Peck, van der Veen, and DeBeer are located at the transition region to the initial stage of the final linear region of the curve. These interpretations yielded at mean displacements that do not exceed a range from 15 to 25 mm for drained soil conditions. 3. For undrained soil conditions, the methods of L2 , Davisson, slope-tangent, Terzaghi and Peck, van der Veen, and DeBeer yielded at mean displacements ranging from 21 to 34 mm. 4. The methods of Fuller and Hoy, DIN4026, and Chin have overestimated capacities and thus were un-conservative in interpreting the capacity of barrette piles for both drained and undrained soil conditions. These methods have mean displacements exceeding 40 mm and 53 mm for drained and undrained soil conditions, respectively. 5. Normalized load-displacement curves and their respective equations have been pre- sented to be utilized for future designs of barrette piles in different soil conditions. ρ Drained soil conditions yielded an equation of QQL2 = 3.21+0.82ρ with an r 2 = 0.99; ρ while undrained soil conditions yielded an equation of QQL2 = 7.09+0.78ρ with an r 2 = 0.99. 6. In order to increase the reliability of the analysis, additional load test data is necessary for the analysis. Acknowledgements. This study was supported by the Ministry of Science and Technology of Taiwan under the contract number MOST 110–2221-E-033–010-MY2 and the John Su Foundation. 106 Y.-J. Chen et al. References 1. van der Veen, C.: Bearing capacity of a pile. In: Proceedings of the 3rd International Confer- ence on Soil Mechanics and Foundation Engineering, Zurich, Switzerland, pp. 16–27 August 1953. International Society for Soil Mechanics and Geotechnical Engineering, London, vol. 2, pp. 85–90 (1953) 2. Terzaghi, K., Peck, R.B.: Soil Mechanics in Engineering Practice. Wiley, New York, 2 (1967) 3. Chin, F.K.: Estimation of the ultimate load of piles not carried to failure. In: Proceedings of the 2nd Southeast Asian Conference on Soil Engineering, Singapore, pp. 81–90 (1970) 4. DeBeer, E.E.: Experimental determination of the shape factors and bearing capacity factors of sand. Géotechnique 20(4), 387–411 (1970). https://doi.org/10.1680/geot.1970.20.4.387 5. Fuller, F.M., Hoy, H.E.: Pile load tests including quick load test method, conventional methods, and interpretations. Highw. Res. Board 333, 74–86 (1970) 6. Davisson, M.T.: High capacity piles. In: Proceedings of the Lecture Series on Innovations in Foundation Construction, American Society of Civil Engineers, Illinois Section, Chicago, Ill. p. 52 (1972) 7. DIN 4026: Beiblatt, Rammpfahle, Herstellun Bemessung und Zulassige Belastun Eraulterun- gen (1975). https://www.umwelt-online.de/recht/bau/din/402664B._C 8. O’Rourke, T.D., Kulhawy, F.H.: Observations on load tests on drilled shafts. In: Proceedings of Drilled Piers and Caissons II. Edited by C.N. Baker. American Society of Civil Engineers, New York, pp. 113–128 (1985) 9. Hirany, A., Kulhawy, F.H.: Interpretation of load tests on drilled shafts. II: axial uplift. In: Kul- hawy, F.H. (ed.) Proceedings of the Foundation Engineering: Current Principles and Practices. GSP 22, pp. 1150–1159. American Society of Civil Engineers, New York (1989) 10. Hirany, A., Kulhawy, F.H.: On the interpretation of drilled foundation load test results. In: O’Neill, M.W., Townsend, F.C. (eds.) Proceedings of Deep Foundations. GSP 116, pp. 1018– 1028. American Society of Civil Engineers, Reston, Va (2002) 11. Long, J.H., Shimel, S.: Drilled shafts – A database approach. In: Proceedings of the Foundation Engineering Congress, pp. 1091–1108 (1989) 12. Wysockey, M.H., Long, J.H.: Utility of drilled shaft load test results. In: Proceedings of the International Conference on Design and Construction of Deep Foundations, pp. 1789–1803 (1994) 13. Marcos, M.C., Chen, Y.-J., Chang, K.C.: Evaluation of interpretation criteria for piles under compression loading. J. Adv. Eng. 9, 177–182 (2014) 14. Chen, Y.-J., Liao, M.-R., Lin, S.-S., Huang, J.-K., Marcos, M.C.: Development of an integrated web-based system with a pile load test database and pre-analyzed data. Geomech. Eng. 7(1), 37–53 (2014). https://doi.org/10.12989/gae.2014.7.1.037 15. Kumari, A., Thakare, S.W., Dhatrak, A.I.: Lateral and uplift capacities of barrette pile in sandy soil. In: Latha Gali, M., Raghuveer Rao, P. (eds.) Construction in Geotechnical Engineering. LNCE, vol. 84, pp. 215–235. Springer, Singapore (2020). https://doi.org/10.1007/978-981- 15-6090-3_15 16. Chen, Y.-J., Chu, T.-C., Topacio, A., Marcos, M.C.: Interrelationships of load and displace- ment of barrette piles for various interpretation criteria under drained loading. In: Proceedings of the 2021 International Conference on Civil, Materials, and Environmental Engineering, Malaysia, pp. 1–7 (2021) 17. Jeon, S.S., Kulhawy, F.H.: Evaluation of axial compression behavior of micropiles. In: Bran- don, T.L. (ed.) Proceedings of the Foundation and Ground Improvement. GSP 113. American Society of Civil Engineers, Reston, Va (2001) 18. Cushing, A.G., Kulhawy, F.H.: Drained elastic behavior of drilled shafts in cohesionless soils. In: O’Neill, M.W., Townsend, F.C. (eds.) Proceedings of DEEP Foundations. GSP 116, pp. 22–36. American Society of Civil Engineers, Reston, Va (2002) Interrelationships of Load and Displacement of Barrette 107 19. Kulhawy, F.H.: On the axial behavior of drilled foundations. In: Turner, J.P., Mayne, P.W. (eds.) Proceedings of Geosupport 2004: Drilled Shafts, Micropiling, Deep Mixing, Remedial Methods, and Specialty Foundation Systems. GSP 124, pp. 34–51. American Society of Civil Engineers, Reston, Va (2004) 20. Chen, J.-R.: Axial behavior of drilled shafts in gravelly soils. Ph.D. thesis, Department of Civil and Environmental Engineering, Cornell University, N.Y. (2004) 21. Chen, Y.-J., Chang, H.-W., Kulhawy, F.H.: Evaluation of uplift interpretation criteria for drilled shaft capacity. J. Geotech. Geoenvironmental Eng. 134(10), 1459–1468 (2008). https://doi. org/10.1061/(ASCE)1090-0241(2008)134:10(1459) 22. Chen, Y.-J., Fang, Y.-C.: Critical evaluation of compression interpretation criteria for drilled shafts. J. Geotech. Geoenvironmental Eng. 135(8), 1056–1069 (2009). https://doi.org/10. 1061/(ASCE)GT.1943-5606.0000027 23. Chen, Y.-J., Chu, T.-H.: Evaluation of uplift interpretation criteria for drilled shafts in gravelly soils. Can. Geotech. J. 49, 70–77 (2012) 24. Chen, Y.-J., Lin, W.-Y., Topacio, A., Phoon, K.-K.: Evaluation of interpretation criteria for drilled shafts with tip post grouting. Soils Found. J. 1–16 (2021) 25. Phoon, K.K., Tang, C.: Effect of extrapolation on interpreted capacity and model statistics of steel H-piles. Georisk: Assess. Manag. Risk Eng. Syst. Geohazards 13(4) 291–302 (2019). https://doi.org/10.1080/17499518.2019.1652920 Research on the Influence of Water Horse on the Vortex Induced Resonance Response of Bridges Bifeng Liu1,2 and Changzhao Qian1,2(B) 1 School of Civil Engineering and Architecture, Xiamen University of Technology, Xiamen 361024, Fujian, China

2 Fujian Provincial Key Laboratory of Wind Disaster and Wind Engineering, Xiamen 361024, Fujian, China Abstract. Vortex-induced resonance is the most common wind-induced vibration phenomenon in bridge structures. On May 5, 2020, the vortex-induced resonance phenomenon occurred on Humen Bridge. The reason is that the water horses continuously installed along the side guardrails of the bridge changed the steel box. The aerodynamic shape of the beam causes the wind-induced vortex-induced resonance phenomenon in the bridge due to the incoming wind. Therefore, it is necessary to study the influence of the water horse on the vortex-induced resonance performance of the bridge structure. In this paper, with the Xiamen Haicang Bridge as the background, a vortex-vibration wind tunnel test of a segment model with a scale ratio of 1:25 is carried out. The main girder segment model is used to arrange water at the windward side railings of the bridge when the wind attack angle is 0°. The effect of water horses on the railings on the windward and leeward side and the water horses on the railings on both sides on the bridge vortex induced resonance. Keywords: Wind tunnel test · Segment model · Bridge structure · Water horse · Vortex induced resonance 1 Introduction Vortex-induced resonance is a wind-induced vibration phenomenon that is very likely to occur in long-span bridges at low wind speeds [1]. It is a self-excited limiting vibration. It is caused by airflow passing through the surface of the main beam. Caused by vortices that fall off regularly at a certain time interval. In a certain range of wind speed, when the frequency of the vortex shedding is close to a certain natural frequency of the bridge, the bridge will produce vortex vibration. At the same time, the bridge vibration will give feedback to the shedding vortex, making a certain wind speed.The frequency of vortex shedding in the interval is locked around the natural frequency of the bridge and does not change with the change of wind speed [1]. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 108–119, 2023. https://doi.org/10.1007/978-981-19-3983-9_10 Research on the Influence of Water Horse on the Vortex 109 Vortex-induced resonance has occurred in many bridges at home and abroad, such as the Xihoumen Bridge in China, the Humen Bridge in China, the Tokyo Bay Passage Bridge in Japan, and the Kossock Cable-stayed Bridge in the United Kingdom etc. Therefore, in order to ensure the aerodynamic stability of the bridge structure, the study of vortex vibration is essential when the bridge is designed for wind resistance. At present, the research on vortex vibration phenomenon is difficult to use analytical methods. Therefore, the use of wind tunnel test technology is the most widely used method in the study of bridge vortex vibration phenomenon, and the segment model wind tunnel test is one of the most conventional test methods [2]. In order to study the vortex vibration performance of long-span bridges more accu- rately, the wind tunnel test of the segment model with large scale ratio is more and more adopted. When Larose et al. conducted a wind tunnel test of a segment model of the Stonecutter Bridge with a large scale ratio, they found that the Reynolds number has a direct effect on the vortex detachment characteristics when the bridge reaches the vortex lock interval [3]. Sun Yanguo, Liao Haili, etc. conducted wind tunnel tests on a segment model of a long-span suspension bridge with a scale ratio of 1:20 to study the vortex vibration experiment of the segment model, and analyzed the damping and wind attack. The different influences of other factors such as angle on the vortex vibration response of the bridge [4]; Li Yongle, Hou Guangyang etc.took a long-span highway suspension bridge as an example, produced a section model of the main girder with a scale ratio of 1:45 and performed the vortex vibration wind The tunnel test is used to study and analyze the influence of bridge railings, deflectors, different inspection lane positions and wind attack angles on the vortex-induced resonance response of the main girder. At the same time, aerodynamic control measures to improve the vortex vibration of the main girder are also proposed [5]. On May 5, 2020, the deck of the suspension bridge of the Humen Bridge in Guang- dong experienced large vibrations, which aroused strong attention from the academic community. According to the discussion results of the experts, the main reason for the vibration of the Humen Bridge this time was that the water horses were continuously installed along the side guardrails of the bridge, which changed the aerodynamic shape of the steel box girder. The phenomenon of vortex induced resonance. Therefore, this paper uses the Xiamen Haicang Bridge as the research background, and uses a 1:25 scaled segment model to conduct wind tunnel tests to study the response of the water horse on the bridge deck to the vortex induced resonance of a long-span bridge. 2 Engineering Background The research in this paper is based on the Xiamen Haicang Bridge. The Xiamen Haicang Bridge is a three-span continuous full-floating steel box girder suspension bridge con- necting Xiamen’s main island and Xiamen’s Haicang District. The span is 230 m + 648 m + 230 m, which is a large-span flexible bridge structure. Due to the geographical environment and wind load, the wind-induced vibration response of the bridge is very obvious, and it causes structural fatigue damage. In order to ensure the safety of the structure of the bridge against wind, the wind tunnel test of the Haixifeng Engineer- ing Research Center of Xiamen University of Technology was carried out to study the 110 B. Liu and C. Qian vortex induced resonance of the bridge, and the effect of water horses on the vortex vibration response of the bridge was extended to different positions on the bridge deck. The standard cross section of the bridge stiffener is shown in Fig. 1. Fig. 1. Standard cross section view of stiffening beam 3 Overview of Wind Tunnel Test The vortex-induced resonance wind tunnel test of the 1:25 main girder segment model of the Haicang Bridge was carried out in the Wind Tunnel Laboratory of the Haixifeng Engineering Research Center of Xiamen University of Technology, as shown in Fig. 2. This segment model A total of 4 test conditions were studied in wind tunnel tests, including bare bridge status, water horses at the windward side railings, water horses at the leeward side railings, and water horses at both side rails. The water horses are also in accordance with actual engineering. The size used is made according to the scale ratio of 1:25, and its size is: 4 m × 0.032 m, and the test conditions involved are all completed in a uniform flow field at a wind attack angle of 0°. Fig. 2. Segment model wind tunnel test Research on the Influence of Water Horse on the Vortex 111 3.1 Segment Model Design The geometric scale ratio of the main girder segment model used for the vortex wind tunnel test is λL = 1:25. In order to reduce the influence of the three-dimensional flow at the end of the segment model, the length of the main girder model is taken as L = 4 m and the main girder width B = 1.28 m, the model height is H = 0.12 m, and the model aspect ratio is about 3.125. Use ANSYS software to establish the structural model of the Haicang Bridge, and perform dynamic response analysis to obtain the structural dynamic characteristic parameters, so that the correspondence between the main parameters of the real bridge and the main parameters of the segment model can be determined (Table 1). In order to make the segment model’s own structure have good rigidity performance, the main girder segment model is composed of high-strength steel frame and light wood cladding to fully ensure the similarity of geometric shapes. The railings on the main girder are made of acrylic panels in proportion to the fineness. The shape and air permeability of the railing are simulated. Table 1. Design parameter table of segment model Parameter Symbol Unit Design value Length of main beam L m 4 Width B m 1.28 Heigth H m 0.12 Equivalent mass meq kg/m 229 Mass moment of inertia Imeq kg·m2 /m 49.33 Verticalfundamental frequency fh Hz 1.625 Torsion fundamental frequency ft Hz 3.629 Torsion frequency ratio ε / 2.233 Vertical bending wind speed ratio mh / 2.5 Torsional wind speed ratio mt / 3.3 3.2 Allowable Value of Vortex Induced Resonance Amplitude According to the “Code for Wind Resistance Design of Highway Bridges” [6], the allowable amplitude values of vertical vortex-induced resonance and torsional vortex- induced resonance when the existing bridge of Xiamen Haicang Bridge is completed are: [ha ] = 0.04/fv = 0.2384 m [θa ] = 4.56/Bfv = 0.2890◦ 112 B. Liu and C. Qian 4 Analysis of Test Results 1. Figures 3 and 4 are the vertical displacement amplitude-wind speed curve and the torsional displacement amplitude-wind speed curve of the main beam segment model at 0° wind attack angle. Fig. 3. Vertical displacement amplitude-wind speed Fig. 4. Torsional displacement amplitude-wind speed Figure 5 is the vertical vibration time history diagram corresponding to the four working conditions when the wind speed is 5 m/s. Figures 6 to 8 are the bare bridge and the windward side railing at the wind speed 5 m/s with the water horses and the two side railings. Vertical vortex frequency corresponding to water horse. It can be seen from Figs. 3 and 5 that when the wind speed is 5 m/s, the bare bridge, the water horses at the windward side railings, and the water horses at the railings on both sides all Research on the Influence of Water Horse on the Vortex 113 have vertical vortex-induced resonance, and the amplitude is 0.01153 m respectively, 0.109 m, 0.1176 m, the frequency when the vibration occurs and the vertical fundamental frequency of this wind speed are both 0.1625 Hz (as shown in Fig. 7–9), and the amplitude of the vertical vortex-induced resonance displacement of the main beam is less than “The allowable value of the “Code for Design of Wind Resistance of Highway Bridges”, while the bridge structure does not have vertical vortex-induced resonance when the water horse is arranged on the leeward side. Fig. 5. 5 m/s vertical vibration time history chart Fig. 6. 5 m/s vertical vortex frequency of bare bridge 114 B. Liu and C. Qian Fig. 7. Vertical vortex frequency of windward side Fig. 8. Vertical vortex frequency of leeward 2. When the water horse is arranged at the railing on the windward side of the main beam segment model, Fig. 9 and Fig. 10 are the torsional vibration time-history diagram and torsional vortex vibration frequency corresponding to the main beam when the wind speed is 8.75 m/s, as shown in Fig. 4, It can be seen that the torsional vortex induced resonance occurs in the wind speed range of 7.5 m/s–10 m/s, with an amplitude of 0.1066°, and its vibration frequency is consistent with the torsional fundamental frequency at this wind speed, which is 0.4799 Hz (Fig. 10) Shown). Figure 11 and Fig. 12 are the torsional vibration time-history diagram and torsional vortex frequency corresponding to the main beam when the wind speed is 15 m/s. It can be seen from Fig. 4 that torsion occurs in the wind speed range of 11.25 m/s– 16.25 m/s For vortex-induced resonance, the wind speed with the largest amplitude of 6 m/s is selected for analysis. Its amplitude is 0.3362°, which is greater than the allowable value of the specification. The vibration frequency and the torsional fundamental frequency at this wind speed are both 0.4799 Hz (as shown in Fig. 12). Research on the Influence of Water Horse on the Vortex 115 Fig. 9. 8.75 m/s time history diagram of torsional vibration Fig. 10. 8.75 m/s torsional vortex frequency Fig. 11. 15 m/s time history diagram of torsional vibration 116 B. Liu and C. Qian Fig. 12. 15 m/s torsional vortex frequency 3. When water horses are arranged at the railings on both sides of the main girder segment model, Fig. 13 and Fig. 14 are the torsional vibration time-history diagram and torsional vortex vibration frequency corresponding to the main girder when the wind speed is 9.9 m/s. It can be seen that the torsional vortex induced resonance occurs in the wind speed range of 8.25 m/s–11.55 m/s, with an amplitude of 0.0492°, and its vibration frequency is consistent with the vertical fundamental frequency at this wind speed, which is 0.4799 Hz (As shown in Fig. 9). Figure 15 and Fig. 16 are the torsional vibration time-history diagram and torsional vortex vibration frequency corresponding to the main beam when the wind speed is 23.1 m/s. It can be seen from Fig. 4 that it occurs in the wind speed range of 16.5 m/s–24.75 m/s For torsional vortex induced resonance, the wind speed with the largest amplitude of 23.1 m/s is selected for analysis. Its amplitude is 0.5806°, which is much larger than the allowable value of the specification. The vibration frequency and the vertical fundamental frequency at this wind speed are both 0.4799 Hz (as shown in the Fig. 16). Fig. 13. 9.9 m/s time history diagram of torsional vibration Research on the Influence of Water Horse on the Vortex 117 Fig. 14. 9.9 m/s torsional vortex frequency Fig. 15. 23.1 m/s time history diagram of torsional vibration Fig. 16. 23.1 m/s torsional vortex frequency 118 B. Liu and C. Qian 5 Conclusion The vortex induced resonance phenomenon of the main girder structure generally occurs in the lower wind speed range, and it is very sensitive to the changes in the aerodynamic shape of the bridge. The water horses arranged on the bridge will change the aerodynamic shape of the bridge structure, making the main girder structure It is no longer the original streamlined section. When subjected to wind load, due to being blocked by the water horse, a larger vortex will be formed. The original bare bridge is very different. When the frequency of the force is close to the natural frequency of the bridge structure, the bridge will have a vortex-induced resonance phenomenon, and the amplitude is very large. Comparing the wind tunnel test on the influence of the vortex-induced resonance response of the bridge when the water horses are arranged at different positions at a wind attack angle of 0°, the following conclusions can be drawn. 1) When water horses are arranged at the windward side railings of the bridge deck, water horses are arranged at the railings on both sides, and the bare bridge is in the state of the main girder when the wind speed is 5 m/s, vertical vortex induced reso- nance occurs, and the amplitudes are 0.109 m and 0.1176 m respectively, 0.01153 m are within the allowable value of the specification, and the corresponding vertical vortex frequency and vertical fundamental frequency are consistent with 0.1625 Hz, 2) When the water horse is arranged at the railing on the windward side of the bridge, the main girder has torsional vortex induced resonance, and the vortex vibration lock interval is 7.5 m/s–10 m/s and 11.25 m/s–16.25 m/s, the largest The torsion amplitude is 0.3362°, which is greater than the allowable value in the specification. When the water horses are arranged at the railings on both sides of the bridge deck, the main beam also has torsional vortex induced resonance, and the vortex vibration lock interval is 8.25 m/s–11.55 m/s and 16.5 m/s–24.75 m/s, the maximum torsion The amplitude of 0.5806° is much larger than the allowable value of the specification. It can be concluded that: arranging water horses on the windward side railings and the railings on both sides of the bridge will cause the main beam to produce vertical vortex induced resonance and torsional vortex induced resonance, and the torsional vortex induced resonance is the largest in the two states The amplitude exceeds the allowable value of the specification. At the same time, the vertical vortex-induced resonance amplitude and the torsional vortex-induced resonance amplitude of the water horses arranged at the railings on both sides are larger than the vortex vibration amplitudes of the water horses arranged at the windward side railings. 3) When the water horse is arranged on the leeward side railing of the bridge deck, there is no vortex-induced resonance phenomenon in the main beam. It may be that when the water horse is arranged on the leeward side, the bridge deck does not easily cause aerodynamic flow, resulting in no vortex-induced resonance of the main beam. Acknowledgement. The authors would like to acknowledge with great gratitude for the supports of the Science and Technology Project in Xiamen (Grant No: 3502Z20183050), National Science Foundation of Fujian Province (Grant No: 2019J01866), National Science Foundation of China (Grant No: 52178510), Science and Technology Project of Construction in Xiamen (Grant No: XJK-2021–12). Research on the Influence of Water Horse on the Vortex 119 References 1. Huayang, L.: Analysis of Vortex-Induced Vibration Response of Long-Span Suspension Bridge Suspenders. Zhengzhou University, Zhengzhou (2018) 2. Li, Z., Yaojun, G.: Research on vortex induced resonance test of segment model of Shanghai bridge vehicle-bridge system. Chin. Civil Eng. J. 40(8), 61–66 (2007) 3. Larose, G.L., Larsen, S.V., Larsen, A., et al.: Sectional model experiments at high reynolds number for the deck of a 1018 m span cable-stayed bridge. In: Proceeding of 11th International Conference on Wind Engineering. TTU Press, Lubbock, pp. 373–380 (2003) 4. Yanguo, S., Haili, L., Mingshui, L.: Suspension bridge vortex vibration suppression measures based on segment model test. J. Southwest Jiaotong Univ. 47(2), 218–223 (2012) 5. Li, Y., Hou, G., Xiang, Q., et al.: Wind tunnel test study on optimization of vortex vibration performance of steel box girder of long-span suspension bridge. Acta Aerodyn. 29(6), 702–708 (2011) 6. Ministry of communications of the People’s Republic of China. JTG/T D60–01–2004 Code for Wind Resistance Design of Highway Bridges. People’s Communications Publishing House, Beijing (2004) The Degradation of Avalanche Anchorage Systems C. Paglia(B) and C. Mosca DACD, University of Applied Sciences of Southern Switzerland, Trevano, CP 12, 6952 Canobbio, Switzerland

Abstract. Avalanche protection systems are decisive to protect urban zones in mountain locations during winter. The anchor rods are firmly inserted in the ground down to the base of the rock and are mainly made of steel incorporated in injection cementitious grouts. In the present work, 30 to 40 years old anchorages from the Alps were extracted until 4 m in depth and the degradation was investigated. They generally exhibited a good conservation state. This was also due to the relatively low aggressiveness of ground-infiltrating waters. The carbonation of the injection mortars was very limited and corrosion was present only at an initial stage, along zones no longer than 30 cm and not around the entire steel bar perimeter. Exfo- liation and localized corrosion with small craters were rarely present. This latter intensified corrosion was identified in zones where an incomplete wrapping of the anchor bar with injection grout was observed. The same behavior was present where the polymeric nets did not expand during the grout injection process and remained in contact with the steel. This was particularly seen in the lower parts of the anchorages down to 3.0 m, where the cementitious material has more dif- ficulties to be injected and to embed the steel. In this concern, Oxygen-rich or depleted zones did not create corrosion macro-elements. Degradation susceptible zones, such as threaded heads portions or steel protection sheaths exhibited a slight uniform corrosion, but no reduction of the anchorage diameters. The bolts did not shifted during the tensile extraction tests and loads around 300 kN were registered. Therefore, uniform corrosion on threaded anchorage heads did not influence the bolt resistance. Keywords: Avalanche · Steel · Anchorage · Degradation 1 Introduction Pre-stressed cables and anchorages are widely used in the construction sector. They are placed in bridges, within cooling towers, in water tanks, off-shore tank, water plants, wall anchorages and foundations. They are mainly composed of an anchorage head with a fixed plate, a twisted structure with an injection/deareation channel an enlargement and a steel bar protection sheath. A cement-based injection grout is injected around the metal bars to protect them from corrosion. The durability of anchorage systems is influenced by the development of high strength cementitious injection materials, the © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 120–127, 2023. https://doi.org/10.1007/978-981-19-3983-9_11 The Degradation of Avalanche Anchorage Systems 121 content of corrosion inhibitors or silica fume and steel alloying elements [1]. Several alternative protection systems to steel are used: PVC, HDPE and PP sheaths. Addition- ally non-destructive diagnostic methods such as corrosion potential, polarization resis- tance, galvanostatic pulse, impedance spectroscopy, ultrasonic measurements, impact echo, radiography, radar and others contribute to control the degradation [2, 3]. Lately, anchorage or pre-stressed cables are put in place with an electrical separation from other metal parts of the structure, in particular rebars or foundation. Steel anchorage are often used in walls or in slope static issues. In this case, the anchorage-mortar system are in contact with the ground or the rocks. The steel anchorage are important parts of avalanche defense structures. These are widely widespread in mountain regions, in particular along the Alps. Under these extreme atmospheric conditions, the durability assumes a decisive role in the safety and functionality of the structures. In order to gain experience in the degradation of the steel anchorage and sheaths up to 40 years old, systems have been extracted from the soil and destructive investigations were carried out. 2 Experimental Procedure The anchorage were extracted from the soil. Twelve samples up to 4 m long, including the embedded cementitious material were brought to the laboratory. Visual and optical (binocular lens) investigations were carried out. The carbonation depth in the mortar was detected with Phenolphtalein [4], while the microstructure was observed with an optical microscope. The anchorage bolt extraction tests were carried out with a conventional tensile testing equipment. The Brinell hardness was measured according to the norm EN ISO 6506 [5] on the surface and on the anchorage section. For the comparison between hardness and tensile strength were used the tables displayed in the norm EN 18265 [6]. The main parameters of the chemistry of groundwater and a geology overview were also analyzed. 3 Results and Discussion The old anchorages exhibited a nominal diameter of 36 mm and were still partially wrapped in the mortar. The mortar diameter reached values up to 12 cm. In some cases, a decentralized position of the anchorage with respect to the mortar was observed (Fig. 1). This may be accounted for a difficult embedment and penetration of the cementitious material along the entire length of the system, especially with depth. The mortar cover thickness varied from 0 to 120 mm, depending on the position. Locally, steel parts were directly exposed to the surface, i.e. to the soil and groundwater. The steel parts embedded in mortar did not exhibit a significant corrosion. This was also due to the general low carbonation rate of the injection grout into the ground, with 3–4 mm over 40 years. Mortar micro-cracks longitudinal, transverse and radial to the anchorage were also observed. These may be caused in situ during slope slowly movements or more 122 C. Paglia and C. Mosca Fig. 1. Cross section of the anchorage. Note the decentralized position of the steel bar. likely during extraction tests or transportation, since no particular enhanced ingress of aggressive agents or degradation is observed on steel surface in the crack regions. The mortar leaching phenomena were not significant, despite the sporadic presence of water. The water exhibited a pH around 8, corrosive for steel. The water carbonate hardness was 27.3 °F, typical for bicarbonate-rich water and not particularly aggressive for mortars (leaching). The water sulphate content of 46.5 mg/l (200–600 mg/l) as well as the Ammonium content <2.5 mg/l (15–30 mg/l) were below the limit of a slightly aggressive environment for cementitious material [7]. These latter chemical parameters were a consequence of the geology. The first 40 cm of the soil were composed of aerated blocks and debris, while below 40 cm, the ground was humid and clay-rich. These conditions may promote the formation of corrosion macro-elements. Nonetheless, no signs of such interaction was observed in the corrosion feature of the anchorage examined. The systems with a steel sheath protection exhibited a slight uniform corrosion on the external side. On the internal side, the corrosion was only at an initial stage (Fig. 2). Fig. 2. Systems with a steel sheath protection. External side (upper part), internal side. The Degradation of Avalanche Anchorage Systems 123 The sheaths had a diameter of 64 mm and a wall thickness of 3 mm. The steel sheath were Zinc coated and the gap between the steel bar and the sheath was filled with mortar. The latter was still very present, not leached and no significant sign of degradation was present. Some systems exhibited a polymer net that partially did not expand during the mortar injection and remained in contact with the anchorage (Fig. 3). Fig. 3. Anchorage system with polymer net. In this case, the anchorage diameter was 26 mm. Where the net remained in contact with the bar, corrosion was increased. This was due to the lack of mortar basicity around the metal (Fig. 4). Fig. 4. Contact zone between the polymer net and the steel bar. Corrosion is present. For Zinc-coated steel wired anchorages a variable intensity of corrosion was also observed, mainly due to the different incorporation of the anchor rod in the injection mor- tar. In general, their degradation also depended on the system used: polymeric material- based, protection sheath presence and zinc-coated steel wires. All these systems may also play a decisive role. The general conservation state of the anchorages was generally good. Despite the systems reach up to 40 years. Nonetheless, a more detail look in the corrosion add some 124 C. Paglia and C. Mosca further information. The upper part of the anchorage exhibited a general slight uniform corrosion, in particular the threaded heads with the bolts (Fig. 5). Fig. 5. Upper part of the anchorage around the head-plate fixing system. The corrosion was different with depth. Generally, the upper parts, such as threaded heads, exhibited a uniform corrosion, mostly caused by the atmospheric condition (humidity and oxygen). In the middle parts, the corrosion was often absent, while a more advanced stage was present in the lower regions within the ground (Fig. 6). Fig. 6. Corrosion of the anchorage at different depths (left to right). The Degradation of Avalanche Anchorage Systems 125 In an initial corrosion stage, the ribs of the steel bars were preferentially attacked (Fig. 7). Fig. 7. Corrosion starting on the rib’s edge. In the advanced stage, a localized corrosion was observed, in particular exfoliation and small craters formation (Fig. 8). Fig. 8. Exfoliation corrosion (left, rib’s width 2 mm) foils ca. 200 m thick and localized corrosion phenomena with craters (right, craters dimeter 0.4 mm). Similar advanced corrosion features were present in zones of the steel in contact with the polymer net. Additionally, polymer debris also remained attached to the metal surface, thus promoting the acidity and the degradation. Generally, a change from a slight uniform corrosion near the ground surface, to a more localized corrosion with craters with depth (acqueous substances attack) was observed. The extraction tensile tests carried out on the anchorage bolt reached loads around 300 kN. Only duptile deformation (necking) below the bolts was observed for all anchorage (Fig. 9). 126 C. Paglia and C. Mosca Fig. 9. Extraction tensile tests on anchorage bolts and fracture location. The bolts did not shifted during the tests. Therefore, the uniform corrosion on threaded heads did not influence the bolt resistance (Fig. 9). The hardness measure- ments on the anchorage surface and on its section exhibited values decreasing from the external to the internal region of the anchorage because of a surface treatment. The values varied from 255 HB (external zone) to 200 HB (intermediate zone) and 165 HB (central zone). Comparison tables between hardness and tensile strength [6] indicated values from 550 N/mm2 for the central zone, 700 N/mm2 for the intermediate zone and 850 N/mm2 for the anchorage perimeter. No hardness and no tensile strength difference were detected between corroded and uncorroded regions. A further confirmation of the reduced influence of the general slight corrosion observed on the mechanical properties. 4 Conclusions The anchor rods exhibited a general good conservation state with corrosion only at an initial stage. The carbonation underground was very limited. The localized corrosion was only present with exfoliation and small craters, in regions no longer than 30 cm on a total length of almost 400 cm. Intensified corrosion was present where the anchorage was not embedded in the injection cementitious grout. The injection grout to reach the lowest parts of the anchorage) or in parts where the polymeric net did not expand during the grout injection process and remained in contact with the steel anchorage. This was particularly observed in the lower part of the anchorage. The corrosion in the lowest part of anchorage was not to be influenced by macro-elements (interaction between oxygen-rich or depleted zones). Often corrosion did not involve the entire perimeter of the anchorage. The bolts did not shifted during the tensile extraction tests and loads around 300 kN were registered. Uniform corrosion on threaded anchorage head did not influence the bolt resistance. Uniform corrosion in the upper parts of anchorage (threaded heads portions, protection sheaths) did not reduce the diameter of the steel bars. Acknowledgements. The authors would like to thank S. Antonietti for the tensile testing and M. Heimgartner for the samples. The Degradation of Avalanche Anchorage Systems 127 References 1. Holst, A.: Korrosionsmonitoring und Bruchortung vorgespannter Zugglieder in Bauwerken, Institut für Baustoffe, Heft 197, Braunschweig (2007) 2. Elsener, B., Böhni, H., Bräunlich, R., Markees, A.: Zestörungsfreie Spannkanbelprüfung mit Reflektometrischer Impulsmessung, USTRA, Forschungsauftrag 81/93, Arbetisgruppe Brückenforschung, Mai 1997, 528 3. Faller, M., Werner, R.: Temporärer Korrosionsschutz von Spannglieder (TEKplus), Wis- senschaftlich Technische Begleitung der Umsetzung in die Praxis, EMPA Dübendorf, USTRA, Forschungsauftrag AGB 2002/026 (AGB) (2007) 4. Concrete construction, materials tests, porosity, norm SIA 162/1, test No. 7 5. EN ISO 6506 Metallic materials, Brinell hadrness test (2019) 6. EN 18265 Metallic materials, conversion of hardness values (2013) 7. EN 206/1, Concrete - Part 1: Specification, performance, production and conformity (2006) Modeling and Analysis of Shanghai Central Tower Yuchi Liu1(B) , Yiwen Chen2 , and Peitong Li3 1 Chang’an Dublin International College of Transportation, Chang’an University, Xi’an 710018, China

2 College of Architectural Science and Engineering, Yangzhou University, Yangzhou 225100, Jiangsu, China 3 UNITEC Institute of Technology, Auckland 1025, New Zealand Abstract. Shanghai Central Tower, as China’s landmark super high-rise building, has attracted wide attention since its construction. Because it is much higher than ordinary buildings, it also poses a challenge to construction and later maintenance. Therefore, the structural safety of Shanghai Center Tower needs to be highly valued. This article builds a model similar to Shanghai Central Tower by using ANSYS software. Through modal analysis, a series of analysis are carried out on the model. In the analysis process, the analysis data are compared with the actual situation, the model can be close to the real situation, and can provide some reference for the analysis of Shanghai Central Tower. At the same time, it is found that when the model is at some corresponding natural frequencies, the overall structure is more prone to dangerous states. These frequencies need to be avoided as much as possible. In addition, the effects of torsion and non-torsion of the structure are also discussed in the article. Keywords: Modal analysis · Structural analysis · Shanghai Central Tower · Ansys 1 Introduction Shanghai Center Tower, as a high-profile high-rise building, has its function and impor- tance. At the same time, because it exceeds the height and design form of ordinary buildings, its architectural structure form is also extremely complex. Shanghai Central Tower is the highest building of China and it located in Lujiazui Shanghai. The height of Shanghai Central Tower is 632 m, and the assumed weight is 800,000 tones. The land area of the project is 30,370 square meters, the above-ground construction area is 380,000 square meters, the underground construction area is 160,000 square meters. The design team is Gensler from America. It is also the newest landmark of Shanghai. Shanghai Center Tower’s structural facade is a unique curved surface, with curtain walls rotating [1]. Wind load is one of the main controlling factors of Shanghai Center Tower. Due to the geographical location of Shanghai in coastal areas, typhoon weather often occurs more frequently here [2]. Some researchers have found that the ratio © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 128–137, 2023. https://doi.org/10.1007/978-981-19-3983-9_12 Modeling and Analysis of Shanghai Central Tower 129 of hourly average wind speed to gradient time average wind speed in Shanghai is about 0.68 [3]. At the same time, the use of radiosonde also helps researchers obtain the data of typhoon wind profile [4]. At the same time, in view of the external influence of high-rise buildings, the field measurement method is also used in the research. These established systems provide useful information on wind effects and the dynamic characteristics of high-rise buildings [5]. This article mainly uses ANSYS software for modeling. Compare the data obtained from the model analysis with the real measurement data, and constantly adjust the parameters to be close to the real situation as much as possible. The analysis method mainly uses modal analysis to obtain the natural frequency and other data of the model. Under some natural frequencies, the structure of the model may be affected, which needs to be avoided in reality. 2 Methodology 2.1 Modal Analysis Overview One of the most basic and important types of structural dynamics analysis is “structural modal analysis”. Modal analysis is mainly used to calculate the vibration frequency and vibration pattern of the structure, so it can also be called frequency analysis or vibration analysis. Dynamical analysis can be divided into time domain analysis and frequency domain analysis, and modal analysis is the basic analysis type of dynamical frequency domain analysis. 2.2 Related Theory 2.2.1 Kinetic Control Equations The kinetic control equations can be expressed as differential equations. [M ]{ü} + [C]{u} + [K]{u} + {F} = 0 (1) where [M] = structural mass matrix, [C] = structural damping matrix, [K] = structural stiffness matrix, {F} = external load function with time, {u} = nodal displacement vector, {u} = nodal velocity vector, and {ü} = nodal acceleration vector. There is no need to consider the effect of external forces in the modal analysis of the structure; therefore, the dynamical control equation for modal analysis can be expressed as follows: [M ]{ü} + [C]{u} + [K]{u} = 0 (2) Ideally, the structure is vibrating without considering the damping effect, which is the so-called free vibration case, and the modal analysis can be described again as [M ]{ü} + [K]{u} = 0 (3) 130 Y. Liu et al. Further analysis of Eq. 3, assuming that the free vibration is harmonic response motion, that is, u = u0sin(ωt), Eq. 3 can be further described as ([K] − ω2[M ]){u} = 0 (4) Solving Eq. 4, the root of the equation is ωi2, the eigenvalue, where i ranges from 1 to the number of degrees of freedom N of the structure (in FEA, the number of degrees of freedom N generally does not exceed three times the number of nodes of the analytical model mesh). The square root of the eigenvalue is ωi, which is the intrinsic circular frequency, so that the structural vibration frequency (structural intrinsic frequency) fi can be obtained by the formula fi = ωi/2π. Finite element modal analysis can be performed to obtain fi or ωi, both of which can be used to describe the vibration frequency of the structure. The characteristic vector corresponding to the eigenvalues is {u}i. The eigenvector {u}i represents the vibration shape (mode) of the structure when it vibrates at the intrinsic frequency fi. 2.2.2 Vibration Frequency Influencing Factors Considering the simplest form of transformation for modal analysis (Eq. 4), its further transformation gives ω2 = [K]/([M]){u}). Thus, the modal frequency is proportional to the stiffness of the structure and inversely proportional to the mass of the structure. 2.3 Model Building 2.3.1 Method and Tool The finite element method is used to analyze the dynamic characteristics of the main tower section of the Shanghai Tower. For a more complex structure, it is difficult to assume the exact overall displacement function. In this case, the structure can be divided into a finite number of units, and a unified and relatively simple displacement function can be used inside the unit, while the structure is analyzed as a collection of these units. This analysis method is called the finite unit method. The CAE software we use is ANSYS, which is powerful, easy to operate, and has become the most popular FEA software in the world, and has been ranked first in the FEA competition for many years. We mainly use workbench of ANSYS to perform modal analysis. Models according to the general dimensions in Shanghai Tower, which are relatively less precise, but easy to generate meshes and calculate results. One of them is distorted and the other one is not, other parameters are the same, and the results of our analysis are based on these two models. The geometric properties of the model is first determined and adjust it to the actual mass and frequency. To simplify the analysis, the gap at the curtain wall is ignored because we believe that it has little effect on the structure. Figure 1 shows two basic models of Shanghai Center Tower. One is the model after torsion and the other is the model without torsion. These two models are both drawn by ANSYS. Modeling and Analysis of Shanghai Central Tower 131 Fig. 1. Shanghai Central Tower basic model 3 Result and Discussion Model analysis method is used in the analysis of Shanghai Central Tower model. Before using the first mock exam function of ANSYS, the basic parameters of the model need to be modified through engineering data module. Taking the density of the model as an example, the whole model needs to be selected first and the software will automatically identify the total volume of the model. The total weight of Shanghai Central Tower is eight hundred thousand tons [6]. According to the mass volume formula, the density value of the model can be roughly obtained. After obtaining the density value, the density value should be input into the corresponding module for subsequent analysis. After adjusting the parameters required by the model, the software can automatically start the analysis. In this analysis, the default mode order is 6, which means only the first six modes have to be analyzed. Through the software display window, the vibration modes and corresponding specific values of the model under different modes are pro- vided. In the process of comparing the value obtained from the analysis of the model with the real value, there may be a large gap compared to the real value. Therefore, in this case, the elastic modulus of the model needs to be adjusted. If the period is too large, it reflects that the material stiffness is small, and the elastic modulus needs to be increased [7]. 3.1 Model Analysis of Torsional Model Figure 2 mainly shows the vibration of the torsional model under different modes. According to the Fig. 3, and compared with the natural frequency of Shanghai Central Tower in reality, the first order frequency deviation is less than 1%, and the second frequency deviation is about 8%. Basically, within the acceptable range of deviation, it shows that the model can reflect the situation of Shanghai Central Tower approximately. 132 Y. Liu et al. Fig. 2. Vibration of different modes of torsional model 1 0.8 0.6 0.4 0.2 0 Mode 1 Mode 2 Mode 3 Mode 4 Mode 5 Mode 6 Fig. 3. The value of the natural frequency of torsional model Modeling and Analysis of Shanghai Central Tower 133 The natural frequency of the model is the largest at the sixth order. However, only from the digital images, it cannot fully reflect the vibration of the model. At this time, the more comprehensive vibration of the model through the solution information table of ANSYS becomes a better choice. Vibration of the model can be judged through three different directions. These three directions are automatically divided into x, y and Z axes in the software. Generally speaking, the change severity of the model along the Y axis can reflect the possible results of the model under the wind. By comparing the modes of different orders to determine, people can avoid the building reaching this frequency in future design. The participation factor can be also used as a good parameter to reflect the vibration severity of the building in this mode. As shown in the following Table 1, the absolute value of the participation factor is the largest in forth mode, and the ratio is larger than that of other modes. The results reflect that the frequency obtained in the fourth mode of the model needs to be avoided as much as possible, and it is more likely to be affected by the wind in reality. The building may be in dangerous condition. Table 1. Y-axis participation factor of torsion model Model Participation factor Ratio 1 35.563 0.1882 2 −91.197 0.4826 3 152.77 0.8084 4 −188.98 1.0000 5 −96.645 0.5114 6 39.675 0.2099 For the x-axis and z-axis, these two axes can mainly reflect the possible impact on the foundation of the building due to vibration. In the y-axis direction, the changes caused by vibration can be clearly displayed, while the changes of X and Z-axes are not easy to detect directly. In this case, the change of effective mass can be analyzed and shown in the Table 3. When the model vibrates, the internal structure and materials will swing, resulting in changes in the partial mass. The change of effective mass refers to the change of forces on different parts of the model due to the deviation of the building during vibration. If the force changes too much at this frequency, the building may be in danger of overturning. This situation should be avoided in engineering design and manufacturing. Table 2 shows that the effective stress of the building is the largest in the second mode in the X direction. The maximum effective stress in Z direction appears in the first mode. Therefore, the frequencies in model 1 and model 2 need to be avoided as much as possible. The building may be dangerous at these two frequencies. 134 Y. Liu et al. Table 2. The value of the effective stress Effective mass X-DIR Ratio% Z-DIR Ratio% 0.6849E + 08 8.57 0.2584E + 09 32.33 0.2521E + 09 31.55 0.7368E + 08 9.22 0.1156E + 09 14.46 0.2717E + 08 3.40 0.2409E + 08 3.02 0.1185E + 09 14.82 0.7694E + 08 9.63 0.5712E + 07 0.71 0.6422E + 07 0.80 0.7240E + 08 9.06 3.2 Model Analysis of Model Without Torsion Shanghai Central Tower has a special torsional structural character compared to the general structure. Therefore, an untwisted model is established and compared it with the previous twisted model. Fig. 4. Vibration of different modes of untwisted model Modeling and Analysis of Shanghai Central Tower 135 1.2 1 0.8 0.6 0.4 0.2 0 Mode 1 Mode 2 Mode 3 Mode 4 Mode 5 Mode 6 Fig. 5. The value of the natural frequency of untwisted model Figure 4 mainly shows the vibration of the untwisted model in different modes. Moreover, in Fig. 5, to make a more intuitive comparison with the torsional model analyzed before, the first six modes are selected in this study when analyzing the non- torsional model. The analysis data show that the natural frequency of the model without torsion is greater than that of the model after torsion. Table 3. Y-axis participation factor of untwisted model Model Participation factor Ratio 1 −49.250 1.0000 2 40.180 0.8158 3 36.801 0.7472 4 −28.820 0.5852 5 −23.116 0.4694 6 17.582 0.3570 136 Y. Liu et al. Table 4. The value of the effective stress of untwisted model Effective mass X-DIR Ratio% Z-DIR Ratio% 153.0 0.00 0.3250E + 09 36.79 0.3276E + 09 37.08 130.1 0.00 86.29 0.00 0.1855E + 09 21.00 0.1869E + 09 21.16 76.57 0.00 41.47 0.00 0.1030E + 09 11.66 0.1037E + 09 11.74 40.77 0.00 Table 3 and 4 show the information of untwisted model especially in Y axis and the value of the effective stress of untwisted model. The maximum value of the participation factor in the Y direction of the non-torsional model appears in the first mode, which is different from the torsional model. At the same time, the numerical value of participation factor is generally smaller than that of twisted model. In terms of effective quality, the results are also significantly different from the previous model. The maximum value appears in the second-order mode in X direction and the first-order mode in Z direction respectively. The foundation part of the non-torsional model may be affected in these two natural frequencies. 3.3 Comparison of Torsional and Non-torsional Models When the model is not twisted, the natural frequency of the structure is greater than the twisted model. The natural frequency of the structure is related to the stiffness and mass of the object. It is the inherent characteristic of the structure itself and is not affected by external forces. This shows that the torsion of the structure may affect the stiffness, resulting in the change of natural frequency. At the same time, in terms of the participation factor in the Y direction, the value of the participation factor of the untwisted model is too small, indicating that the untwisted model may have a stronger ability to resist the wind in reality. The effective mass of the model without torsion is also different from that after torsion. It indicates that torsion may change the properties of the structure to a great extent. 4 Conclusion Shanghai Central Tower as a landmark super high-rise building, its completion represents the advanced level of construction technology. In this research, model analysis method is used to analyze the natural vibration characteristics of Shanghai Central Tower model and the following conclusions are drawn. Compared with the natural frequency of Shanghai Central Tower in reality, it is basically within the acceptable range of deviation. It shows that the model can reflect the situation of Shanghai Central Tower approximately. In the Modeling and Analysis of Shanghai Central Tower 137 fourth mode, the vibration of the model changes greatly along the y-axis direction, and the wind may have a great impact on this frequency. In the torsion model, the effective mass change of X and Z axes reflects that the foundation part is more vulnerable to the influence of these two axes frequencies and reaches the maximum in the second-order mode and the first-order mode respectively. By comparing the torsion and non-torsion of the model, the natural frequency of the model will change after torsion. It indicates that torsion may lead to deviation of the natural frequency by affecting the structural stiffness. In future research, the model can still be improved to obtain more accurate data. If the basement is not considered in the model, the pile-soil area is equivalent. The internal force and deformation of pile-soil area cannot be obtained. Therefore, the model can be further improved in the future research, so as to strengthen the research on each structural part. Acknowledgement. Y. Liu, Y. Chen and P. Li contributed equally to this work and should be considered co-first authors. References 1. Xin, Z., et al.: The wind resistant design of Shanghai Center Tower. J. Build. Struct. 32(7), 1–7 2. Katagiri, J., Marukawa, H., Ohkuma, T., Fujii, K.: Effects of natural frequency ratio of across- wind to torsion on wind-induced responses of a high-rise building. Nihon Kenchiku Gakkai kōzōkei ronbunshū 63(507), 21–27 (1998) 3. Xu, H., Lin, N., Huang, M., Lou, W.: Design tropical cyclone wind speed when considering climate change. J. Struct. Eng. 146(5), 4020063 (2020) 4. Kepert, J.: The dynamics of boundary layer jets within the tropical cyclone core. Part I: linear theory. J. Atmos. Sci. 58(17), 2469–2484 (2001) 5. Fu, J.Y., Wu, J.R., Xu, A., Li, Q.S., Xiao, Y.Q.: Full-scale measurements of wind effects on Guangzhou west tower. Eng. Struct. 35, 120–139 (2012) 6. Jiemin, D., et al.: Critical issues of structural analysis for the Shanghai Center project. J. Build. Struct. 31(6), 1–3 (2010) 7. Wu, J., Xu, H., Zhang, Q.: Dynamic performance evaluation of Shanghai Tower under winds based on full-scale data. Struct. Des. Tall Spec. Build. 28(9) (2019) From Digitized Systemization to New Era of Autonomous Materials Facilitation in Architectural Design and Actuation Mohammed Abbood(B) and Ghada Al Slik Department of Architectural Engineering, College of Engineering, University of Baghdad, Baghdad, Iraq {mohammad.aboud,ghada.alslik}@coeng.uobaghdad.edu.iq Abstract. Based on the architectural approach of Human Acceptability towards built environments, the functionalization of advanced and renewable materials through part-autonomy to general systemization and structure-making could be considered as one of “the key factors” in sustainability. The contemporary states of material automation and autonomy methods are passed through several featuring orders, but in general, they are still not qualified to be considered as domesticated aspects of dealing with the natural resources through the build and construction orders in “the 4th industrial revolution”. This is the main problem of this paper and so, it will clarify the modeling, individual processing, and the emergence of such issues to rich the humanistic acceptability and interaction “as support- ive aspects”. Towards that, this paper approaches some of the basic methods of contemporary facilitation, formulation, and reformulation of enhanced materi- als and natural resources through the architectural domains to clarify the role of industry-architecture relation, which at the first look of the human-centered set make challenges for active/operative systemization. So, such a procedure also needs to provide a better understanding of the contemporary potentials of sus- tainable applied practices furthermore. In addition, the embodied intelligence of respondent materials would be characterized as a cornerstone of structural con- trol and micro/nanoelectromechanical systems of the upcoming era of individual autonomy, which would be covered in this paper as a new method of nature- based logics classification. Through four featural points of modeling technics, beauty characterization state in sustainable applications of post-digitalism, bio- based logic as the principle of additive design, and the smart processing as basic of functionalization of additive design, this paper provides a primary insight towards the faster, more efficient and effective solutions of fully autonomous driven of sustainable actuation and design prospects architecturally. Keywords: Autonomous materials intelligence · Bio-based logic · Ontological genotype 1 Introduction The first two decades of the 21st century were the in-between time zone of construction materials and the methodology of their applicating as enhancement processes between © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 138–145, 2023. https://doi.org/10.1007/978-981-19-3983-9_13 From Digitized Systemization to New Era 139 two main digital and post-digital eras of general actuation technologies. These two inte- grational periods provide effective potentials for the architectural design, technologies, fabrication, spatial actuation, and practices to build a contemporary platform of material utilization technologies and new properties understandings through the constructional orders. As clarification of the word ‘potential’ in this matter, Cambridge dictionary defines the term as (possible when the necessary conditions exist), or (someone’s or something’s ability to develop, achieve, or succeed), or (possible but not yet achieved), or (able to develop into something in the future when the necessary conditions exist), or (the possibility of something developing or happening in a particular way), and (natural qualities or abilities that mean that someone or something may or should succeed or achieve something) [1]. The outcomes of post digitalism as a new set of individual potentialities of actuation needed sort of appropriate featuring and guidelines to be raised as applicable method- ologies in the architectural domains. The experimental phase of aggregative processing reveals the advanced key aspect of controlling both expressive and operative divisions of individuality through the productional series of architectural effects. This paper explains the main approaches those together edit the structure of contemporary automation, as gradational progress from the ordinary type, to the combined and the autonomous ones by considering the role of nature in papering the shared logic between them. Also, provide a primitive insight towards the aspect of adapting the independency as a key factor for the futuristic prospect of architectural practices and facilitating the advanced intellectual sense of nature as its base for more fundamental investments. 2 Modeling of Posthuman Beauty Due to Ontological Factor Sample Based on what is known as the model of a thing, the model-making was for representa- tional order, rather than revealing the potentials, abilities or dysfunctionality, or disorder. The Model Ontology division of the theoretical foundation demonstrates the importance of positioning the true representation of the final object or the model by considering its own true-scale for projecting its real nature and manner in order to be efficiently invested [2]. As it is known, until the end of the middle ages in Europe, the practitioners were using the ancient Greek-based method of preparing a minimized model, that known at that periods as ‘paradeigmata’, in order to represent their final piece or prototype or object as ‘documentation of the contractual consideration’ procedure. Also, we can mention one of the earlier model-builders, those not only distinguished because of their conceptual manifestations for the architectural field, but for their craft or ‘techne’ of processing, Vitruvius, the most known through the history of architecture. Also, Alberti was using the modeling for explaining the ontology of the system through the simulation of the three-dimensional, strength of the proposed structure and related solidity form as mass objecting of the prototype of the project as possible as he and the technology enabled for. So, the mentioned examples were just representing the transitional stage from the graphical representation and individuality of analyzing of the next stage of the enabling the faster and easier understudying of the next stage of the properties and potentials, but still with wide range of individuality and propositional sort of scaling, 140 M. Abbood and G. Al Slik performing, resisting, detailing and the most important aspect of using the real and the same amount and covering area and combination of the determined materials [2]. The model in it prototyped phase, that just delivering some unrevealed prospects of the final object, system or project has an alluring nature to what is wanted and unrelated to the fact potentialities, that mostly overshadowing the recipient. In this matter, architects must under title the finalized models of their project as a ‘preliminary object’, and not a final one. For the real and efficient validation and analyze, as responsible way for true representation of the primary morphology of the project and also as Massimo Scolari states about that; they are “miniature isomorph of the building” in their scale and properties, not as final set of the gained potentialities [2]. But the phase of the included beauty of the order through faster achievement of all aspects of the final object is introduced via simulation technique which gives us the benefits as; 1. The easier capability of dealing with, because of smallness of the scale, 2. It provides variety of possibilities of fabricating and assembling parts, junctions and details, depending on the variety of material and methods. Researchers saw this as emergence of new forms of (posthuman beauty), when architecture moves the locus of its experimentation from the authority of the abstract drawing and expressions of truth in materials and assembly to the speculative reality of the model [2]. According to the explained sight of (ontological analyzation), the resulted prototype or object as model aimed to challenge the recipient mind in order to be able to see images and differentiate between a toy and real model. That is because of the difference and the patchiness of the used materials in construction (as real object) or on the surfaces (as drawing), those leading to the forming patterns according to their properties but not on the same quantity and even quality of designed for the final and on the ground object. Thus, the result of simulating object or prototype will not prove any objective performance, because of its absence on the ground as existence of reality, and the mimicry of the final order, and could only counted as alteration of individuality of some limited process towards the reality. Yet in this course, the (ontology) of the individual model or prototypes would demonstrate some new aspects, even if the prototype is designed to the proper human scale for revealing some techniques of operational orders or constructional methods as; Setting the partial and spatial relationship as alterative scaling of needed order, revealing new set of functional and spatial aesthetics, and enabling for some unusual sorts of physical editing and dealing with for both partial and comprehensive manner of manufacturing, fabricating and assembling and remaining under the term of representational aspect of the modeling that not necessarily giving true advantages. Here, the individual roles of both joints and parts of the model are taking place for revealing strange manner of beauty via the facilitating the ‘meta-seams’ method of material or junction, that leads to the ‘tectonically phase’ of the expressed beauty as affective aspect, and the unexpected functioning of connection via the use of ‘super- components’ of the beauty, that could charm the recipient by the unexpected state. The researchers see that mentioned means of achieving the state of the beauty, would be accessible by considering the harmony between the ‘true functioning’ and operating according to the exact scale of manufacturing or fabrication that designed for, in order From Digitized Systemization to New Era 141 to develop the optimum results and demonstrating ‘the difference between the fact and fiction’, which refers also to the essence of necessity and the shading by simulative projection and propositions. 3 Smart Processing of Additive Design and Manufacture The contemporary architectural disciplines of automation and individual programming have been characterized by the algorithmic manner of intellectual enrolment. This man- ner is considered as basic analytic procedures of training software and machine learning as well. According to the need of enhancing the range of accuracy in final actuation and its impact, the utilized material in machine learning order, which depends on software or preprogramming of material training, leads to advance accuracy and deeper insight of problem deduction, analysis, and solution-makings and finally actuate as much as prop- erly [3]. As explained in Fig. 1 the developed by Benjamin Jennett and Kenneth Cheung example of versatile and generative phase of digital processing and production at MIT and NASA, BILL-E, distribute robots those are able to build, inspect and edit discrete lattice structures. Compared to continuous fabrication methods such as 3D printing, this approach offers the possibility to adapt, maintain and repair structures in real-time [3]. Fig. 1. MIT Center for Bits and Atoms and NASA Ames Research Center, BILL-E robotic platform, 2017 [4]. Also, for controlling the final properties of the produced model, discreteness pro- vides the access to the ability to scale mass-deformation by the computational process as (what we saw in term of architectural mereology of processing and conducting of parts). The term discrete in computational design and processing reveals the ‘key-aspect’ of both autonomously automotive optimization of the quality of the architectural produc- tion. Here, the role of the part in the structure of the model passes through the historical mean of element towards the unite, then to the cell as contemporary 3D or even 4D printed matter, and eventually to the data, bits or sign as Emmanuelle Chiapone Piriou 142 M. Abbood and G. Al Slik explains the historical evaluation of facilitating the discreteness of the parts in architec- ture. Thus, the architectural state of 21st century’s aesthetics are no longer referring only to the optimization and sophistication of production, but paving the way to reducing the amount of humanistic efforts and costs and directing them more to the social benefits by automation and organization of materials and systems, as the raised practicing platform. 4 Emerging of Post-digital Potentialities with the Concept of Beauty Continuing the theoretical approach towards the achievement of optimization of the architectural order of the contemporary materials and system of the early 21st century through post digitalism, the term of beauty is also linked to the act of control of artificial intelligence mechanism of transporting and transferring big data, in order to enhance the actuation of the functional situation of the final architectural products [5]. Depend- ing on the fundamental aspect of ontology and discreteness of finalizing a system or a model for architectural and also functionally related to operating and effect, which are another aspects of basically rooted to the essence of reality for both mentioned aspects have to be projected as (scale and factor of the beauty) through the (potentiality of programming process). This means that the form and space needed with their aestheti- cal characterization might be uploaded as programmed order through the construction materials. According to the philosopher Max Bense in 1960, the methodology of ‘program- ming the beauty’ in order to function beyond the emergence of qualities of materials and technologies, achieved at that period a wide and open-ended prospect for evaluation and development, the difference between the substances of both designed and constructed terms of the beauty becomes obviously clear through the outcomes. Thus, in order to fol- low the update of the characterization of the mean, the state of beauty demonstrates raised constants, those represent the aesthetics of scaling the range of impacts and capabilities, and so the state of beauty which was affecting the individual set of the understanding. Due to that, it becomes the state of reality that affects the explorational phase of the recipient’s consciousness according to the physical and objective investment of prop- erties of the architectural system. By considering Bense’s sight to the contemporary state of the beauty and its relation to the technological integrations of functioning and facilitating the orders, as he expressed as: “the concept of beauty is losing in substance, but gaining in function”, and Aristotle’s description of the quality towards the quantity of the beauty and its impact by considering the discreetness in sorting each item for assessing the impact of the proportional relationship between them as a mechanism of production and operation, so the act of setting the beauty range depends on the ontologi- cal aspect in the physical process, which reveals the quality of the production, while the discreteness reveals the quantity of the order and the advantages of every single property of the entities [5], such as; Segments, Vortexes, Surface or mediums, and needed time of accomplishment. Those represent the pillars of raised technology of programming materials through the 4D printing technology, and basic element of programming and tangibly functioning and self-actuating the medium of producing (the new concept of the beauty). As well as the quantity of the order here, the quality phase contains discrete features through its own medium of the process [5], such as; The nature of performing From Digitized Systemization to New Era 143 and actuation potentials, the range of ‘effect on’ and ‘affect by’ via the impacting of functioning, permanent method of operating through the mechanism of the system, and state of patchiness towards the emerged conditions. So, the new approach of ‘assessing the beauty’ according to both factors of (qual- ity and quantity) have temporary and permanent features due to the mechanism and enhancement of each factor. The state of beauty here refers to the superior effect and the depth of effect according to the immediate impression that received by the human desire achievement, not only as concept or individuals, but as real novel and unexpectable oper- ating and actuation on the ground due to optimization of the process. According to [5], the beauty aspect demonstrates the state of (deep super-quality) of its order that could be described as the ability to operate in the numerous medium of parametric aspects, such as the special and geometrical aspects as the transferable actuator, due to the memorization and autonomously responding potentials, as a new aesthetics of contemporary version of engaging with the act and respond term of design. Its volatile nature would allow this aesthetic system to interact with programmatic, structural, ecological, or contextual frameworks and inform them. Also, the parametrical mediums and their related entities are playing the main role in directing the comprehensive mechanism of actuating the sys- tem by the manipulation, stimulation, and also the direct informing of the special unite or part of the system to characterize the aggregative manner of operation by combining every single discrete mechanism together. 4.1 The Beauty in Post-digitalism and Bio-Digital Based Sustainability As contemporary pre-planning characteristic order of qualifying the design and oper- ation approach of architectural practice, and facilitating such order on both individual and urban scales of, the so-called ‘blue-green planning’ reveals efficiency in dealing and processing the climate change problem and other special and functional aspects those based on the architectural role of the system. This prospect of confronting the contemporary challenges of the early 21st century via the firm framework of providing real and true answering or problem-solving of the order, based mainly on two incorpo- rated domains of (contemporary bio and digital senses), and actuating new aesthetics on this matter as well [6]. Through the blue-green planning the involved practicians of the architectural field which targeted the (re-greening state of architectural footprint) by sanitizing the ecosystem of the design and production order, to reach the endurance of modernity, inspired by bacteriological aspects and principles of controlling of sanitate actuation to produce a harmonical medium of live-beings, special use, and consumption. Here, the general concept of architectural fabrication, accumulation, formulation pro- duction, and even interaction and response paradigms are wider than the digital manner of individual actuation, ordering, and synthetical methodologies of physical facilitating of mass-material. The new approach demonstrates the ‘deep efficiency’, and ‘system values’ of emerging the biological scope of investing for accuracy and production along with the last achievements of the early 21st century’s period of post-digital state, in order to prepare a new vision of (self-automotive manner of formulation and formation of natural patterns) [6], those crucially guide special and function order on the micro-scale as well ae the humanistic-scale of architectural order through magnifying the range of actuation via the same concept and principles. So, those principles on their new scale 144 M. Abbood and G. Al Slik of functioning could not be considered as linear procedures, and also reveal another division of processing the composition of ‘bio-based feedback strategy’ of contempo- rary material science and the architectural domain of design and operation. According to that, the state of independence here would be characterized as another ‘non-human based strategy’ of production and critical controlling of both ecological and spatial systems and would encourage the role and importance of nature, as (essential aesthetic guide) to the process till getting the final enhanced function. Also, this approach considers the ‘known as west material and pollution’ as another variable of dealing and feeding the productional prosses and also turning them into new materials and nutrients as ‘main resources’ for continuing the order at others. So, this methodology of engaging with nature and the order has the potentiality of being facilitated on both scales of urban and individual or singular state of the architectural production, according to mentioned properties and frameworks of its conceptual division. From this perspective (ecoLogicStudio’s bio-digital architectures) in 2018 promoted a new urban aesthetic centered around a novel appreciation for the micro-scale of bacteria as well as other forms of non-human intelligence [6]. Here, the new vision of formation and actuation establishes a new and prospected state of considering bio-based aesthetics as the essential root of culture for living and production through architecture. Also, revealing new sort of impacts identification of people as ‘bio-citizenship’. Towards that, ecoLogicStudio in 2012 and Xeno Derma (as a research project) in corporation with Urban Morphogenesis Lab are studied experimental, spatial, and functional examples of this approach (Fig. 2). Fig. 2. Interior and Exterior views of Cyber-gardeners ecoLogicStudio, HORTUS XL Astaxan- thin.g, Centre Pompidou, Paris, 2019 [6]. 5 Conclusions The post-digital era of processing data and informative ruling of manifestations, reveals a new typology of dealing with the natural and artificial mechanisms of material activation. From Digitized Systemization to New Era 145 This leads to adopting a new manner of classification and programming the orders as a more efficient technological approach, which directs the contemporary architectural state of practice to an extraordinary level of procedural understandings and controlling the production of the beauty and its impacts. As the result, the procedure of preparing new aesthetics for the upcoming era of architectural practices, those based on mass-production and autonomy would remark new featural points toward guiding the suitability of built environment outcomes and paving the way for a higher reliance on material individual empowerment in forward. Also, on the systemization scale of the next stages of the sustainable approach of advanced material actuation, the unseen beauty of potent codes beyond nature is riching the wider range of controlling the state of intelligence building of any artificial structures as the closest norm of providing efficiency of material activation due to human desire for more sufficient interactions. In addition, this type of active norm are giving a deeper connection with the ontological aspect of material actuation for riching the optimum programmed for functionalization. References 1. C.E. Corpus: Cambridge Dictionary, 21 May 2019. https://dictionary.cambridge.org/dictio nary/english/potential 2. Wiscombe, T.: A specific theory of models: the posthuman beauty of weird scales, snowglobes and supercomponents. Arch. Des. AD 89(5), 80–89 (2019) 3. Retsin, G.: Special issue: discrete: reappraising the digital in architecture. Arch. Des. AD 89(2), 7–13 (2019) 4. Costa, A., Abdel-Rahman, A., Jenett, B., Gershenfeld, N., Kostitsyna, I., Cheung, K.: Algo- rithmic approaches to reconfigurable assembly systems. In: IEEE Aerospace Conference (2019) 5. Rutzinger, S., Schinegger, K.: Deep immediacy: programming beauty. Arch. Des. AD 89(5), 98–103 (2019) 6. Pasquero, C., Poletto, M.: Beauty as ecological intelligence: bio-digital aesthetics as a value system of post-anthropocene architecture. Arch. Des. AD 89(5), 58–65 (2019) Geological Exploration and Earthquake Engineering Analytical Investigation of the Influence of Seismicity on the Construction Cost of the R/C Load-Bearing Structure of a 15-Storey Building Georgios Papadimitriou and Theodoros Chrysanidis(B) International Hellenic University, P.O. Box 141, 26335 Sindos, Greece

Abstract. Greece is divided into three seismic hazard zones ZI, ZII, ZIII. In the present research work, the same building is modeled, analysed and dimensioned for the three seismic zones of Greece and then the construction cost of its load- bearing body is estimated. The building modeling was performed in SAP2000 software using linear finite elements. The analysis of the building was performed by dynamic spectral analysis methods using the design spectrum of EC8. A 15- storey building with a standard floor plan per floor was used. The purpose of this research paper is through comparative analytical estimation of construction costs to demonstrate whether the cost of construction of the bearing structure of a reinforced concrete building is affected by the area seismic hazard, if this influence is significant and to what extent. Useful conclusions are drawn regarding the influence of seismicity on the construction cost of the load-bearing structure of reinforced concrete buildings. Keywords: Construction cost · Seismic hazard · Reinforced concrete · Building 1 Introduction It is world-wide accepted that a multi-storey reinforced concrete (R/C) building should be adequate against all types of loads [1, 2]. Special provisions are considered for seismic actions in the relevant codes and regulations. These provisions have to do, among other things, with the seismic hazard zone that each structure should be designed in. Usually, the relevant regulations like, e.g., EC8 [3] or the Greek Seismic Code 2000 [4], etc. include specific maps showing each region in which seismic zone belongs to. According to these maps, the consulting engineer chooses the seismic zone that the building belongs to and designs it accordingly to the earthquake action provided for the seismic zone in question. Having said that, it is noteworthy to be said that the engineer can choose a higher seismic zone if he wishes to for reasons of extra safety. Of course, such an action by the engineer has an effect to the construction cost, too apart from the safety [5–12]. In that case, other phenomena have to be taken into account to, e.g., the lateral buckling of R/C walls, etc. [13, 14]. The present study investigates the influence to the construction cost of the load-bearing structure of a 15-storey building when moving from one seismic © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 149–162, 2023. https://doi.org/10.1007/978-981-19-3983-9_14 150 G. Papadimitriou and T. Chrysanidis zone to a higher one. Useful conclusions arise from the present research which could be proven helpful to practice engineers world-wide. The codes used are Eurocode 2 (EC2) [26] for the design of concrete and Eurocode 8 (EC8) [3] for the earthquake loads. 2 Structure Description This study concerns a ten-storey building without a basement. It has a rectangular plan 25 × 25 m with a total floor plan area of E = 625 m2 . There are two walls at the perimeter of the floor plan and in the centre, there is the staircase and the elevator surrounded by eight walls, forming two symmetrical cores. The height of each floor is 3.0 m except for the ground floor which is 4.5 m and the dimensions of the columns change in height and are reduced per floor by 5 cm. Also, the thickness of the walls changes in height and reduces per five floors by 12.5 cm. Table 1 displays the geometrical characteristics for all structural elements and for all floors of the building, while Fig. 1 displays the floor plan. The concrete used is quality C30/37 and according to EC2 its modulus of elasticity is equal to Ecm = 33 GPa while the steel quality is B500C. SAP2000 software is used for the building modelling. Figure 2 displays the model of the building. Table 1. Geometrical characteristics of structural elements. Floor Height Beam dimensions (cm) Wall thickness Columns (m) Perimetric Internal (cm) (cm) 1st 4.5 25 × 70 25 × 60 50 100 × 100 2nd 3.0 25 × 70 25 × 60 50 100 × 100 3rd 3.0 25 × 70 25 × 60 50 95 × 95 4th 3.0 25 × 70 25 × 60 50 90 × 90 5th 3.0 25 × 70 25 × 60 50 85 × 85 6th 3.0 25 × 70 25 × 60 37.5 80 × 80 7th 3.0 25 × 70 25 × 60 37.5 75 × 75 8th 3.0 25 × 70 25 × 60 37.5 70 × 70 9th 3.0 25 × 70 25 × 60 37.5 65 × 65 10th 3.0 25 × 70 25 × 60 37.5 60 × 60 11th 3.0 25 × 70 25 × 60 25 55 × 55 12th 3.0 25 × 70 25 × 60 25 50 × 50 13th 3.0 25 × 70 25 × 60 25 45 × 45 14th 3.0 25 × 70 25 × 60 25 40 × 40 15th 3.0 25 × 70 25 × 60 25 35 × 35 Analytical Investigation of the Influence of Seismicity 151 Fig. 1. Building’s floor plan. Fig. 2. Views of the 3D building model: (a) Linear finite elements, (b) Sections. 152 G. Papadimitriou and T. Chrysanidis 3 Analysis of Results 3.1 Concrete Measurements Table 2 displays the measurement results for concrete material. Figure 3 displays the concrete percentage per building floor, while Fig. 4 shows the concrete participation per type of structural element. From these diagrams, it is concluded that the first and second floors have the highest percentage of concrete. On the contrary, the percentage of concrete from the third to the eighth floor decreases by 1% per three floors, then on the ninth and tenth floors, it remains constant, then decreases by 1% and from the eleventh floor to the roof remains constant. It is observed that the slabs occupy the largest percentage of concrete equal to 42%, in the second place come the columns with 26%, then the walls with 22% and finally the beams with 10%. Table 2. Concrete measurements for each structural element per floor. Concrete measurements per floor Storey Slabs Beams Columns Walls TOTAL (m3 ) (m3 ) (m3 ) (m3 ) (m3 ) 1 90 21 104 63 278 2 90 21 104 63 278 3 90 21 94 63 268 4 90 21 85 63 259 5 90 21 75 63 249 6 90 21 67 47 225 7 90 21 59 47 217 8 90 21 51 47 209 9 90 21 44 47 202 10 90 21 38 47 196 11 90 21 32 32 174 12 90 21 26 32 169 13 90 21 21 32 164 14 90 21 17 32 159 15 90 21 13 32 155 TOTAL 1350 314 830 710 3203 Analytical Investigation of the Influence of Seismicity 153 Concrete percentage per floor 1 2 3 4 5% 5% 9% 5 5% 9% 5% 6 5% 8% 7 6% 8% 8 6% 9 8% 7% 10 7% 7% 11 12 13 14 15 Fig. 3. Concrete percentage per floor. Structural element parƟcipaƟon in concrete 22% SLABS 42% BEAMS 26% COLUMNS 10% WALLS Fig. 4. Structural element participation in concrete percentage. 154 G. Papadimitriou and T. Chrysanidis 3.2 Steel Measurements Table 3 displays the measurement results for reinforcement steel material, while Fig. 5 shows the total steel weight of structural elements for all three seismic zones measured for the ground floor using a column diagram. Table 3. Measurements of reinforcement steel for ground floor. Steel measurements Zone Slabs Beams Columns Walls Total steel (kg) (kg) (kg) (kg) (kg) I 9066 5563 10837 4059 29525 II 9066 5563 10837 4323 29790 III 9066 5571 10837 5256 30730 Total steel weight of structural elements for ground floor (kgr) 31000 30500 30000 29500 29000 28500 Ι ΙΙ ΙΙΙ Fig. 5. Total steel weight of structural elements on the ground floor for all three seismic zones. Analytical Investigation of the Influence of Seismicity 155 3.3 Percentage Increase in Steel Weight Between the Zones Percentage increase in total steel weight between zones 11% 50% 39% Ζ1-Ζ2 Ζ2-Ζ3 Ζ1-Ζ3 Fig. 6. Percentage increase in total steel weight between zones for ground floor Percentage increase in beam steel weight between zones 6% 50% 44% Ζ1-Ζ2 Ζ2-Ζ3 Ζ1-Ζ3 Fig. 7. Percentage increase in beam steel weight between zones for ground floor. 156 G. Papadimitriou and T. Chrysanidis Percentage increase in wall steel weight between zones 13% 49% 38% Ζ1-Ζ2 Ζ2-Ζ3 Ζ1-Ζ3 Fig. 8. Percentage increase in wall steel weight between zones for ground floor. From the above diagrams (Figs. 6, 7, 8), the following are concluded: 1. The transition of the load-bearing system from zone I with seismic acceleration αg = 0.16 g to zone II with seismic acceleration αg = 0.24 g results in: • 11% increase in the total steel weight of all structural elements • 6% increase in the steel weight of the beams • 13% increase in the steel weight of the walls 2. The transition of the load-bearing system from zone II with seismic acceleration αg = 0.24 g to zone III with seismic acceleration αg = 0.36 g results in: • 39% increase in the total steel weight of all structural elements • 44% increase in the steel weight of the beams • 38% increase in the steel weight of the walls 3. The transition of the load-bearing system from zone I with seismic acceleration αg = 0.16 g to zone III with seismic acceleration αg = 0.36 g results in: • 50% increase in the total steel weight of all structural elements • 50% increase in the steel weight of the beams • 49% increase in the steel weight of the walls Analytical Investigation of the Influence of Seismicity 157 4. The increase in seismic acceleration from: • Zone I to Zone II is of the order of ((0.24 − 0.16)/0.16) * 100 = 50% • Zone II to Zone III is of the order of ((0.36 − 0.24)/0.24) * 100 = 50% • Zone I to Zone III is of the order of ((0.36 − 0.16)/0.16) * 100 = 125% 5. The steel weight for the columns as well as for the slabs remains unchanged during the transition of the load-bearing system from zone to zone. The other structural elements, the beams and the walls, their steel weight increases from zone I to zone II and to zone III, as a result of which the total weight of the ground floor steel increases. 3.4 Steel Percentage Per Structural Element for Each Zone Steel percentage of structural elements (Zone I) 14% 31% Slabs Beams Columns Walls 36% 19% Fig. 9. Steel percentage of structural elements for ground floor of Zone I. 158 G. Papadimitriou and T. Chrysanidis Steel percentage of structural elements (Zone II) 15% 30% Slabs Beams Columns Walls 36% 19% Fig. 10. Steel percentage of structural elements for ground floor of Zone II. Steel percentage of structural elements (Zone III) 17% 30% Slabs Beams Columns Walls 35% 18% Fig. 11. Steel percentage of structural elements for ground floor of Zone III. Analytical Investigation of the Influence of Seismicity 159 From the above diagrams (Figs. 9, 10, 11), the following are concluded: 1. In all seismic zones, the columns occupy the largest percentage in steel: • Zone I = 36% • Zone II = 36% • Zone III = 35% 2. In all seismic zones, the slabs occupy the second largest percentage in steel: • Zone I = 31% • Zone II = 30% • Zone III = 30% 3. In all seismic zones, the beams occupy the second smallest percentage in steel: • Zone I = 19% • Zone II = 19% • Zone III = 18% 4. In all seismic zones, the walls occupy the smallest percentage in steel: • Zone I = 14% • Zone II = 15% • Zone III = 17% 5. The beams retain their transverse reinforcement in the critical area during their transition to the three seismic zones while in the non-critical area from zone I to Zone III the longitudinal reinforcement shows some increase. 6. The columns in each seismic zone retain the same reinforcement. 7. The walls between the seismic zones show increases in shear reinforcement. 4 Conclusions The building studied is a 15-storey building of square conventional floor plan without a basement. The building was analysed using the spectral dynamic analysis according to EC8 [3]. After the analysis has finished, the dimensioning of the building took place according, again, to the provisions of EC2 [26] and EC8 [3]. The last step of analysis involved the measurement of the quantities demanded for the construction of the load- bearing structure of the building, as far as the concrete and steel reinforcement materials are concerned. During its design in the three seismic zones, the following conclusions have risen: 160 G. Papadimitriou and T. Chrysanidis 1. The walls show a large percentage increase in steel weight with the increase of seismic acceleration. • From Zone I (0.16 g) to Zone II (0.24 g): Increase 13% • From Zone II (0.24 g) to Zone III (0.36 g): Increase 38% • From Zone I (0.16 g) to Zone III (0.36 g): Increase 49% 2. In the beams the steel rises almost equally to the walls: • From Zone I (0.16 g) to Zone II (0.24 g): Increase 6% • From Zone II (0.24 g) to Zone III (0.36 g): Increase 44% • From Zone I (0.16 g) to Zone III (0.36 g): Increase 50% 3. In the columns, on the other hand, the steel percentage remains stable for all three seismic zones. This was expected since the structural walls have a dominant position in the load-bearing system. 4. The increase of seismic acceleration: • From Zone I to Zone II, it is of the order of 50%, while the percentage increase in the total demand of steel is 11%, i.e., much smaller. • From Zone II to Zone III, it is of the order of 50%, while the percentage increase in the total demand in steel is 39%, i.e., much smaller. • From Zone I to Zone III, it is of the order of 125%, while the percentage increase in the total demand in steel is 50%, i.e., much smaller. 5. The conclusion is that for a 15-storey building with strong walls and a strong core of walls, the percentage increase of materials required is proportional to the seismic risk, but not to the extent someone would expect, thus the cost increase can be considered as not too much, especially compared to the corresponding increase of the seismic acceleration. 6. The challenges still existing is the analysis, dimensioning and material measure- ment for more buildings with different parameters. For example, different number of storeys can be used and different floor plans should be used, too. Furthermore, the influence of the foundation and the soil could be taken into account to, instead of using only rigid supports for all vertical elements at the base of the ground floor. These challenges could be confronted in a future research. References 1. Penelis, G., Stylianidis, K., Kappos, A., Ignatakis, C.: Reinforced Concrete Structures. A.U. Th. Press, Thessaloniki (1995) 2. Penelis, G.G., Kappos, A.J.: Earthquake-Resistant Concrete Structures. E & F N SPON (Chapman & Hall), London (1996) 3. European Committee for Standardization: EN 1998–1:2004, Eurocode 8: design of structures for earthquake resistance - Part 1: General rules, seismic actions and rules for buildings (2004) Analytical Investigation of the Influence of Seismicity 161 4. Ministry of Environment Planning and Public Works: Greek earthquake resistant design code (2003) 5. Chrysanidis, T., Panoskaltsis, V., Tegos, I.: Parametrical cost analysis of an ultra high-rise building: Detailed design. Int. J. Appl. Eng. Res. 11, 9644–9650 (2016) 6. Chrysanidis, T., Panoskaltsis, V., Tegos, I.: Preliminary design and analysis of cost parameters of a high-rise building: braced shear wall core system. Int. J. Civil Eng. Technol. 7, 137–152 (2016) 7. Chrysanidis, T., Tegos, I.: Cost comparison and parametrical investigation of the R/C shear wall core of a tall building. Int. J. Eng. Res. Technol. 5, 592–595 (2016) 8. Papageorgiou, G., Papadimitriou, E., Alamanis, N., Xafoulis, N., Chouliaras, I., Lazogiannis, K.: Construction cost comparative analysis of highways in Greece. PRIME Int. J. Pract. Res. Innovative Manag. Entrepreneurship 12, 68–85 (2019) 9. Tsiknas, A., Athanasopoulou, A., Papageorgiou, G.: Evaluation of flexible pavement con- struction cost according to the design method. In: Proceedings of the Institution of Civil Engineers (ICE) – Transport, 173, 3–12 (2020) 10. Papageorgiou, G., Alamanis, N., Chouliaras, I., Kapsali, P.: Decision making for designing infrastructure projects: the case of the city of Larissa. Greece J. Eng. Archit. 7, 115–125 (2019). https://doi.org/10.15640/jea.v7n1a13 11. Srivastava, M., Kumar, V.: The methods of using low cost housing techniques in India. J. Build. Eng. 15, 102–108 (2018). https://doi.org/10.1016/j.jobe.2017.11.001 12. Jasvi, A., Bera, D.: Sustainable use of low cost building materials in the rural India. Int. J. Res. Eng. Technol. 4, 534–547 (2015). https://doi.org/10.15623/ijret.2015.0425077 13. Chrysanidis, T.: The effect of longitudinal reinforcement ratio on the lateral buckling behavior of R/C walls modelled using prism elements. J. Build. Eng. 42, 102456 (2021). https://doi. org/10.1016/j.jobe.2021.102456 14. Chrysanidis, T.A.: Evaluation of out-of-plane response of R/C structural wall boundary edges detailed with maximum code-prescribed longitudinal reinforcement ratio. Int. J. Concr. Struct. Mater. 14(1), 1–21 (2020). https://doi.org/10.1186/s40069-019-0378-4 15. Chrysanidis, T.: Influence of elongation degree on transverse buckling of confined boundary regions of R/C seismic walls. Constr. Build. Mater. 211, 703–720 (2019). https://doi.org/10. 1016/J.CONBUILDMAT.2019.03.271 16. Chrysanidis, T., Tegos, I.: Size of seismic tensile strain and its influence on the displacements due to transverse buckling of ultra-highly reinforced structural walls. ARPN J. Eng. Appl. Sci. 11, 13884–13890 (2016) 17. Haro, A., Kowalsky, M., Chai, Y., Lucier, G.: Boundary elements of special reinforced concrete walls tested under different loading paths. Earthq. Spectra 34, 1267–1288 (2018). https://doi. org/10.1193/081617EQS160M 18. Welt, T., Massone, L., LaFave, J., Lehman, D., McCabe, S., Polanco, P.: Confinement behavior of rectangular RC prisms simulating wall boundary elements. J. Struct. Eng. 143, 1–12 (2017). https://doi.org/10.1061/(asce)st.1943-541x.0001682 19. Herrick, K., Kowalsky, M.: Out-of-plane buckling of ductile reinforced structural walls due to in-plane loads. J. Struct. Eng. 143, 1–15 (2016). https://doi.org/10.1061/(asce)st.1943-541x. 0001660 20. Chai, Y., Kunnath, S.: Minimum thickness for ductile RC structural walls. Eng. Struct. 27, 1052–1063 (2005). https://doi.org/10.1016/j.engstruct.2005.02.004 21. Chrysanidis, T., Tegos, I.: Axial and transverse strengthening of R/C circular columns: con- ventional and new type of steel and hybrid jackets using high-strength mortar. J. Build. Eng. 30, 101236 (2020). https://doi.org/10.1016/j.jobe.2020.101236 22. Terzioglu, T., Orakcal, K., Massone, L.M.: Cyclic lateral load behavior of squat reinforced concrete walls. Eng. Struct. 160, 147 (2018). https://doi.org/10.1016/j.engstruct.2018.01.024 162 G. Papadimitriou and T. Chrysanidis 23. Christidis, K., Trezos, K.: Experimental investigation of existing non-conforming RC shear walls. Eng. Struct. 140, 26–38 (2017). https://doi.org/10.1016/j.engstruct.2017.02.063 24. Ho, N.M., Doh, J.H., Fragomeni, S.: Instability analysis of reinforced concrete walls with various support conditions. Struct. Des. Tall Spec. Build. 26, e1353 (2017). https://doi.org/ 10.1002/tal.1353 25. Lim, J.-J., Park, H.-G., Eom, T.-S.: Cyclic load tests of reinforced concrete columns with high-strength bundled bars. ACI Struct. J. 114, 197–207 (2017). https://doi.org/10.14359/516 89254 26. European Committee for Standardization: EN 1992–1–1:2004, Eurocode 2: design of concrete structures - Part 1–1: General rules and rules for buildings (2004) The Influence of Buried Depth on the Stress Mode of Independent Foundation in Rock Foundation Yunpeng Dong1,2(B) and Yin Ke1,2 1 College of Civil Engineering, Chongqing University, Chongqing 400045, China

2 Key Laboratory of New Technology for Construction of Cities in Mountain Area of Ministry of Education, Chongqing University, Chongqing, China Abstract. Reinforced concrete independent foundation has the advantages of simple structure, low economic cost and convenient design. In order to study the influence of buried depth on the stress mode of independent foundation in rock foundation, the stress and deformation characteristics of independent foundation in rock foundation under buried depth are analyzed through theoretical discussion and numerical simulation, and the differences of reaction force and internal stress field of foundation under different constraints are compared. The results show that the independent foundation is actually a stress model between the deep beam and the compression member under the constraints of the bottom and side walls; In addition, when the independent foundation is under different constraints, the mechanical properties are different, so the influence of various constraints on the independent foundation should be considered in the design. Keywords: Rock foundation · Independent foundation · Buried depth · Force mode · Design calculation 1 Introduction Rock foundation usually has the characteristics of shallow buried depth, high stiffness and high strength. As a widely used form of extended foundation, reinforced concrete independent foundation has the advantages of simple construction, cost economy and easy design. Therefore, given the engineering value of adopting independent foundations in rock foundations, scholars have studied the force models and calculation methods of independent foundations on rock foundations in addition to treating independent foundations as deep bent members or shallow beams. Zhang Bin [1] believes that the shear force of the independent foundation on the rock foundation is large and the bending moment is small, which is similar to the rectangu- lar short beam and inconsistent with the bending shear of the simply supported beam without belly bars. The same type of calculation model is considered to be similar to the rectangular short beam shear, and the bearing capacity design method is proposed to give full play to the shear strength of concrete [2–4]. Lai Qingwen [5] considered the © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 163–171, 2023. https://doi.org/10.1007/978-981-19-3983-9_15 164 Y. Dong and Y. Ke independent foundation as a deep beam model subjected to continuous normal loads for analysis while ignoring the bonding force and friction between the rock foundation and the foundation. Li Cuncun [6] proposed that the load on the independent foundation on the rock foundation is borne by the core part and circumferential concrete, and its design model is an expansion model formed by the core part constrained by the bottom and circumferential under high pressure. Li Xinchun [7] and Zhang Youran [8] proposed the force models of flexural extension-expansion and deep bending, considering the joint effect of rock foundation and independent foundation under column and the width-height ratio parameters of foundation. According to the field test results and numerical simula- tion, Zhang Yuanzhen [9] combined the calculation models of deep beam, shallow beam and short beam, and proposed an empirical calculation method with different aspect ratios. Through experiments and theoretical analysis, Simoes et al. [10–12] proved that pure bending failure would occur when there were few bending bars at the bottom of an independent foundation, and in most cases, bending-shear beam model failure would occur. Jianfei g. et al. [13] regarded the independent foundation as an overground beam and carried out the reaction test. All the above studies regard the independent foundation as the model of beam com- pression shear action on rock foundation. However, in practical engineering, the inde- pendent foundation is buried at a certain depth, and the influence of the buried rock on the lateral wall and bottom of the foundation is much greater than that of the soil foun- dation. In addition, there are differences in foundation settlement deformation caused by the two kinds of foundations [14, 15]. Therefore, it is necessary to study the influence of independent foundation depth in rock foundation. In this paper, the stress and deforma- tion characteristics of independent foundation under the buried depth of rock foundation are analyzed, and then the differences of stress field of independent foundation under different constraints are compared by numerical simulation. This provides a reasonable basis for future design. 2 The Influence of the Buried Depth of the Foundation on the Force and Deformation of the Independent Foundation in the Rock Foundation When an object is compressed, it will produce co-directional compression deformation and lateral expansion deformation. In actual engineering, independent foundations gen- erally have different embedment depths. When the foundation bears the superstructure load transmitted by the concrete column, due to the low compressibility of the rock foundation, the lateral deformation of the foundation will be strongly restrained by the side foundation rock. The independent foundation on the rock foundation generally has a larger height and a smaller bottom area. When the foundation reaches a certain height, the upper load can be more evenly transmitted to the foundation. Therefore, it can be considered that the independent foundation of rock foundation with a certain buried depth is not only subjected to the pressure transmitted by the upper column, but also subjected to relatively uniform base reaction, foundation bottom friction, foundation side normal reaction and lateral tangential friction. As shown in Fig. 1. The Influence of Buried Depth on the Stress Mode of Independent Foundation 165 F —Column transfer pressure p n 1 —Base reaction f1 —Bottom friction p n 2 —Side normal reaction f 2 —Lateral tangential friction Fig. 1. Stress diagram of single-column footing within rock foundation Aiming at the basic force state, this paper analyzes its calculation model. When sim- ply supported beams with a span-to-height ratio of less than 2 or multi-span continuous beams with a span-to-height ratio of less than 2.5 no longer satisfy the assumption of flat section, they should be regarded as deep beam members. When the span-to-height ratio is smaller, especially when the bottom area of the foundation is equivalent to the top loading area, the foundation is like a column under concentric compression. Under heavy load, the independent foundation of complete rock foundation has the characteristics of relatively small area and large height, and its aspect ratio is usually less than 2 or even smaller. Therefore, the independent foundation can be regarded as a model in which the top is under central compression, the bottom and side walls are constrained, and are placed between the deep beam and the compression member. It is often inaccurate to think of them as simple deep beams or compression members. 3 Independent Foundation Numerical Simulation 3.1 Introduction to Model Parameters and Working Conditions The numerical model is mainly composed of concrete columns, independent foundations and rock foundations. The columns and independent foundation are both made of C30 concrete. The column size is 500 * 500 * 250 mm3 and the base size is 2000 * 2000 * 1700 mm3 , representing an aspect ratio of 0.45 independent base. According to the engineering geological survey report, physical and mechanical parameters of numerical simulation are shown in Table 1. Mohr-coulomb criterion is adopted for rock foundation, and concrete plastic damage model is adopted for independent foundation. The simulated conditions were divided into the following four types: 1. Condition 1: The independent foundation is buried in the rock foundation, and the normal and tangential forces of the side wall and the foundation are considered. 166 Y. Dong and Y. Ke Table 1. Material parameters. Material Elastic modulus Density Poisson’s ratio Friction angle Cohesive force GPa kg/m3 ° MPa Foundation 27.75 2400 0.27 \ \ concrete Rock 40 2540 0.27 31 0.6 2. Condition 2: The independent foundation is buried in the rock foundation, and the action of the normal force of the side wall, the normal reaction force of the base and the tangential friction force of the side wall are taken into account, but the tangential friction force of the side wall is not taken into account. 3. Condition 3: The independent foundation is placed on the rock foundation without burial depth, and the effect of normal reaction force and tangential friction of the foundation is not considered. 4. Condition 4: The independent foundation is placed on the rock foundation without burial depth, and the side wall binding force is not considered, but only the normal reaction force of the foundation is considered. 3.2 Results and Analysis of Independent Basic Properties Under Different Conditions 3.2.1 Base Reaction In order to magnify the influence of various constraints on the foundation, the force far greater than the test force, about 20 MPa, will be applied to the column in the simulation. The distribution diagram of the reaction force of the independent foundation under different constraints is shown as follows (see Fig. 2): Fig. 2. Base reaction curve The Influence of Buried Depth on the Stress Mode of Independent Foundation 167 By observing the shape of the above curves, it is found that the basic reaction curve is relatively flat and the stress concentration at the edges is large. In addition, lateral friction and bottom friction will increase the stress concentration of foundation edge. The lateral wall normal constraint makes the middle and both sides of the foundation reaction distribution curve smaller, while the stress at the edge of the foundation becomes more concentrated and the distribution shape of the whole curve becomes more obvious. 3.3 The First Principal Stress Distribution Considering whether the foundation is subjected to shear failure or flexural failure, the bending moment and shear value of the central axis of the foundation are both the largest.In addition, the shape of the foundation at the four corners has mutation, which is easy to cause stress concentration. Therefore, the central axis of the foundation and the edge of the foundation are extremely dangerous parts all around. In this paper, the distribution of the first principal stress inside the foundation is analyzed from four aspects: the central axis of the foundation, the angular Angle of the foundation, the axis of the foundation bottom and the diagonal of the foundation bottom.The specific location is shown in the Fig. 3. Fig. 3. Schematic diagram of position 3.3.1 The First Principal Stress Distribution Curve Along the Central Axis of the Foundation The Fig. 4 shows that the lateral friction has little effect on the distribution of the first principal stress along the central axis, while the bottom friction has a certain decrease on the principal stress near the base, and the further away from the bottom, its effect decreases. At the same time, considering the normal binding force of the side wall, except for the first principal stress near the top of the foundation, the principal stress at other positions is reduced to a certain extent. In general, the tension zone on the central axis is larger than the compression zone, and the maximum stress value in the tension zone is much smaller than the compression zone. 168 Y. Dong and Y. Ke Fig. 4. The first principal stress distribution curve along the central axis of the foundation 3.3.2 The First Principal Stress Curve Along the Height at the Base Corner The Fig. 5 shows that the friction between the side wall and the bottom has little effect on the principal stress at the corners of the independent foundation, but the normal force of the side wall does the opposite. When the side wall binding force is considered, the pressure near the corner of the independent foundation will be greater. Fig. 5. The first principal stress curve along the height at the base corner 3.3.3 The First Principal Stress Curve Along the Bottom Axis The lateral friction of independent foundation has little effect on the first principal stress of foundation bottom axis. The friction of the bottom surface of an independent foundation will reduce the first principal stress uniformly along the whole axis. When the normal binding force of the side wall is considered, the principal stress of bottom axis decreases uniformly and to the greatest extent. Observe the principal stress curve of the whole bottom axis, its distribution is relatively uniform, all are tensile stress (see Fig. 6). The Influence of Buried Depth on the Stress Mode of Independent Foundation 169 Fig. 6. The first principal stress curve along the bottom axis 3.3.4 The First Principal Stress Curve Diagonally Along the Underside On the diagonal, the effect of the bottom friction on the first principal stress decreases as the edge of the independent foundation approaches. At the same time, it can also be seen that the compressive stress of condition 3 and condition 4 is larger at the corner of inde- pendent foundation. In addition, each constraint has the same effect on the independent foundation as the bottom axis (see Fig. 7). Fig. 7. The first principal stress curve diagonally along the underside 3.3.5 Maximum Principal Stress Position of Independent Foundation The numerical results also show that the maximum tensile stress of the independent foundation is located at the boundary between the foundation and the column regardless of the constraint. Therefore, if the excessive force at the junction between the foundation and the column can be solved, further reducing the height of the foundation can be achieved. 170 Y. Dong and Y. Ke 3.4 Shear Stress Distribution at the Junction of Column and Independent Foundation The existing design is mostly based on the shear capacity of the inclined section at the junction of the foundation and the column, and according to the mechanical knowledge, the shear force here is the largest. Therefore, in order to study the difference of shear stress distribution in this part under different conditions, the shear stress diagram is shown as follows (see Fig. 8): Fig. 8. Shear stress distribution curve of foundation The figure above shows that the lateral wall constraint has a great influence on the shear stress distribution at the junction between foundation and column. Therefore, if the existing design is carried out without considering the sidewall constraints, the foundation may be unsafe and the economic benefits will be low. 4 Conclusion This paper summarizes the existing independent foundation design model, and points out the shortcomings of the model according to the deformation and stress characteristics of the independent foundation under column on rock foundation. Meanwhile, in order to compare with the original model, combined with numerical simulation, the founda- tion reaction and internal stress field under different constraints are studied. The main conclusions are as follows: (1) The independent foundation on the rock foundation cannot be regarded as a pure beam or compression member, while ignoring the sidewall and bottom constraints of the rock. Especially when there are sidewall normal constraints, the first prin- cipal stress inside the foundation is reduced, and the first principal stress near the four corners of the foundation changes from a smaller tensile stress to a larger compressive stress. The Influence of Buried Depth on the Stress Mode of Independent Foundation 171 (2) Whether sidewall constraint is considered has great influence on the internal shear stress distribution of independent foundation. Moreover, as concrete is a brittle material with very low tensile strength, it is not reasonable to design an inde- pendent foundation on rock foundation only according to the shear stress at the connection between foundation and column. It is suggested that the design should be carried out according to the stress characteristics of independent foundation and the corresponding first principal stress under actual conditions (3) In practical engineering, the bearing capacity of independent foundation in rock foundation is controlled by local pressure at the connection between column and foundation, so it can be considered to carry out certain reinforcement at the top of foundation. References 1. Bin, Z., Changan, X., Chengwei, X.: Discussion on shearing of individual footing of column based on rock foundation. J. Guizhou Univ. Technol. Nat. Sci. Ed. 31(6), 52–55 (2002) 2. Fangshi, D.: Research on shear design of independent foundation under column of rock foundation. Guangxi University (2002) 3. Chengwei, X.: Discussion on independent foundation shear resistance design. Low Carbon World 23, 339–340 (2014) 4. Wen, H.: Study on shear behavior of independent foundation in rock foundation. Heilongjiang Sci. Technol. Inf. 5(18), 264–265 (2011) 5. Qingwen, L., Hongsheng, Z., Honglinl, S.: Discussions on calculation methods for shear bearing capacity of footings on rock foundation. Ind. Constr. 32(8), 32–34 (2002) 6. Cuncun, L.: Failure model of independent foundation upper columns and lower columns of rock foundations and its related research. Chongqing University (2016) 7. Xinchun, L., Ke, Y., Jihui, H.: Bending-expansion tensile failure of single-column footing on rock foundation. Zhongguo keji lunwenm, 12(13), 1520–1525, 1542 (2017) 8. Youran, Z.: Study on the force characteristics and destruction model of under-column independent foundation on rock foundation. Chongqing University (2019) 9. Yuanzhen, Z., Ke, Y., Jihui, H.: Study on shearing resistance design of single column spread footing on rock foundation. Chin. J. Undergr. Space Eng. 15(05), 1511–1518 (2019) 10. Simões, J.T., Faria, D.M., Ruiz, M.F.: Strength of reinforced concrete footings without transverse reinforcement according to limit analysis. Eng. Struct. 112, 146–161 (2016) 11. Siburg, C., Josef, H.: Experimental investigations on punching behavior of reinforced concrete footings with realistic dimensions. Struct. Concr. 15(3), 331–339 (2015) 12. Hegger, J., Ricker, M., Ulke, B.: Investigations on the punching behaviour of reinforced concrete footings. Eng. Struct. 29(9), 2233–2241 (2007) 13. Jianfei, G., Hao, L., Guangyu, H., Huili, X., Jingwen, K.: Experimental study on failure behavior of plain concrete independent spread foundation under column. Build. Sci. 36(S1), 25–34 (2020) 14. Aurelio, M.: Punching shear strength of reinforced concrete slabs without transverse reinforcement. ACI Struct. J. 105(4), 440–450 (2008) 15. Einpaul, J., Bujnak, J., Ruiz, M.F.: Study on influence of column size and slab slenderness on punching strength. ACI Struct. J. 113(1), 135–145 (2016) Study on the Strength Characteristics of Metamorphic Slate Based on Point Load Tests Wei Xingcan1 , Qi Shaoli2 , Yang Baowen3 , Wang Wenpo4 , and Zhong Zhibin3(B) 1 Chengdu Engineering Corporation Limited, PowerChina, Chengdu 610072, China 2 Sinohydro Bureau 7 Corporation Limited, Chengdu 610231, China 3 College of Environment and Civil Engineering, Chengdu University of Technology, Chengdu 610059, Sichuan, China

4 School of Civil Engineering, Southwest Jiaotong University, Chengdu 610031, Sichuan , China Abstract. The mechanical properties of layered slate in Western Sichuan Plateau is significant anisotropic, which leads to the unstable surrounding rock of diver- sion tunnels. Since the integrity of thin slate is poor, it is impossible to obtain the strength parameters of rock through traditional uniaxial compression test. In this paper, based on the diversion tunnel project of Guzeng hydropower station in Western Sichuan, thin-layer carbonaceous slate and sandy slate of the lower Ordovician Wachang formation (O1 w) exposed by excavation are taken as the objects. Firstly, the main mineral composition and microstructure characteristics of the rock are analysed by XRD and thin section projection microscopy, and then point loading tests were performed on irregular rock blocks. Finally, 10 typ- ical empirical formulas were selected to investigate the applicability of different empirical formulas to predict the uniaxial compressive strength of slate, and the strength characteristics of two slates were analysed. The results show that: (1) the slate is a typical thin-layer structure, which mainly composed of quartz and a small amount of albite, muscovite, pyrite, etc.; (2) the point load strength of the two kinds of slate is significantly anisotropic. The overall strength of sandy slate is larger than that of carbonaceous, while the compressive strength of sandy slate along the bedding is smaller, and the discreteness is also greater; (3) for the selected 10 kinds of empirical formula, exponential function and logarithmic function predict the worst, while linear function is depended on the selected pro- portion coefficient. The linear function recommended by ISRM with lower limit proportion coefficient shows better prediction, while the power function recom- mended by Standard for engineering classification of rock mass is conservative. The results can provide a reference for stability evaluation of a large number of deep buried underground projects in Western Sichuan Plateau. Keywords: Rock mechanics · Slate · Point load tests · Strength characteristics · Anisotropic 1 Introduction The eastern margin of the Qinghai-Tibet Plateau is one of the most complex geological structures in China and even in the world. Under the long-term internal and external © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 172–189, 2023. https://doi.org/10.1007/978-981-19-3983-9_16 Study on the Strength Characteristics of Metamorphic Slate Based on Point Load Tests 173 dynamic action, the geological structure and stratigraphic lithology in the region are complex and changeable, which is the main reason for engineering disasters such as rock burst and large deformation in the excavation surrounding rock of deep under- ground projects such as large and long tunnels of Sichuan-Tibet Railway, underground powerhouses of hydropower stations and diversion tunnels. Slate, as a typical metamor- phic rock, is widely distributed in the western Sichuan Plateau, with lamellar bedding structure. It is soft, and the bedding is closely closed under high ground stress, which often shows rheological properties. However, after excavation and unloading, the inter- layer relaxation and large deformation are generated, and the water rationality is poor. The strength has obvious anisotropic characteristics. The stability of surrounding rock after tunnel excavation is closely related to the bedding angle [1–5]. Xie et al. [1] believed that carbonaceous slate was mainly composed of quartz, mica and albite through micro- analysis, and its strength was significantly affected by soft rock effect of water. Cai et al. [5] found that the slate strength of layered structure is closely related to load and plane angle. Garcia-Fernandez et al. [6] studied the deterioration characteristics of air humidity on the tensile strength of slate through Brazilian splitting test. The strength of rock is an important parameter for the stability analysis and support design of underground engineering excavation. As an economical, convenient and effi- cient indirect test method for rock strength, point load test has been widely accepted in the world since it was proposed in 1972. In 1985, the International Society of Rock Mechanics and Engineering formulated the standard of point load test method for rock [7]. Chinese scholars have also carried out a large number of point load tests for dif- ferent rocks and their correlation with uniaxial compressive strength of rock [8–14]. The specific test methods are specified in Standard for test methods of engineering rock mass (GB/T 50266–2013) and Standard for engineering classification of rock mass (GB/T 50218–2014). However, despite nearly 50 years of research, the application of point load test in practical engineering still has limitations, especially the uncertainty of the corresponding relationship between point load strength and uniaxial compres- sive strength of rock [15, 16]. Mahmoodzadeh et al. used artificial neural network to predict uniaxial compressive strength of slate based on point load strength value I s(50) ; Azimian et al. [17] summarized the method of obtaining the standard uniaxial compres- sive strength (UCS) of rock through the conversion of point load index I s(50) , including linear function, exponential function, logarithmic function, power function, and consid- ered that the conversion correlation of power function was the highest. For anisotropic rock, the error rate of uniaxial compressive strength predicted by point load test can even reach 100% [18]. The point load test is faster and more economical than other tests. For the rock which is not easy to obtain the standard sample to carry out indoor uniaxial compression test, especially the slate due to the existence of the layer, it is difficult to prepare the standard rock sample. Du Yaohui [2] carried out point load test on carbonaceous slate of Muzhailing Tunnel on Lanzhou-Chongqing Railway, analyzed the mechanism of large deformation of surrounding rock through point load strength value, and guided the design of support measures. Lin Chao [12] analysis of physical and mechanical properties and mineral composition of carbonaceous slate in Xingzishan tunnel of Da-Lin Railway shows that the joint fissure, strain softening characteristics of rock mass and uneven 174 W. Xingcan et al. distribution of stress and strain of surrounding rock are important factors for progressive failure of surrounding rock. However, there are few research results on the systematic microscopic view and point load strength characteristics of thin slate. Therefore, based on the diversion tunnel project of Guzeng Hydropower Station in Muli River, Liangshan Prefecture, this paper carried out XRD test and microscopic thin section identification of carbonaceous slate and sandy slate of the lower Ordovician Wachang Formation (O1 w) widely distributed in the tunnel site area, and analyzed the mineral composition and microstructure characteristics of slate. Furthermore, the irregu- lar carbonaceous slate blocks and sandy slate drilling cores were collected, and the point load tests in the vertical and parallel directions were carried out respectively. Based on the point load strength value, the applicability and characteristics of different conver- sion empirical formulas of uniaxial compressive strength were studied, and the strength characteristics of carbonaceous slate and sandy slate were analyzed. The research results have reference significance for the stability analysis of surrounding rock of slate under- ground engineering excavation in the eastern edge of Qinghai-Tibet Plateau, such as Sichuan-Tibet railway tunnel and hydropower engineering diversion tunnel. 2 Engineering Geology Condition Guzeng Hydropower Station is located on the main stream of Muli River in Muli County, Liangshan Prefecture, Sichuan Province. It is the fifth cascade of the one reservoir and six levels hydropower planning of the main stream of Muli River (Shangtongba - Abudi section). It adopts the diversion development mode. The upstream is Egongbao Hydropower Station, and the downstream is connected with the Lizhou Hydropower Station. The installed capacity of the power station is 172 MW, the length of the diversion tunnel is about 11.15 km, the reference flow is 185.2 m3 /s, the normal water storage level of the reservoir is 2215.00 m, the total reservoir capacity is 484000 m3 , and the utilization drop of the power station is 127 m. 2.1 Regional Geological Environment The project area is located in the southeast of Qinghai-Tibet Plateau and the east of Hengduan Mountains. In the Sichuan-Yunnan rhombic fault block located in the Muli River Basin, the secondary Daocheng fault block, Qianshan anticline and the NW end of Jiagou syncline are located. The active faults in the area are developed, which is one of the most complex geological structures in China and even in the world. Ground geological survey found that small and medium faults and interlayer dislocation zones along the diversion tunnel are more developed, and small folds such as wrinkles and flexure are common. The occurrence of rock strata changes violently, and structural joints are extremely developed (Fig. 1). The terrain in the area tilts from north to south, which is a high mountain deep cutting area. The river is strongly eroded and cut down, which belongs to the high mountain canyon landform. The elevation of the mountain top is generally 3500–4500 m, and the relative elevation difference is 1500–2500 m. The Muli River flows through the engineering area from north to south, and the valleys are mostly asymmetric “V” valleys. Study on the Strength Characteristics of Metamorphic Slate Based on Point Load Tests 175 Fig. 1. The local folds in project area. The strata in the project area are dominated by regional shallow metamorphic trough sedimentary formation with weak magmatic activity. The strata belong to Muli district of Zhongdian district, where the Ordovician and Triassic strata are more developed and widely distributed. There is a set of regional shallow metamorphic rock series along the diversion tunnel, from old to new mainly include: the Lower Ordovician Rengong Formation (O1 r), the Lower Ordovician Wachang Formation (O1 w), the Lower Silurian System (S1 ), the Lower Carboniferous System (C1 ), the Upper Triassic Qugasi Formation (T3 q), there are four large-scale faults. Among them, the fault is mainly composed of mylonite, breccia, a small amount of fault mud, fracture dense zone, etc. Due to multi- stage and different scale geological structure, the rock mass in the area is extremely broken, especially the layered and tabular rock mass is squeezed into intestinal shape, and the integrity is very poor (Fig. 1). In addition to long-term metamorphism, the rock mass has low strength and poor water rationality, and the stability of surrounding rock is very poor during excavation. 2.2 Physical Properties of Typical Slate According to the preliminary geological survey and excavation, it is found that the lower Ordovician Wachang Formation (O1 w) strata are widely distributed along the diversion tunnel, with a thickness of 450–720m. The lithology is mainly metamorphic quartz sandstone and feldspar quartz sandstone with gray, gray and white medium-thickness - thin-layer texture, and gray, gray and dark thin-layer - very thin-layer slate and phyllite. Among them, slate is widely distributed along the tunnel. The joint fissures of this kind of rock mass are extremely developed, and the integrity is poor. The angle between the level and the axis direction of the tunnel changes dramatically. In particular, the dark black thin carbonaceous slate is peeled off along the level after excavation unloading. It is easy to soften under the action of water, and is in the form of fragmentation. It is a typical V-type surrounding rock. The excavation self-stability is extremely poor, and local collapses occur repeatedly, which is the control rock layer of excavation construction [19] (Fig. 2). 176 W. Xingcan et al. Fig. 2. The foliated carbonaceous slate in excavation. In order to fully reveal the physical and mechanical properties of such carbonaceous slate, typical rock samples were collected, and the rock composition and microstructure characteristics were tested and analyzed by XRD and thin section projection microscopy. According to the microscopic analysis of rock samples (Fig. 3), the rock is granular crystal and plate structure. They are composed of siliceous, carbonaceous, a small amount of calcite and muscovite, and some reticular quartz veins can be seen (the colorless part of thin section in Fig. 3a), which has similar composition with that of the Cambrian Qiujiahe Formation in Qingchuan County by Sun et al. [20]. Some of them are vein quartz semi- directional distribution. The silica is colorless and mostly cryptocrystalline, partially recrystallized to form fine-grained quartz and mixed with carbonaceous distribution (black part of the flake). The quartz particle size is less than 0.05 mm, a few particles slightly coarse up to 0.1 mm. The carbonaceous is black mixed with dusty and siliceous. According to the main mineral composition and structure of slice, the rock is named black carbonaceous slate and gray black sandy slate. (a) Black carbonaceous slate (b) Gray black sandy slate Fig. 3. Microstructure of typical rock samples. Study on the Strength Characteristics of Metamorphic Slate Based on Point Load Tests 177 In order to obtain the mineral composition of the carbonaceous slate, a typical rock sample of 10–20 g was selected by using XPERT PRO equipment. The mineral com- position and content of the whole rock were determined by X-ray diffraction (XRD) after drying, grinding and passing 0.05 mm sieve. Table 1 shows the XRD results of two rock samples, and Fig. 4 is the corresponding XRD spectrum diagram. It can be seen that the main mineral compositions of the two typical carbonaceous slate samples are similar, both of which contain a large number of detrital mineral quartz (the main component is SiO2 ), accounting for 97.8% and 91.3%, respectively. They also contain a small amount of pyrite and flake-like extremely cleavage silicate mineral muscovite and clinochlore, which are similar to that of Yuanbaoshan Tunnel in Lixiang Railway [1]. The results of XRD are consistent with microstructure analysis. The carbonaceous slate is mainly siliceous, which is formed by the dynamic metamorphism of argillaceous sandstone rich in carbon. The quartz inherits the quartz sand in the original rock [20]. When the metamorphism is shallow, the rock is gray and sandy, on the contrary, it is dark black and carbonaceous. In addition, the muscovite and chlorite are hydrophilic because of water-soluble albite, it can be judged that the physical and mechanical properties of the slate will be significantly affected by water. Table 1. XRD mineral analysis results of carbonaceous slate. Numbering Quartz Albite Pyrite White mica Clinochlore Sandy slate 97.8 - 1.3 0.9 - Carbonaceous slate 91.3 3.3 1.5 0.3 3.6 Combined with the test results of microstructure and mineral composition of the carbonaceous slate exposed, the rock is mostly thin - extremely thin slice structure and extremely broken with smooth and metallic luster plane (Fig. 2). The slate rock is tightly closed under the action of three-dimensional stress without excavation, and then the rock mass is relaxed and opened into thin slices along the plane due to the unloading effect of excavation. Meanwhile, besause of the muscovite, chlorite and other minerals, the interlayer cementation ability is reduced, and the rock mass is further weathered and gradually disintegrated into loose under the action of water and atmospheric weathering. The rock mass is structurally lost and can be crushed, showing an approximately loose soil. 178 W. Xingcan et al. (a) Sandy slate (b) Carbonaceous slate Fig. 4. XRD results of Metamorphic slate. 3 Point Load Test 3.1 Experiment Design The uniaxial compressive strength (σucs ) of rock is an important parameter to evaluate the engineering characteristics, stability and chassification of surrounding rock. In practical engineering, it is mostly obtained by uniaxial compression test of standard rock samples. However, the integrity of carbonaceous slate in the lower Ordovician Wachang Formation (O1 w) exposed in this project is extremely poor, with thin – extremely thin layer structure, and the thickness of the layer is less than 10 mm. After excavation, the rock mass is relaxed into a loose structure, and hard to perform standard uniaxial compression tests by drilled cores. Study on the Strength Characteristics of Metamorphic Slate Based on Point Load Tests 179 Point load test is an indirect test method for measuring rock strength. The correspond- ing test methods and standards are given by the ISRM and Standard for test methods of engineering rock masss in China. Its advantages are simple operation, convenient sam- pling and short test cycle. It has been widely used in practical engineering, especially for rocks with poor integrity and difficult to prepare standard samples [9]. Point load test rock samples include two categories: (1) For the gray sandy slate, it is relatively integrated, but also hard to obtained standard cylindrical specimen for uniaxial compression test due to the developed rock layer. Therefore, hand-held drilling rigs is used in the field to drill cores in large blocks, and then processed into a rock with a ratio of height to diameter about 1.0 (Fig. 5a). (2) For thin carbonaceous slate, due to its low strength and poor integrity, it is impossible to obtain the complete rock sample by drilling. The rock mass with a square shape, a size of about 50 mm and size ratio of about 1.0 is selected near the tunnel face (Fig. 5b). The WH-1000 digital display point load tester of Chengdu University of Technology (Fig. 5c) is used to perform the point load test on the natural water-bearing rock mass reference to the Standard for test methods of engineering rock mass (GB/T50266–2013). Considering the obviously anisotropic characteristics of slate, the test loading is divided into two directions which are parallel and perpendicular to rock layer. The anisotropic characteristics of carbonaceous slate are analyzed by the point load strength index in two directions. Some fractured rocks tested are shown in Fig. 5b. (a) Sandy slate samples (b) Carbonaceous slate samples (c) Point load tester Fig. 5. The samples and apparatus of point load tests. Finally, a totally of 264 point load tests were performed included 19 and 36 of dark black carbonaceous slate parallel and perpendicular to bedding, respectively, and 100 and 116 point load tests of gray sandy slate. 180 W. Xingcan et al. 3.2 Experimental Data Processing According to the obtained failure load P and the equivalent core diameter De , the rock point load strength I s can be calculated: P Is = (1) De2 where, for radially tested De2 = D2 , otherwise De2 = 4WD π , D is the distance between two loading points (mm). Since I s is the point load strength of a specimen of any size, it should be converted into a modified rock point load strength value I s (50) for a specimen with 50 mm between two loading points. The relationship curve between De2 and failure load P should be drawn according to the test results, and then find the P50 corresponding to De2 = 2500 mm2 in the curve. The point load strength index (PLSI) of rock can be calculated according to the following formula: P50 Is(50) = (2) 2500 When the amount of test data is less, the following formula is used: 0.45 De Is(50) = Is (3) 50 Furthermore, the anisotropic index of rock point load strength Ia(50) can be calculated by the Is(50) perpendicular to bedding and Is(50) parallel to bedding: Is(50) Ia(50) = (4) Is(50) 4 Test Results and Analysis Due to the sufficient number of completed point load tests, the maximum and minimum point load strength test values were removed. The point load strength standard values corresponding to different loading conditions are calculated by Formulas (1)–(4), and then statistically analysed. There is a correlation between the standard value of point load strength index I s(50) and the uniaxial compressive strength σucs of rock. Many scholars have established linear and nonlinear relationships between the test results of different types of rock [11, 13, 17, 18, 21, 22].When Broch and Franklin [23] firstly proposed the point load test method, they suggested that the linear empirical formula with a coefficient of 25 should be used to predict the uniaxial compressive strength of rocks. Later, ISRM [7] summarized the recommended value of 20–25. The linear relationship without intercept is widely used at present, but the value of the linear coefficient varies greatly. It is affected Study on the Strength Characteristics of Metamorphic Slate Based on Point Load Tests 181 by lithology, sample size and other factors [24, 25]. Tsiambaos and Sabatakakis [24] believe that the coefficient is related to the soft and hard degree of the rock itself. From soft rock to hard rock coefficient changes from 13–28, the weaker the lithology is, the smaller the linear conversion coefficient. On the other hand, Kahraman [22] believes that the nonlinear relationship of soft rock is more accurate. Some scholars also proposed a variety of nonlinear formulas including quadratic function [13, 26], exponential function [22, 27], logarithmic function [28] and power function [24] through a large number of experiments. Based on the summary and application of a large number of test data by China Railway Eryuan, Standard for engineering classification of rock mass (GB/T 50218–2014) also recommed the power function: σc = 22.82Is(50) 0.75 (5) There is no single corresponding relationship between I s(50) and σucs , but it has been widely accepted that σucs increases with I s(50) [27]. Although, the error rate of σucs of anisotropic rock predicted by point load test can even be 100% [18], for rocks that are hard to preform standard uniaxial compression tests, such errors are also allowed in engineering applications [10]. To this end, after analyzing of a large number of literature data, this paper selected 10 typical linear and nonlinear formulars suitable for weaker rocks (Table 2). It should be noted that, considering the low strength of slate, the linear coefficient recommended for ISRM is set to the lower limit of 20. 4.1 Standard Characteristics of Point Load Strength Figure 6 is the distribution of I s(50) for carbonaceous slate block point load tests. The mean value and standard deviation are 3.5 MPa and 1.9, respectively, and the discrete coefficient is 0.81 under perpendicular loading (Fig. 6a). They are 2.4 MPa, 1.7 and 0.26 for parallel loading (Fig. 6b). It can be seen that the discreteness of vertical loading is larger, and its average strength is also larger. The strength anisotropy coefficient of carbonaceous slate is 1.46, indicating anisotropic characteristics. 10 10 8 8 Sample counts Sample counts 6 6 4 4 2 2 0 0 0 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 8 Point load strength index Is(50)/MPa Point load strength index Is(50)/MPa (a) Vertical loading (b) Parallel loading Fig. 6. Distribution of point load strength index of carbonaceous slate. 182 W. Xingcan et al. Figure 7 is the distribution of I s(50) for sandy slate. The mean value and standard deviation are 4.6 MPa and 2.4, and the discrete coefficient is 0.65 under vertical load- ing (axial) (Fig. 7a). Under parallel loading (radial), they are 1.6 MPa, 1.1, and 0.89, respectively (Fig. 7b). It can be seen that the I s(50) meets the normal distribution, and the discreteness of the test results in the two directions is basically the same. At the same time, the strength anisotropy coefficient of sandy slate reaches 2.88, which is more significant than that of carbonaceous slate. The I s(50) of 522 groups of point load tests on carbonaceous slate and slate indistin- guishable layers of Muzhailing Tunnel on Lanzhou-Chongqing Railway measured by Du [2] are 1.9 MPa and 3.4 MPa, respectively, which are similar to the above results. It can be seen from the I s(50) that slate is significantly anisotropy. Although the strength of sandy slate is larger, its compressive strength along the layer (radial) is smaller than that of carbonaceous slate, and the discreteness is larger. In practical engineering, the relationship between rock layer and loading direction should be paid close attention to. The stability of surrounding rock with the same lithology varies greatly with different angle combinations. 20 30 15 25 Sample counts Sample counts 20 10 15 10 5 5 0 0 0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 Point load strength index Is(50)/MPa Point load strength index Is(50)/MPa (a) Vertical loading ( axial ) (b) Parallel loading ( radial ) Fig. 7. Distribution of point load strength index of sandy slate. 4.2 Uniaxial Compressive Strength Distribution Characteristics Considering that the test results of sandy slate are more reliable and statistically, the uniaxial compressive strength is calculated by the selected 10 empirical formulars listed in Table 2, and the statistical analysis is carried out. Table 4 shows the statistical parameters of normal distribution for axial and radial σucs of sandy slate calculated by the empirical formulars in Table 2. Figure 8 shows the relationship between the statistical mean of each σucs and the dispersion coefficient. It can be seen that: Study on the Strength Characteristics of Metamorphic Slate Based on Point Load Tests 183 200 Axial Radial Average σucs /MPa 150 100 CS 50 0 0.0 0.3 0.6 0.9 1.2 1.5 Coefficient of dispersion Fig. 8. Relationship between mean uniaxial compressive strength and discrete coefficient of sandy slate. (1) According to Table 4, the average value of the axial σucs converted by the 7th exponential function reaches 5191.6 MPa, which is obviously unreasonable and is eliminated in Fig. 8. This is determined by the curve characteristics of the expo- nential function. When the I s(50) is small, the conversion result based on the expo- nential function is also small. However, the conversed σucs increased rapidly with the increasing of I s(50) . The strength of sandy slate in this experiment is higher, and the I s(50) is larger. When I s(50) > 6.0, the converted σucs will increase signif- icantly. When I s(50) > 10.0, σucs has been greater than 500 MPa. In a word, the exponential formula would overestimate the σucs of rocks with high strength. (2) According to Fig. 8, the σucs obtained by different empirical formulas are quite different. The σucs of axial loading σca = 59.4 – 151.1 MPa, and the range reaches 91.7 MPa, while that of radial loading is σcd = 23.9 – 53.6 MPa, and the range is 29.7 MPa. The maximum value is obtained by logarithmic function of the 8th , and the maximum discreteness is obtained by power function of the 5th . Therefore, the greater the rock strength is, the greater the difference of σucs obtained by different empirical formulas, the lower the reliability of the conversion results. (3) Among the 10 selected models, the σucs obtained by power empirical formula rec- ommended by Standard for engineering classification of rock mass is the smallest, which means the result is more conservative. the σucs and the discrete coefficient obtained by linear empirical formula recommended by ISRM are more reasonable when the lower limit is taken. (4) The anisotropic coefficients of sandy slate obtained by different methods are all greater than 2.0, and the anisotropic coefficients obtained by 5th and 9th empirical formular are even greater than 4.0, indicating that the strength of sandy slate is obviously anisotropic, and the σucs perpendicular to the plane is much larger. Point load test has been widely used in practical engineering due to its flexibility, convenience and low cost. Point load test is a test method that can quickly and effectively obtain the uniaxial compressive strength of rock. However, the empirical relationship 184 W. Xingcan et al. between point load strength index and uniaxial compressive strength is complex. There is no recognized and reliable unified empirical conversion formula so far. Most of them prefer to carry out both point load test of rock blocks and uniaxial compression test of standard rock samples at the same time, and then established the empirical formular. Finally, the uniaxial compressive strength of similar rock is calculated by the empirical formular. It is inevitable to carry out the uniaxial compressive tests of standard rock samples, which weaken the advantages of point load test. This is one of the biggest obstacles to limiting the application of point load test in rock engineering. In this paper, based on a large number of point load test results of sandy slate cores, 10 different empirical formulars are used to predict the uniaxial compressive strength. The comparative analysis show that the prediction effect of exponential function and logarithmic function is poor for sandy slate with obviously anisotropic. The effectiveness of linear function depends on the selected proportional coefficient. The lower limit recommended by ISRM is better than those others. The power function recommended by Standard for engineering classification of rock mass (SECR) is conservative. Table 2. Some empirical equations correlating the UCS to point load Strength Index of rock. Serial number Type Empirical Applicable Source relationship conditions 1 Linearity σ c = 20I s(50) Schist ISRM [7] 2 σc = Coal rock Kahraman [18] 23.62I s(50) −2.69 3 σ c = 21.65I s(50) Red sandstone Dai Ling et al. [8] 4 σ c = 13I s(50) Sandstone, I s(50) < Tsiambaos [24] 2 MPa 5 Power σ c = 7.3I s(50) 1.71 Soft rock 6 function σc = Not applicable to Standard for 22.82I s(50) 0.75 conglomerate and engineering σ c < 5 MPa very classification of soft rock rock mass [29] 7 Exponential σ c = 2.27exp(1.04 Dry and saturated Kahraman [22] function I s(50) ) pyroclastic rocks 8 Logarithmic σ c = 100ln(I s(50) ) Sandstone and Teymen [28] function + 13.9 other 17 kinds of rocks 9 Parabola σ c = 3.86I s(50) 2 + Pyroclastic rock Quane [26] 5.65 I s(50) 10 σ c = –0.66I s(50) 2 0 < I s(50) < 15 Zhang Jianming + 21.15I s(50) MPa et al. [13] Study on the Strength Characteristics of Metamorphic Slate Based on Point Load Tests 185 Table 3. Experimental results of Point Load Tests. Category Load Mean Standard Confidence Coefficient Number Coefficient direction strength deviation interval of of of /MPa dispersion effective anisotropy tests Carbonaceous Vertical 3.5 1.9 [2.8,4.2] 0.81 32 1.46 slate Parallel 2.4 1.7 [1.4,3.3] 0.26 15 Sandy slate Vertical 4.6 2.4 [4.1,5.0] 0.65 112 2.88 Parallel 1.6 1.1 [1.4,1.8] 0.89 96 Table 4. Experimental results of uniaxial compressive strength for sandy slate. Coefficient Coefficient Serial Load Mean σucs Standard 95% confidence of of number direction /MPa deviation interval dispersion anisotropy 1 Axial 91.4 47.9 [32.3,41.0] 0.68 2.49 ISRM Radial 36.7 21.8 [32.3,41.0] 0.69 Axial 105.2 56.5 [94.6,115.8] 0.67 2 2.59 Radial 40.6 25.7 [35.4,45.9] 0.74 Axial 98.9 51.8 [89.2,108.6] 0.65 3 2.49 Radial 39.7 23.6 [34.9,44.5] 0.69 Axial 59.4 31.1 [53.6,65.2] 0.65 4 2.49 Radial 23.9 14.2 [21.0,26.7] 0.69 Axial 114 97 [95.9,132.2] 0.96 5 4.58 Radial 24.9 23.5 [20.2,29.7] 1.10 6 Axial 69.5 27.8 [64.3,74.7] 0.50 2.00 SECR Radial 34.8 16.0 [31.5,38.0] 0.53 Axial 5191.6 17628.9 [1890.8,8492.5] 1.30 7 164.81 Radial 31.5 48.3 [21.7,41.2] 0.99 Axial 151.1 56.5 [140.1,161.7] 0.52 8 2.82 Radial 53.6 70.3 [39.3,67.8] 1.00 Axial 128.3 113.1 [107.1,149.4] 0.95 9 4.60 Radial 27.9 24.9 [22.8,32.9] 0.98 Axial 79.1 34.2 [72.7,85.5] 0.55 10 2.21 Radial 35.8 19.9 [31.8,39.8] 0.65 186 W. Xingcan et al. Table 5. Statistical results of converted uniaxial compressive strength of slate. Category Load Empirical σc mean Mean Standard 95% Coefficient Number direction relation value difference/% deviation confidence of of /MPa interval dispersion effective tests Carbonaceous Vertical ISRM 69.9 18.88 37.6 [55.3,84.4] 0.81 32 slate SECR 56.7 23.8 [47.5,65.9] 0.59 Parallel ISRM 47.0 11.06 34.2 [28.0,65.9] 0.26 15 SECR 41.8 21.4 [29.9,53.7] 0.19 Sandy slate Vertical ISRM 91.4 23.96 47.9 [32.3,41.0] 0.68 112 SECR 69.5 27.8 [64.3,74.7] 0.50 Parallel ISRM 36.7 5.18 21.8 [32.3,41.0] 0.69 96 SECR 34.8 16.0 [31.5,38.0] 0.53 4.3 Comparative Analysis Based on the above-mentioned, the linear and power empirical function recommended by ISRM and Standard for engineering classification of rock mass, respectively, have a better prediction effect on the uniaxial compressive strength of sandy slate. Therefore, these two methods are used to analyze the results of point load test of carbonaceous and sandy slate (Table 5), It shows that: (1) The average difference of σucs obtained by the two methods is small. The minimum difference is 5.18% under radial loading and the maximum difference is 23.96% under axial loading. (2) For vertical loading, the strength of sandy slate is significantly higher than that of carbonaceous slate, indicating that the strength of sandy slate block itself is larger. However, for the condition of parallel loading, the strength of carbonaceous slate is slightly higher than that of sandy slate, indicating that the cementation between carbonaceous slate strata is better. 5 Conclusion Based on a diversion tunnel project in western Sichuan, this paper takes the thin carbona- ceous slate and sandy slate of the Lower Ordovician Wachang Formation (O1 w) exposed during excavation as the objects, and comprehensively studies the strength characteris- tics of slate by XRD, thin section projection microscopy and rock block point load test. The following conclusions are obtained: (1) Due to the complex geological structures, the Ordovician Wachang group (O1 w) slate presents poor integrity, thin layer structure and joints development. It is mainly contained quartz, some albite, muscovite, pyrite, etc. (2) The point load strength of carbonaceous slate and sandy slate meet well of the normal distribution. The anisotropy coefficients are 1.46 and 2.88 respectively, which have obvious anisotropy characteristics. The overall strength of sandy slate is larger. Study on the Strength Characteristics of Metamorphic Slate Based on Point Load Tests 187 (3) Among the selected 10 kinds of empirical formulas between uniaxial compres- sive strength and point load index, the exponential and logarithmic function have the worst prediction effect, and the prediction effect of linear function depends on the selected proportional coefficient. The lower limit proportional coefficient recommended by ISRM is better, while the prediction result of power func- tion recommended by Standard for engineering classification of rock mass is conservative. (4) The difference of uniaxial compressive strength obtained by different empirical formulas would be greater for the rock with greater strength, and the reliability of conversion results is lower. (5) The method recommended by ISRM and Standard for engineering classification of rock mass are more suitable for the prediction of uniaxial compressive strength of slate. The results show that the strength of sandy slate is larger than that of carbonaceous slate, while carbonaceous slate has the better cementation. (6) In consideration of the microscopic structure and strength characteristics of the lower Ordovician Wachang Formation (O1 w) slate, in practical engineering, it should be paid more attention to the relationship between the rock layer and load- ing direction, since the stability of surrounding rock with the same lithology varies greatly with the combination of angles. Meanwhile, the water rationality of the slate is poor which the softening effect of water will significantly reduce the strength of rock and interlayer cementation. The results can provide reference for the stability evaluation of surrounding rock for a large number of deep underground engineering in slate area of Western Sichuan Plateau. Due to the extreme bedding structure and strength anisotropy of slate, the angle between the attitude of rocks and tunnel axis is an important factor affecting the engi- neering characteristics of slate in practical engineering. However, the complex tectonic action in western Sichuan has caused drastic changes in the attitude of rock mass. There- fore, it is suggested that dynamic construction and support design should be carried out under the premise of fully studying the strength and deformation anisotropy character- istics of slate in the area, combined with the measurement of exposed surrounding rock, so as to effectively reduce the construction risk. References 1. Yunpeng, X., Qiunan, C., Xiaocheng, H., et al.: An experimental study of microstructure and uniaxial compression test of carbonaceous slate in a deep buried tunnel. Hydrogeol. Eng. Geol. 47(1), 96–102 (2020). (in Chinese) 2. Du, Y.: Study on the mechanical characteristics of support structure and deformation con- trol technology of carbonaceous slate surrounding tunnel. Chang’an University (2017). (in Chinese) 3. Xiaohui, X., Yongjian, B., Yongbo, T., et al.: Geological structure and rock deformation characteristics and their geohazard-controlling mechanism in Yajiang area, west Sichuan. Geoscience 35(1), 145–152 (2021). (in Chinese) 4. Gang, C., Gong Lun, Y., Yu, J., et al.: Study on large deformation characteristics and construc- tion control technology in high altitude slate tunnel. Chin. J. Undergr. Space Eng. 16(S2), 744–751 (2020). (in Chinese) 188 W. Xingcan et al. 5. Cai, J., Du, G., Ye, H., et al.: A slate tunnel stability analysis considering the influence of anisotropic bedding properties. Adv. Mater. Sci. Eng. 2019, 1–17 (2019) 6. Garcia-Fernandez, C.C., Alvarez-Fernandez, M.I., Cardoso, R., Gonzalez-Nicieza, C.: Effect of environmental relative humidity in the tensile strength of layering in slate stone. Bull. Eng. Geol. Environ. 79(3), 1399–1411 (2020). https://doi.org/10.1007/s10064-019-01619-7 7. Franklin, J.A.: Suggested method for determining point load strength. Int. J. Rock Mech. Min. Sci. Geomech. Abstr. 22(2), 51–61 (1985) 8. Ling, D., Huayan, Y., Feiyang, Z., et al.: Experimental study on point load strength of red sandstone with different shapes. Sci. Technol. Eng. 19(07), 209–214 (2019). (in Chinese) 9. Jun, D., Chenchen, W., Siqi, W., et al.: Correlation of compressive strength of irregular coal samples by point load tests. Sci. Technol. Eng. 18(05), 291–296 (2018). (in Chinese) 10. Hu, Q.: Correlation between point load test and standard compressive strength of rock. Water Resour. Hydropower Eng. 24–26 (1997). (in Chinese) 11. Li Hongpeng, W., Yiyang, G.C., et al.: The relation between point load strength and com- pressive and tensile strength of marble. Sci. Technol. Eng. 19(32), 294–299 (2019). (in Chinese) 12. Lin, C.: Study on large deformation mechanism and deformation control technology of broken carbonaceous slate surrounding tunnel. Southwest Jiaotong University (2019). (in Chinese) 13. Zhang, J.M., Tang, Z.C., Liu, Q.S.: Relation between point load index and uniaxial compressive strength for igneous rock. Rock Soil Mech. 36(S2), 595–602 (2015). (in Chinese) 14. Yu, Z., Quan, G., Fei, Y., et al.: Study of weathering grouping and strength characteristics of wudang group schist based on point load testing. Rock Soil Mech. 33(S1), 229–232 (2012). (in Chinese) 15. Guo, M.L.: Discussion on adaptability of point load test of rocks. Rock Soil Mech. 24(3), 488–489 (2003). (in Chinese) 16. Wang, R., Tang, C., Wang, S.: Study on several problems about point load test of rock. J. Northeast. Univ. Nat. Sci. 29(01), 130–132 (2008). (in Chinese) 17. Azimian, A., Ajalloeian, R., Fatehi, L.: An empirical correlation of uniaxial compressive strength with P-wave velocity and point load strength index on marly rocks using statistical method. Geotech. Geol. Eng. 32(1), 205–214 (2014). https://doi.org/10.1007/s10706-013- 9703-x 18. Kahraman, S.: Evaluation of simple methods for assessing the uniaxial compressive strength of rock. Int. J. Rock Mech. Min. Sci. 38(7), 981–994 (2001) 19. Guo, Y., Chen, Y., Jiang, Y., et al.: Study on stability control measures for large deformation of carbonaceous slate at diversion tunnel of Guzeng hydropower station. Pearl River 41(12), 58–64 (2020). (in Chinese) 20. Ping, S.U.N., Yueping, Y.I.N., Shuren, W.U., et al.: Experimental study of microstructure and mechanical properties of rocks from Donghekou landslide. Chin. J. Rock Mech. Eng. 29(S1), 2872–2878 (2010). (in Chinese) 21. Ng, I., Yuen, K., Lau, C.: Predictive model for uniaxial compressive strength for Grade III granitic rocks from Macao. Eng. Geol. 199, 28–37 (2015) 22. Kahraman, S.: The determination of uniaxial compressive strength from point load strength for pyroclastic rocks. Eng. Geol. 170, 33–42 (2014) 23. Broch, E.M., Franklin, J.A.: The point load strength test. Int. J. Rock Mech. Min. Sci. Geomech. Abstr. 9, 669–697 (1972) 24. Tsiambaos, G., Sabatakakis, N.: Considerations on strength of intact sedimentary rocks. Eng. Geol. 72(3–4), 261–273 (2004) 25. Forster, I.R.: The influence of core sample geometry on the axial point-load test. Int. J. Rock Mech. Min. Sci. Geomech. Abstr. 20(6), 291–295 (1983) 26. Quane, S.L., Russell, J.K.: Rock strength as a metric of welding intensity in pyroclastic deposits. Eur. J. Mineral. 15, 855–864 (2003) Study on the Strength Characteristics of Metamorphic Slate Based on Point Load Tests 189 27. Diamantis, K., Gartzos, E., Migiros, G.: Study on uniaxial compressive strength, point load strength index, dynamic and physical properties of serpentinites from Central Greece: test results and empirical relations. Eng. Geol. 108(3–4), 199–207 (2009) 28. Kılıç, A., Teymen, A.: Determination of mechanical properties of rocks using simple methods. Bull. Eng. Geol. Env. 67(2), 237–244 (2008). https://doi.org/10.1007/s10064-008-0128-3 29. National standard of the People’s Republic of China. Standard for engineering classification of rock mass. Beijing: Ministry of Water Resources of the People’s Republic of China (2014). (in Chinese) Soıl Structure Interactıon Effects on the Seısmıc Response of Rc Structure wıth Rıgıd and Flexıble Foundatıons Gokaran Prasad Awadhya, Aayush Jha(B) , Mrinal Thakur, and Sanidhya Sharma Department of Civil Engineering, Delhi Technological University, Bawana Road, Delhi, India

,

Abstract. During earthquakes, soil conditions are extremely important. The study of the energy transmission mechanism from soil to buildings during seismic waves is important for multi-story building seismic design and complements the notion of soil structure for building design. In this paper, G + 10 symmetrical model with fixed (NSSI) and flexible (SSI) bases has been added at the base to investigate the effects of soil structure interaction during seismic waves. Using the Bowles technique area springs and line springs of required stiffness were assigned to the flexible base model. We have proceeded using the response spectrum method in our research because it complies to the linear dynamic features of seismic anal- ysis. Short, non-deterministic, and transient dynamic events such as earthquakes are estimated using the response Spectrum approach. The numerical findings and the graphs obtained using etabs with the SSI model are compared to those obtained using fixed-base support modelling assumptions. The study revealed that soil struc- ture interaction needs to be considered especially while designing buildings in high seismic zones. The SSI responses were studied using the peak responses of story shear, story moment, story displacement, story drift, column moments, base reactions, axial forces, and soil pressure. Keywords: Spectrum · Soil structure interaction · Story drift · Column moments · Seismic prone area 1 Introduction The interaction between the structural elements determines the reaction of the structural system composed of numerous elements. In the case of seismic waves, the soil-structure interaction effect plays a crucial role in influencing the structure’s response. Seismic waves cause motion in structures, which is determined by the structure’s vibrational properties and layout. The magnitude at which earthquake motions alter the features observable at the foundation is entirely determined by the soil’s relative mass and stiff- ness attributes. In this paper we have proceeded using the response spectrum method because it demonstrates the linear dynamic property of seismic analysis. In this paper, the study and comparison of the soil structure interaction under the effect of earthquake according to the Indian Standard Code. A 10-storey building situated in Kathmandu © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 190–211, 2023. https://doi.org/10.1007/978-981-19-3983-9_17 Soıl Structure Interactıon Effects on the Seısmıc Response 191 (Nepal) has been taken for this study. Earthquake load is applied on buildings as per IS Code 1893 part (1): 2006. Analysis is done using ETABS software. Building is anal- ysed using Response spectrum method under-ground and random motion.Movement of the ground surface in the non-existence of the structure and its foundation is termed as free-field motion. This movement is greater than the Foundation Input Motion i.e., when the motion is applied to structure in presence of line springs and area springs attached to the foundation. Because of the passive response of the structure and the deflections that response creates in the well-defined foundation springs, foundation input motion differs from true foundation base motion. Inertial interaction of the structure and foun- dation is the name given to this response. Because of the three-dimensional movement of the ground, kinematic interaction transforms the free-field ground action into foun- dation input motion. Because of the foundation system’s stiffness and resilience, the three-dimensional movement is rounded closer within the building shroud and over the foundation depth within the foundation mark.In some circumstances, Soil Structure Interaction can put up considerable variation in exactly how the structure will perform during the seismic waves and the forces employed for their design. SSI effect shows the difference between the actual response and theoretical response. In this paper, an effort is made to find the detrimental consequence of inertial interaction and are compared for both fixed and flexible base foundations.The effect of SSI subjected to gravity loads and seismic stresses is investigated using ETABS analytical software [1]. The building is modelled with assumed material properties and structural details. The building is pre- sumed to be located in a seismic prone area. The model is simulated under two different situations namely: fixed base (NSSI) and flexible base (SSI). Due to complexities asso- ciated with the modelling of structure using soil structure interaction effect, this practice is often neglected, and certain assumptions are made for simplifying the model [2]. 2 Literature Review In last half century understanding of earthquake has taken a considerable progress, which has given rise to many model’s concepts to improve the seismic performance of building under dynamic condition. The major earthquakes around the world showed the important role of soil-structure interaction (SSI) in determining dynamic behaviour of structure. Many examples may be found, such as the 1985 earthquake in Mexico City, when soft soil enhanced the ground motion and frequency of ground shaking, caus- ing havoc. Similarly, soil motion amplification caused the cypress freeway to collapse during the 1989 Loma Prieta earthquake. [Pallavi Badri et al. 2016] looked at the SSI interaction analysis of an asymmetric building supported on piled raft in the 2015 Nepal earthquake and discovered that the more complex the building, the higher the risk during an earthquake event, and that the building’s response is governed by the peak ground acceleration rather than the magnitude of the earthquake. [Julio et al. 2008] analysed G + 6 storey rcc frame building under the influence of SSI 4. Under the influence of SSI horizontal spectral acceleration values of structure were decreased. Also, the paper stated vibration period and damping were increased when SSI model was introduced to the dynamic behaviour of the building in comparison to the fixed base model [3]. [Anand et al. (2010)] used Etabs to investigate the g + 15 space frame with and without shear 192 G. P. Awadhya et al. wall for various soil conditions [4]. Between two frames, the dynamic quantities were compared. When the soil shifted from hard to medium to soft, the base shear, axial force, and lateral displacement in the column all rose at the same time, according to the report. The effect of SSI on different foundations of the construction was compared by [Priyanka et al. (2012). Response spectrum Method was used to examine the structures. [Gopal 0. Dhoot. et al. (2016)] studied performance-based design of RCC with pushover analysis under the effect of SSI. According to the findings, the seismic design provides a viable methodology for retrofitting the structure to meet the needed goal.[Behzad Fatahi et al. (2011)] investigated 15-story concrete-resisting buildings in Australia’s Ce, De, and Ee soil classifications. According to the article, the influence of SSI on inelastic seismic design based on soil type Ce is minimal, however the model’s performance level signifi- cantly raised in soil types De and Ee. [Ebrahim Rayat Roknabadi et al. (2017)] analysed the dynamic response of building using experimental tests and finite element analysis. Shaking table results were compared to SSI system finite element analysis results, which included shear velocity, frequency, damping ratio, and PSA [5]. The paper concluded SSI effects reduces with the increase of the foundation embedment. [Shylaja N et al. (2017)] analysed SSI behaviour of asymmetric 3d building frames with isolated footing under modified wrinkler model. The fundamental natural period, base shear, lateral dis- placement, axial force, and bending moment were all examined in this article [6]. The paper stated bending moment increased with the lessening in modulus of shear in the soil.According to [Nikhil N. Chopade et al. (2018)], the traditional inelastic design tech- nique without SSI is insufficient to ensure the safety of RC buildings resting on medium and soft soil. SSI impacts on seismic response needs of multi-story MRF building on raft foundation were investigated by [Tarek M. A. Alazrak et al. (2014)]. The equivalent static load, response spectrum approaches, and nonlinear time history analysis were all used to analyze the model. According to the findings, SSI has a considerable impact on the building’s base forces and roof displacement when compared to a model that ignores interaction. 3 Soil-Structure Interaction The responsive effect of soil and sub-grade with structure is referred to as soil struc- ture interaction (SSI). The SSI is the difference between the structure’s actual response and the response when the structure’s base is believed to be rigid. Inertial effects, kine- matic interaction effects, and soil-foundation flexibility effects are the three types of SSI effects.These effects can be describedin terms of engineering analysis as founda- tion stiffness and damping, foundation deformation and difference between free-field ground motion and foundation input motion [7]. At the interface, vibrations, displace- ments, and rotations cause shear, moment, and torsion, which dissipate energy and affect the system’s response. These are caused by the foundation’s elasticity.The flexural, axial, and shear deformations of the structural foundation elements result in forces and dis- placements applied by the building and the soil medium. The kinematic interaction and relative deformation at the soil foundation interface account for foundation deformation and difference between free-field ground motion and foundation input motion. Soıl Structure Interactıon Effects on the Seısmıc Response 193 Two methods are used for evaluating the above effects. In Direct method the soil and structure are assumed to be the same model and the system is analyzed. Because this method takes a long time to compute, it is often avoided. InSubstructure methodthe first step in this method is to compute free field soil motions, followed by the computation of transfer functions, which are utilized to convert free field motions into FIM. Finally, springs are assigned to represent stiffness and damping at the soil-structure contact, and the entire model system is analysed [8]. SSI is determined by structure stiffness in relation to soil stiffness, structure height or slenderness in relation to footing width and the mass of the structure in relation to the mass of the earth that supports the structure. When seismic waves reach the foundation’s bottom, they are separated into two types, transmission waves that enter the structure and reflection waves that bounce back into the ground. When a transmission wave enters a structure, creating vibration, it reflects and travels back to the foundation, causing the soil-structure phenomena. The radiation wave is a portion of the wave that is sent back to the earth. Seismic waves are trapped when radiated waves are insignificant, and the building begins to vibrate continuously for a long time [9]. In comparison to the structure itself, radiation damping generates an increase in the total damping of the soil-structure system. The natural frequency of the soil structure system is also lower than the natural frequency of the soil due to the SSI effect. This con- tact increases the structure’s overall displacement, allowing the foundation to translate and rotate. There are two types of interactions in the soil-structure interaction. Kinematic Interaction which explains how translation excitation causes kinematic interaction and Inertial Interaction which accounts for the fact that the structure’s mass and subsequent hammering can produce soil compliance. Inertial Interaction (SSI) is induced by inertial forces and is associated with the mass of the structure, resulting in overturning moment and transverse shear in the vibrating structure. 3.1 Respone Spectrum Response-Spectrum Analysis (RSA) is a linear dynamic analysis approach for estimating a structure’s peak response from a subset of the system’s odes recovered using the Eigen frequency extraction procedure. It investigates dynamic behavior by determining pseudo- spectral acceleration, velocity, displacement, and damping over a particular time. A reaction spectrum is a graph of the greatest linear SDOF response for a certain component of earthquake ground motion. In this plot, natural period is on the x axis whereas response quantity (max. displacement, max. velocity or max. acceleration) is on the y-axis as shown in Fig. 1 [10]. Maximum response of SDOF system depends only on the natural frequency and damping. 194 G. P. Awadhya et al. Fig. 1. Response spectrum graph. Dynamic analysis is basically performed for regular building in zone IV and V for height greater than 40 m and in zone II and III for height greater than 90 m and for irregular buildings located in zone IV and V for height greater than 12 m and zone II and III for height greater than 40 m as per IS code. Due to lateral load from earthquake forces the building displaces. The response of the structure, due to application of lateral load induces displacement, velocity, and acceleration in the structure. The assessment of these factors is studied in response spectrum method. 4 High Structure-to-Soil Stiffness Ratios Lengthens Period and Change Design Forces If the structure is comparatively stiff in contrast to the soil, foundation rotation and translation comparative to the free-field motion can arise adding up to the structural displacements and expanding or lengthening the structural period of the structure. These increased periods can influence the related spectral accelerations employed for the design purpose. This impact normally arises in structures with a concentrated lateral force- resisting structure. On the other hand, the effect of soil flexibility on the building reaction is generally quite minimal in structures with large, stiff foundations on very stiff soils and flexible structures.The fundamental period of vibration is lengthened by soil flexi- bility. The line springs and area springs attached to the bottom demonstrate the soil and structure’s vertical flexibility. In the horizontal displacement of the model, there might be vertical displacement of the springs positioned at the bottom of the foundation and the rotation of foundation might occur. This increase in displacement corresponds to an increase in the structure’s time period [11]. Soıl Structure Interactıon Effects on the Seısmıc Response 195 The time-period refers to the building’s fundamental translation or natural period. It signifies the natural period of the oscillation of the building when it is subjected to ground shaking during seismic waves. The height of the building taken for analysis is 30.48 m which gives the time-period to be 0.9729 s according to the formula given below as per IS 1893:2016(part 1). Time-period for bare moment resisting frame buildings. Ta = 0.0075 h0.75 ............eqn (1) where, h = height of the building. 5 Site Characterization The site considered for this project lies in the severe most seismic zone i.e., Seismic Zone V with highest seismic zone factor of 0.36. Soil type II is considered for this project as per IS 1893: 2006 part (1). The soil structure interaction model took into account medium and stiff soil with poorly graded sands and gravel (SP) with few particles. The safe bearing capacity of the soil is set at 135 KN/m2 , and the stiffness (ks) of the line and area springs is calculated using the Bowles Method. The foundation of the 10-storey building is positioned on the upper most soil layer with unit weight of soil as 20 KN/m3 as per the statistical records obtained through official website. The site condition consists of a thick deposit of black silty beds, the “Kalimati Formation” for 3 m and Greyish Moderately Weathered Rock beneath. 5.1 Modelling and Formulation The G + 10 Storeys system geometry is located in Kathmandu and has a plan dimension of 14.7486 m × 10.261 m. The structure will be utilized as a dwelling. Reinforced concrete frames support the lateral and vertical load resisting systems. Columns, primary beams, and secondary beams make up the frames. The material and geometric properties are shown in Table 1. (Table 2 and 3). Table 1. Material and geometric properties of frame, raft and concrete S. no Structure Component Details 1 Frame a. Storey height 3.048 m b. Beam size 300 mm × 400 mm c. Column size 450 mm × 450 mm 2 Raft a. Size 243.16 m2 b. Thickness 600 mm c. Beam size 600 mm × 600 mm 3 Concrete a. For columns M25 b. For beams and slabs M25 c. For raft M25 4 Shell a. Slab 125 mm 196 G. P. Awadhya et al. Table 2. List of seismic parameters Category Parameter Zone V Zone factor 0.36 Importance factor 1.2 Response reduction factor 5 Vertical irregularity in geometry NO Soil type II i.e., Medium soil or Stiff soil Table 3. Load combinations S. no Load combination S. no Load combination S. no Load combination 1 1.5DL 8 1.5(DL-EQx) 15 1.2(DL+LL+RSx) 2 1.5(DL+ LL) 9 1.5(DL+EQy) 16 1.2(DL+LL+RSy) 3 1.2(DL+LL+EQx) 10 1.5(DL-EQy) 17 1.5(DL+RSx) 4 1.2(DL+LL-EQx) 11 0.9DL+1.5EQx 18 1.5(DL+RSy) 5 1.2(DL+LL+EQy) 12 0.9DL+1.5EQy 19 0.9DL+1.5RSx 6 1.2(DL+LL-EQy) 13 0.9DL-1.5EQx 20 0.9DL+1.5RSy 7 1.5(DL+EQx) 14 0.9DL-1.5EQy 5.2 Models 5.2.1 NSSI (Non-soil Structure Interaction) In this G + 10 model, the structure’s base is rigid, i.e., a fixed support is allocated to the structure’s base as shown in Fig. 2. This model is assigned different load scenarios, such as dead load, live load, and seismic loads, employing 20 distinct load combinations, as previously indicated [12]. The response spectrum method was used to model this construction, and it passed the concrete frame design examination [13]. Soıl Structure Interactıon Effects on the Seısmıc Response 197 Fig. 2. NSSI model. 5.2.2 SSI (Soil Structure Interaction) In this G + 10 model at the base of the structure raft foundation is designed using Bowles method. Then using the empirical formulae foe stiffness of the spring, area springs were assigned underneath the raft foundation and the line springs were also assigned at the base of the raft foundation along the beams connecting the columns. The load combinations were same as in the fixed support model. The model shown in Fig. 3 personifies the soil structure interaction study [14]. 198 G. P. Awadhya et al. Fig. 3. SSI model. 6 Results and Discussion 6.1 Story Drift The largest comparable displacement of each level divided by the height of the floor in question is the story drift ratio. For a 10-story building, the maximum story drift response plots of fixed and flexible bases are evaluated and scrutinized. The response spectrum approach is used to calculate the seismic response plots in accordance with IS 1893:2006 part 1. For fixed and flexible base structures, the maximum story drift response values are 0.001448 and 0.001735, as shown in Fig. 4 and Fig. 5 rspectively [15].The drift ratio increases gradually and reaches its maximum value at the second storey level, as seen in the graph. The intensity of the story drift ratio increases as soil stiffness decreases. Because the underneath soil state varies from stiff to soft, the story drift ratio is higher for flexible mat foundation. The upward trend is more pronounced in the upper and lower storeys. As the number of stories rises, the Soil Structure Interaction effect on the tale drift ratio becomes increasingly significant [16]. Soıl Structure Interactıon Effects on the Seısmıc Response 199 Fig. 4. Story drift (SSI) Fig. 5. Story drift (NSSI). 200 G. P. Awadhya et al. 6.2 Story Shear The lateral load imposed by seismic waves is depicted per storey by story shear [17]. The maximum storey shear value for fixed and flexible foundation buildings can be calculated using the response spectrum approach. The narrative shear values rise as soil stiffness falls. The flexible base foundation’s minimum story shear is -623.423703 KN (See Fig. 6), which appears in the first floor of a 10-story structure, while the rigid base foundation’s minimum story shear is -637.05047 KN (See Fig. 7), which appear in the ground floor of the building [18]. Fig. 6. Story shears (SSI) Fig. 7. Story shears (NSSI) Soıl Structure Interactıon Effects on the Seısmıc Response 201 6.3 Maximum Story Displacement The SSI model amplifies lateral displacements and inter-story drifts, especially for MRF buildings resting on relatively soft soil. This increase in lateral deformations may affect the building frames’ performance. For a G + 10 storey building, the maximum story displacement response plots of fixed and flexible foundation are explored. The displace- ment increases in SSI models; the displacement grows more in foundations located on soft soil, and this value reduces as soil stiffness increases. For fixed and flexible foun- dation constructions, the maximum story displacement response values are 18.68255 and 56996658655 respectively (See Fig. 8 and Fig. 9). Because of the flexible basis, the story displacement value for the flexible base construction exceeds the IS code limit. In accordance with IS code 1893:2016 (part 1), maximum story displacement should be less than 0.004 times the height of the building [19]. Fig. 8. Maximum story displacement (SSI) Fig. 9. Maximum story displacement (NSSI) 202 G. P. Awadhya et al. 6.4 Story Overturning Moment Moments, uplift forces, and shears applied to structures cause the footing to become unstable, resulting in the footing turning over and the structure failing during high- intensity seismic waves. In the case of rigid foundations, however, these overturning moments have no influence on the structure. Resisting moments at the soil-foundation contact help to sustain the footing and prevent it from overturning. Fixed and flexible bases have maximum overturning moments of 0 KNm and 24.846326 KNm, respectively (See Fig. 10 and Fig. 11). These fixed and flexible base values demonstrate the need of addressing soil structure interaction when constructing a structure in order to achieve a stable and efficient structure. The overturning moment increases as the stiffness of the soil increases, as seen by these data and respective graphs. At the edge of the footing, these overturning tests are presented [20]. Fig. 10. Story Overturning Moment (SSI) Soıl Structure Interactıon Effects on the Seısmıc Response 203 Fig. 11. Story overturning moment (NSSI) 6.5 Soil Pressure The precise estimate of contact pressure for flexible foundations is a difficult task. As a result, Terzaghi has assumed that the subgrade response modulus (Ks) for the entire footing remains constant. This means that the pressure ratio, as well as the factor of safety and settlement, will remain constant at all positions of the footing. The maximum soil pressure intensity, 140 KN/m2 , is found near the center of the footing (See Fig. 12). As a result, the highest displacement will occur at this spot, which is known as the dished effect. As we travel away from the center, both the soil pressure and the displacement decrease. Only 25% of the mat foundation surface area is influenced by each node of a perfect square or rectangular foundation baseThe soil pressure at each site can alter- natively be determined by dividing the support base response by the tributary area at that point. Because the rigid foundation settles flat, the distribution of the subgrade reac- tion across the foundation base must be planar for permanent foundations [22]. Flexible foundations, on the other hand, do not have a linear subgrade reaction since their base is dependent on the foundation’s compressibility and structural stiffness. For studying the effect of Soil and Structure Interaction, it is critical to design a flexible foundation that takes into account the effect of modulus of subgrade reaction. The relative displacement of the structure decreases as the structural rigidity of the foundation increases, and the soil pressure beneath the structure decreases. Even with higher modulus of subgrade reaction the pressure distribution also reduces. Which means the Ks will take up more stress at the point of load application [21]. 204 G. P. Awadhya et al. Fig. 12. Soil pressure of SSI model 6.6 Column Moments The main sources of moments in the columns under gravity action are lateral stresses from wind or seismic waves, as well as eccentricity of the supported beams. The strength and bending qualities of the columns are monitored by the concrete’s compressive strength, column shape, and even the longitudinal strength of the reinforcements [23]. Moment 3-3 is the bending of a beam or frame around the 3-3 axis (See Fig. 14 and Fig. 16), while Moment 2-2 denotes the bending of a column on the positive and negative 2 faces around the 1 axis of consideration (See Fig. 13 and Fig. 15). The largest peak moment occurs along the 2-2 axis in the local direction. Because of larger displacement, overturning moments, and other factors, these moments are higher in flexible base structures. Soıl Structure Interactıon Effects on the Seısmıc Response 205 Fig. 13. M 2-2 (SSI) 206 G. P. Awadhya et al. Fig. 14. M 3-3 SSI Soıl Structure Interactıon Effects on the Seısmıc Response 207 Fig. 15. M 2-2 NSSI 208 G. P. Awadhya et al. Fig. 16. M 3-3 NSSI Soıl Structure Interactıon Effects on the Seısmıc Response 209 7 Conclusion Seismic forces were applied to a G + 10 storeyed building frame with a fixed and flexible base, which were analyzed and developed using ETABS software. Using the plots indicated above, the seismic responses of the building frames such as story shear, story moment, story displacement, story drift, column moments, base reactions, axial forces, and soil pressure were analyzed. When soil structure interaction is taken into account, the study of the soil-structure model yields a wide range of responses. The fixed support model’s values were within limitations, i.e., maximum storey displacement was less than 0.004 * building height, and storey drifts were less than 0.004. Fixed base buildings had lower lateral deflection, storey drifts, base shear, and moment values than flexible foundation buildings. When the soil condition at the site is not adequate, i.e., when the structure rests on loose soil, the soil structure interaction effect must be considered.This research could be useful in developing safe design guidelines for seismic design of structures that take into account soil flexibility. The Response Spectrum approach is used to investigate the effects of slab, column, mat foundation, structure, and soil model components (IS 1893 part1:2006). The findings obtained using a fixed base model and a flexible base model assumption are compared to analyze soil structure interaction. The stiffness of the foundation base reduces as the soil springs are added, but the fundamental period of the structure increases, implying that the fundamental period of the structure is not only a function of structure height, but also of soil condition. As a result, it improves the concept of modeling the structure while considering the soil-structure interaction concept. SSI models will thus have a longer period than NSSI models. The story displacement response of SSI models with flexible base foundations is higher than that of fixed base models (NSSI). The consequences of soil structure interaction become exacerbated as the structure’s storey increases. Soil structure interaction has a greater impact on the bottom storey displacements than on the rest of the storeys. As a result, if the SSI principle is not considered while designing multistorey buildings, they are far more likely to fail. With a decrease in soil stiffness and even an increase in the structure’s storey, the story drift ratio rises. The tale drift ratio of the SSI model with flexible base was higher than the fixed base model (NSSI). The effects of SSI on overturning moments, base reactions, and column moments are then evaluated using this model. In comparison to the conventional hypothesis, which ignored soil structure interaction, the investigation demonstrates that the SSI has a significant impact on every structural response. The structure’s dynamic responses differ from the fixed base condition. The seismic behavior of mid-rise structures is heavily influenced by these dynamic responses. Thus, SSI effects are critical for seismic design of mid-rise structures, especially when they are built on soft soil bases and in high seismic zones like Seismic Zone IV and V. If SSI is not taken into account during analysis and design, the precision in determining structural safety when confronted with seismic waves may not be consistent. As a result, while designing building frames for seismic stresses, a proper foundation system that takes into account the influence of soil stiffness ”must be used. 210 G. P. Awadhya et al. References 1. Wong, F.S.: Uncertainties in dynamic soil-structure interaction. J. Eng. Mech. 110(2), 308– 324 (1984). https://doi.org/10.1061/(asce)0733-9399(1984)110:2(308) 2. Wong, B.H.L.: “In the absence of soil-structure interaction effects have been presented by Mita and Luco (1990a, 1990 ) and Luco et al. (1991). The form and effectiveness of active control in the presence of soil-structure interaction effects are examined by considering structures with different heights supported on soils with different rigidities. The effects on the response of the structures and on the amplitudes of the required control Structural Response with Absorbing Boundary To formulate and solve the complete problem, we first consider the motion of the superstructure for a given translation and rotation of the base. Let u ( x ) e’ ° equal the harmonic translation of the structure with respect to an inertial frame of reference . For small displacements, the total displacement u ( x ) can be decomposed into u ( x ) = u B + xd B + u R ( x ),” vol. 117, no. 10, pp. 2237–2250 (1992) 3. Pilecki, T.J.: Interface model for dynamic soil-structure interaction. J. Geotech. Eng. 113(5), 537–538 (1987). https://doi.org/10.1061/(ASCE)0733-9410(1987)113:5(537) 4. Dicleli, M., Albhaisi, S., Mansour, M. Y.: Static soil–structure interaction effects in seismic- isolated bridges, 10(1), 22–33 (2005) https://doi.org/10.1061/(ASCE)1084-0680(2005)10 5. S. S. I. Analysis and P. For, . “i: a,” vol. 125, no. January, pp. 26–37, (1999) 6. Jonathan, P., Gregory, L., Raymond, B., Stewart, B.J.P., Fenves, G. L., Seed, R.B.: S Eismic S Oil -S Tructure I Nteraction in B Uildings, vol. 125, pp. 38–48 (1999) 7. Renzi, S., Madiai, C., Vannucchi, G.: A simplified empirical method for assessing seismic soil-structure interaction effects on ordinary shear-type buildings. Soil Dyn. Earthq. Eng. 55, 100–107 (2013). https://doi.org/10.1016/j.soildyn.2013.09.012 8. Ghosh, B., Madabhushi, S.P.G.: Centrifuge modelling of seismic soil structure interaction effects. Nucl. Eng. Des. 237(8), 887–896 (2007). https://doi.org/10.1016/j.nucengdes.2006. 09.027 9. Wolf, J.P., Song, C.: Some cornerstones of dynamic soil-structure interaction. Eng. Struct. 24(1), 13–28 (2002). https://doi.org/10.1016/S0141-0296(01)00082-7 10. Spyrakos, C.C., Maniatakis, C.A., Koutromanos, I.A.: Soil-structure interaction effects on base-isolated buildings founded on soil stratum. Eng. Struct. 31(3), 729–737 (2009). https:// doi.org/10.1016/j.engstruct.2008.10.012 11. Tang, Y., Zhang, J.: Probabilistic seismic demand analysis of a slender RC shear wall con- sidering soil-structure interaction effects. Eng. Struct. 33(1), 218–229 (2011). https://doi.org/ 10.1016/j.engstruct.2010.10.011 12. Ganjavi, B., Hao, H.: A parametric study on the evaluation of ductility demand distribution in multi-degree-of-freedom systems considering soil-structure interaction effects. Eng. Struct. 43, 88–104 (2012). https://doi.org/10.1016/j.engstruct.2012.05.006 13. Stefanidou, S.P., Sextos, A.G., Kotsoglou, A.N., Lesgidis, N., Kappos, A.J.: Soil-structure interaction effects in analysis of seismic fragility of bridges using an intensity-based ground motion selection procedure. Eng. Struct. 151, 366–380 (2017). https://doi.org/10.1016/j.eng struct.2017.08.033 14. Sucasaca, J., Sáez, E.: Topographical and structure-soil-structure interaction effects on dynamic behavior of shear-wall buildings on coastal scarp. Eng. Struct. 247, 113113 (2021). https://doi.org/10.1016/j.engstruct.2021.113113 15. Kausel, E.: Early history of soil-structure interaction. Soil Dyn. Earthq. Eng. 30(9), 822–832 (2010). https://doi.org/10.1016/j.soildyn.2009.11.001 16. Zogh, P., Motamed, R., Ryan, K.: Empirical evaluation of kinematic soil-structure interaction effects in structures with large footprints and embedment depths. Soil Dyn. Earthq. Eng. 149, 106893 (2021). https://doi.org/10.1016/j.soildyn.2021.106893 Soıl Structure Interactıon Effects on the Seısmıc Response 211 17. Matinmanesh, H., Asheghabadi, M.S.: Seismic analysis on soil-structure interaction of build- ings over sandy soil. Procedia Eng. 14, 1737–1743 (2011). https://doi.org/10.1016/j.proeng. 2011.07.218 18. Luco, J.E., Trifunac, M.D., Wong, H.L.: Isolation of soil-structure interaction effects by full- scale forced vibration tests. Earthq. Eng. Struct. Dyn. 16(1), 1–21 (1988). https://doi.org/10. 1002/eqe.4290160102 19. Wong, H.L.: Wong-SFSIThesisCaltech.pdf 20. Stewart, J.P., Fenves, G.L.: System identification for evaluating soil-structure interaction effects in buildings from strong motion recordings. Earthq. Eng. Struct. Dyn. 27(8), 869–885 (1998). https://doi.org/10.1002/(SICI)1096-9845(199808)27:8%3c869::AID-EQE762%3e3. 0.CO;2-9 21. Baker, J.W.: Measuring bias in structural response caused by ground motion scaling. In: Pacific Conference on Earthquake Engineering, no. 056, pp. 1–6 (2007). https://doi.org/10.1002/eqe 22. Stewart, J.P.: Soıl-Structure Interactıon Effects by Raymond, B., Seed and Gregory, L. Fenves Report No . PEER-98/07 Pacific Earthquake Engineering Research Center University of California Berkeley, California November 1998 ii, Environ. Eng., no. November (1998) 23. Isbiliroglu, Y., Taborda, R., Bielak, J.: Coupled soil-structure interaction effects of building clusters during earthquakes. Earthq. Spectra 31(1), 463–500 (2015). https://doi.org/10.1193/ 102412EQS315M Study on Treatment of Ground Subsidence in Goaf Hong Qiang Zhang(B) , Ling Gao, and Ming Yue Feng Hebei Provincial Communications Planning, Design and Research Institute Co., Ltd, Shijiazhuang 050011, China

Abstract. In the construction of expressway, the ground collapse caused by under- ground mining restricts the route selection of expressway. Therefore, before the construction of expressway, it is necessary to find out the impact of Goaf Collapse on expressway construction. In order to reduce the risk of geological disasters, safeguard social and public interests and alleviate social contradictions, it is very necessary to carry out gob survey and treatment research. Find out the distribution of goaf through data collection, investigation and visit, measurement, engineer- ing geological mapping, geophysical exploration, drilling, in-situ test, sampling test and other methods, and obtain the optimal treatment scheme in combination with the actual situation of the site and relevant engineering experience for peer reference. Keywords: Goaf · Ground Subsidence · Expressway · Treatment 1 Introduction Due to the failure to fill in time after the underground goaf is formed, the original stress balance state of the overburden is damaged. When highway construction is carried out above the goaf, when the influence depth of subgrade load overlaps with the caving zone and fracture zone of the goaf, the goaf may be “activated”. The “activation” process of goaf is difficult to observe directly. Its “activation” process is affected by a variety of natural and human factors. The impact on the surface is quite complex and a long- term process, which may occur a few years or decades or even hundreds of years after mining. As long as there is residual deformation in the “activated” goaf, it is still possible to reactivate, The main threat objects are the life and property safety of passing vehicles and pedestrians and the stability of expressway subgrade. Therefore, it is necessary to control the goaf [1–3]. This paper investigates and studies the treatment scheme for the ground collapse of goaf passing by an expressway. 2 Engineering Geological The overlying loess soil layer in the exploration area is 0.5–12 m thick, with slight to medium collapsibility and a small amount of gravel in some parts; The overlying © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 212–220, 2023. https://doi.org/10.1007/978-981-19-3983-9_18 Study on Treatment of Ground Subsidence in Goaf 213 rock mass in the goaf is mostly gray and gray white strongly ~ moderately weathered dolomitic limestone with brecciated dolomitic limestone and limestone. Affected by the intrusion of tectonic magma, the rock mass is relatively broken ~ extremely broken, mostly in fragment shape, a small amount in column shape, and mixed with multi-layer iron bearing amphibolite and diorite vein, which is irregularly distributed. 2.1 Meteorology and Hydrology The exploration area has a temperate monsoon climate, with large temperature differ- ence and four distinct seasons. It is dry and windy in spring, hot and rainy in summer, cool and cool in autumn, cold and dry in winter with little rain and snow. The annual average temperature is 13.5 °C, and the average temperature in the coldest month is − 2.3 °C. The average temperature in the hottest July is 26.9 °C, the extreme minimum temperature is −19.9 °C, and the extreme maximum temperature is 42.5 °C; The annual average precipitation is about 549 mm, with uneven distribution within the year and large interannual variation. The annual precipitation is concentrated from June to September, accounting for about 75% of the total annual precipitation, and mostly in the form of rainstorm. The annual sunshine is 2200−2700 h, and the frost free period is 191−215 d. Affected by the high-pressure air flow in Inner Mongolia in winter, this area is mostly northwest wind in winter and southwest wind in spring and autumn. Strong winds of force 7−8 often occur, with an annual average wind speed of 2.6 m/s. The maximum frozen soil depth is 0.53 m. Nanming River is about 2 km to the southwest of the exploration area, which origi- nates from the south foot of Motianling in the territory, turns from northwest to southeast to northeast, and intersects Beiming River to the east of Yonghe village, with a total length of 95 km. This river is a seasonal river. Except for a small amount of spring water in the upstream guantaochuan, the rest are dried up for a long time, and the flood is discharged only in flood season, with an annual runoff of 71.8 million m3 . Surface water flows in the river in rainy season and fluctuates with rainfall. 2.2 Topographic Features The project area is located at the eastern foot of Taihang Mountain, which is surrounded by xiaomotianling and Shibapan mountains, with an Intermountain basin in the middle. The exploration area is located in the gentle slope zone of low mountains, heavy hills and belongs to the low mountain area (II 2) with erosion structure dominated by carbonate rocks. The terrain is high in the West and low in the East. The elevation is 433.2−448.3 m. The surface vegetation is not developed, and more crops are planted. 2.3 Formation Lithology The bedrock at the top of the exploration area is exposed, and most of it is covered with quaternary Middle Pleistocene slope diluvium. The strata from new to old are described as follows: Quaternary Middle Pleistocene proluvial and Deluvial loess like silty clay and silty clay (Q2 pl + dl ): distributed in the slope zone of the hillside, with maroon silty clay at the 214 H. Q. Zhang et al. bottom and gravel layer. The soil layer has vertical joints, the bedding is not obvious, has the general characteristics of Malan loess, and has slight to medium collapsibility. The thickness varies, mostly 0.5−12 m. Middle Ordovician Majiagou Formation (O2): gray and gray white dolomitic lime- stone with breccia dolomitic limestone. Affected by structural intrusion, the rock mass is relatively broken ~ extremely broken, mostly in fragment shape, a small amount in column shape, and mixed with multi-layer iron bearing amphibolite and dioritevein. Yanshanian magmatic rocks (δ52 ): magma intrudes along the NNE trending fault beam, and the surface is irregularly distributed locally. The lithology is complex, mainly diorite and amphibolite, intercalated with multi-layer iron bearing amphibolite. Ore bodies occur in veins in amphibolite metamorphic strata and are irregularly distributed. 2.4 Neotectonic Movement and Earthquake There is a small north-east transverse concealed normal fault 500 m south of the bridge site area. The overall strike of the fault is 23−36° NNE and the dip angle is 55−75°. The hanging wall of the fault is Sinian chert dolomite and dolomitic limestone, and the footwall is Ordovician Majiagou Formation limestone. In the route area, the southwest fault trend gradually changes from NNE to NE, and is concealed under the overburden. Through this survey, the fault is not a Holocene Active fault. Affected by the fault, structural breccia is developed in the Middle Ordovician (O2 ) limestone layer in the bridge site area, structural fractures are developed along the rock contact zone, water erosion marks on the contact surface are obvious, and dissolution is generally developed. Geological Structure. The exploration area is located in the North Korean paraplatform I2 of the first-class tectonic unit, the Xishan fault uplift II 23 of the second-class tectonic unit, the Taihang arch short beam III 211 of the third-class tectonic unit, and the Wu’an concave fault beam IV 233 of the fourth-class tectonic unit. Most faults strike NNE and NW, and the development degree is weak in the West and strong in the East. Most magmatic rocks intrude along the faults in the East. The intrusive rocks in the exploration area intrude along the NNE trending fault bundle, and the surface presents irregular local distribution. Ore bodies occur in veins in amphibolite metamorphic strata and are irregularly distributed. Earthquake. According to the county chronicles, there has been no earthquake above magnitude 6 in and around the exploration area, which belongs to a relatively stable area. According to the seismic ground motion parameter zoning map of China (GB 18306– 2015), the basic seismic fortification intensity in the exploration area is VII and the basic seismic acceleration is 0.15 g [4]. 2.5 Hydrogeological Conditions Surface Water. Groundwater in the exploration area is mainly bedrock fissure water and pore phreatic water in Quaternary loose overburden. Phreatic water mainly occurs in the contact part between slope diluvium and bedrock. Due to the high terrain, there is little Study on Treatment of Ground Subsidence in Goaf 215 occurrence; Bedrock fissure water mainly occurs in shallow moderately weathered rock fissures and goaf. The recharge source is atmospheric precipitation, which is discharged to low-lying areas, and the dynamic change of groundwater is greatly affected by climate factors. There was no surface water during the survey. After the rainstorm in June 2017, the deep goaf was filled with groundwater. According to the water quality analysis results of water samples taken from nearby sites, groundwater is slightly corrosive to steel bars in concrete structures and reinforced concrete structures. To sum up, the conditions of groundwater infiltration, runoff and discharge in the exploration area are good, and its circulation and discharge are fast. The rapid flow of groundwater makes the shallow rock and soil in the goaf soften, disintegrate and sink. Especially under the condition of rainstorm, the rainwater pours back, and the messy pillars in the original goaf are washed away and collapsed. In addition, the surface scouring induces the further development of ground cracks and collapses. 3 Ground Subsidence in Goaf 3.1 Overview of Mineral Exploitation According to the surface mapping and survey visit, the iron ore mining is composed of four main mines in the periphery, one main mine in the middle and three to four small mines, of which the main mine 1 is located near 70 m to the right of K25 + 042; The main mine 2 is located near 30 m to the left of K25 + 054; Main mine 3 is located near 60 m to the left of K25 + 228; The main mine 4 is located near 153 m to the right of K25 + 184; The middle main mine 5 is located near 26 m to the right of K25 + 172. The small mines are basically buried, and the four main mines in the periphery are mined in the middle. The mining methods are shaft and ramp, and the mining roadway and main mining area are basically without support or temporary support. The main mining area of the iron mine is banded, located in the west of K25 + 100−K25 + 200, with a width of 60−100 m, a buried depth of 60−100 m and a maximum mining thickness of 20 m. All mines are mined at this banded vein, and a small amount of chicken nest mining is carried out in other places, which is irregular. 3.2 Distribution Characteristics of Goaf Surface Collapse Collapse area: it is banded, and it basically intersects vertically at K25 + 160−k25 + 170. The intersection is narrow and 10 m wide. The main collapse part is 60−100 m west of K25 + 100−K25 + 200, 50–60 m wide. Two collapse pits are formed in the middle of ore body mining. Collapse pit 1 is 60−90 m right of K25 + 120−K25 + 160, about 20−30 m wide. The edge surface is mostly messy, and a collapse pit is formed at the bottom, with a depth of 15−20 m, It is now a garbage dump, which has not been filled; The collapse pit 2 is located at 120−130 m on the right side of K25 + 150−K25 + 170, 10 m wide and 5−10 m deep. It has been filled with garbage and there is no obvious trace on the surface. 216 H. Q. Zhang et al. Crack area: located at the edge of the collapse area and the extension part of the depression, 5 cracks are mainly distributed. Crack 1 is located 40m to the left of K25 + 133, intersects the route K25 + 125, runs 40° to the northwest, nearly parallel to the collapse edge, extends 140m to 118 m to the right of K25 + 125, with a width of 10−50 cm and a visible depth of 1−3 M; Crack 2 is located 56.8 m to the left of K25 + 150, intersects the route K25 + 148, extends 52° to the northwest, extends 55 m to the collapse edge, with a width of 3−10 cm and a visible depth of 0.5−2 m; Crack 3 is located 58 m to the left of K25 + 164, intersects the route K25 + 174, extends 45−50° to the northwest, extends 120 m along the collapse edge to the tail end of crack 1, with a width of 5−20 cm and a visible depth of 0.5–3M; Crack 4 is located 41 m to the left of K25 + 225, intersects K25 + 210, extends 72 m to the northwest 45° to the collapse edge. Generally, the main mine 3 extends to 1−2 small mines, with a width of 10−50 cm, the widest part is nearly 1m, the visible depth is 2−4 m, and the surrounding loess has been covered by collapse; Crack 5 is located 12 m to the left of K25 + 232, intersects the route K25 + 230, extends 22° to the northwest, extends intermittently from main mine 3 to main mine 4, and is mostly covered by loess collapse. The local visible width is 3−6 cm and the visible depth is 0.5−1.5 m. In addition, some fine cracks are distributed at the collapse edge, which is caused by shallow collapse and tension crack. 3.3 Cause Analysis of Goaf Ground Collapse Most of the mining roadway and goaf have no support or temporary support, generally no filling, and the roof of the goaf is free to collapse. The overlying rock mass of the goaf is strongly ~ moderately weathered dolomitic limestone with breccia limestone. Affected by the intrusion of tectonic magma, the rock mass is relatively broken ~ extremely bro- ken, mixed with multi-layer iron bearing amphibolite and diorite vein. The engineering geological properties are poor, and the soil is above it. Therefore, the goaf surface will produce strong displacement, settlement and deformation. When the collapse zone and fracture zone develop into the soil layer, collapse will occur under the action of strong permeability of drainage. Since the mining has been terminated for nearly 15 years, and after the rainstorm in June 2017, the rainwater is poured along the cracks or collapse of the goaf, and there is no obvious change in the collapse area, indicating that the two collapse areas have completely collapsed and have no further development. Therefore, the hidden danger body in the current goaf has been basically stable, the impact of continuous deformation is light, and the impact on the geological environment is small. However, when engineering construction is carried out above the goaf, when the influence depth of building load overlaps with the caving zone and fracture zone of the goaf, it may cause “activation” of the goaf, cause large land subsidence and deformation, and cause building damage. 4 Ground Subsidence in Goaf According to the survey visit and survey results, combined with the collapse and fracture development form, it is determined that the goaf belongs to small kiln goaf. According Study on Treatment of Ground Subsidence in Goaf 217 to technical rules for highway design and construction in goaf (JTG/T d31–03-2011), the limit equilibrium analysis method is adopted for the roadway in goaf, and the critical depth Hcr to maintain the stability of roadway roof is calculated according to appendix d.0.5, so as to obtain the stability coefficient of roof in goaf [5]. After exploration and analysis, the cavities under the subgrade are divided into three types: residual shallow branch roadway, deep branch roadway, deep main roadway and goaf. The width of branch roadway is 2−5m, and the width of main roadway and goaf is 5−10 m. According to the technical rules for design and construction of goaf highway (JTG/T d31–03-2011), the limit equilibrium analysis method is adopted for goaf roadway. When h reaches a certain depth, the roof rock stratum just maintains natural balance (i.e. q = 0), at this time, h is called the critical depth Hcr. Calculate the critical depth Hcr to maintain the stability of roadway roof according to appendix d.0.5, as shown in formula 1. μ Bγ + B2 γ 2 + 4Bp0 γ 1−μ tan ϕ Hcr = (1) μ 2γ 1−μ tan ϕ where: B - roadway width in goaf (m); P0 - subgrade base pressure of highway, including driving load, subgrade and pavement load (kPa); γ —Weighted average weight of overburden (kN/m3); ϕ—Weighted average internal friction angle of overlying strata (°); μ—Weighted average Poisson’s ratio of overlying strata. Among them, the roadway width adopts the upper limit of the range known in the investigation, and the highway base pressure is the sum of the pressure generated by the Quaternary overburden (10 kPa) and the driving load (20 kPa). According to the geotechnical physical and mechanical indexes, the weighted average gravity, average internal friction angle and average Poisson’s ratio of the overburden are taken as 25 kN/m3, 55° and 0.20 respectively. The average internal friction angle of overlying strata is given by referring to the similar friction angle of limestone in the engineering geology manual. The similar internal friction angle takes into account the cohesion of rock. It is concluded that the roof stability evaluation result of shallow support roadway is unstable, which needs to be treated by engineering measures, and the roof of deep main roadway is basically stable, which cannot be treated; There is no need to deal with the stability of goaf roof in deep branch roadway. 5 Study on Treatment Scheme of Ground Collapse in Goaf The small pile abutment is flat, with pebbles on the surface and bedrock under it. Good abutment stability. The slope of abutment with large chainage is about 37°, the surface is exposed as syenite porphyry, the rock stratum is stably distributed, and no adverse geology such as landslide and collapse is found. Good abutment stability. Pile foundation is adopted in abutment design, which can effectively increase abutment stability and reduce abutment slope sliding risk. If the gravity abutment is adopted, the sliding and 218 H. Q. Zhang et al. overturning stability shall be checked. Slope stability shall be considered in abutment subgrade excavation and bridge head cone slope design. 5.1 Governance Principles According to the surface mapping and survey visit, the iron ore mining is composed of four main mines in the periphery, one main mine in the middle and three to four small mines, of which the main mine 1 is located near 70 m to the right of K25 + 042; The main mine 2 is located near 30 m to the left of K25 + 054; Main mine 3 is located near 60 m to the left of K25 + 228; The main mine 4 is located near 153 m to the right of K25 + 184; The middle main mine 5 is located near 26 m to the right of K25 + 172. The small mines are basically buried, and the four main mines in the periphery are mined in the middle. The mining methods are shaft and ramp, and the mining roadway and main mining area are basically without support or temporary support. Ensure the scientificity, feasibility and rationality of the prevention and control project. The gob collapse prevention and control project shall be planned uniformly, highlight the key points according to the geological environment background conditions, adjust measures to local conditions, and take comprehensive treatment measures. The prevention and control scheme shall have the characteristics of reliable technology, reasonable economy, simple structure and strong operability. Comprehensive prevention and control should implement the principle of combining engineering measures with administrative measures. 5.2 Common Treatment Schemes Grouting Method. Grout materials with filling and cementation properties are injected into cavities, collapse zones, fracture zones and geotechnical voids in goaf manually, so as to increase their strength after hardening, support overlying strata, and control and reduce the damage of ground deformation to ground buildings. It is applicable to the treatment of goaf and roadway without underground construction conditions. Dry Masonry Method. The abandoned but not fully collapsed roadway with under- ground construction conditions shall be backfilled and built manually with rubble. Mortar Method. It is mainly used for the roadway with defective supporting structure. Excavation and Backfilling Method. Excavate the shallow buried goaf or roadway, and then backfill with dry masonry, mortar masonry or other methods. Bridge Span Method. Span the influence range of goaf or roadway in the form of bridge. It is applicable to goaf or roadway with shallow buried depth (several meters to tens of meters). Study on Treatment of Ground Subsidence in Goaf 219 5.3 Governance Scheme Adopted for the Project In recent years, with the development of China’s expressway construction, Shanxi and Henan have many successful examples in the management of Expressway goaf, such as Zheng Shao line in Henan, Chang Jin line, Yang Hou line, Fen off-line line, Lijun line and Dayun line in Shanxi. The buried depth of the managed goaf ranges from tens of meters to more than 500 m, which have obtained successful experience. The buried depth of the goaf of the project is large, the width of the influence zone of the goaf is large, the roof has collapsed after mining, and obvious ground collapse has been formed. Obviously, mortar masonry, dry masonry, excavation and backfilling are not available. If the bridge span method adopts pile foundation, the stable stratum of the goaf floor needs to be used as the pile end bearing stratum. The pile length is too large, and it is difficult to cast into pile due to the existence of cavity, which is not technically feasible. If long-span bridges are used, the construction is difficult and the cost is high due to the large span; The goaf does not have the conditions for underground construction. Combined with the successful experience of previous projects, the grouting method full filling pressure grouting method should be selected for the project. 6 Abutment Stability Evaluation The small pile abutment is flat, with pebbles on the surface and bedrock under it. Good abutment stability. The slope of abutment with large chainage is about 37°, the surface is exposed as syenite porphyry, the rock stratum is stably distributed, and no adverse geology such as landslide and collapse is found. Good abutment stability. Pile foundation is adopted in abutment design, which can effectively increase abutment stability and reduce abutment slope sliding risk. If the gravity abutment is adopted, the sliding and overturning stability shall be checked. Slope stability shall be considered in abutment subgrade excavation and bridge head cone slope design. 7 Conclusion The exploration of the project has collected the preliminary work results, basic geolog- ical data and hydrogeological data around. Through investigation and visit, geological mapping, geophysical exploration, drilling and other exploration means, the geological environment background conditions of the exploration area have been basically found out. Through the analysis of the ground collapse stability of goaf, the treatment schemes of ground collapse in different goaf areas have been studied, Finally, the grouting method - full filling pressure grouting method is selected for treatment. References 1. Gou, D., Tian, J., Li, J., et al.: Grouting treatment technology in goaf of small coal mine under expressway tunnel. J. For. Eng. 3(5), 142–149 (2018) 2. Yan, X., Ke, B., Luo, l., et al.: Construction and quantitative analysis of stability evaluation index system of a goaf. Mod. Min. (8), 194–196 (2016) 220 H. Q. Zhang et al. 3. Yin, L., Li, W., Yang, X., et al.: Application of grouting method in foundation treatment engineering in goaf 4. Seismic ground motion parameter zonation map of China (GB 18306–2015) 5. Technical rules for design and construction of goaf highway (JTG/T d31–03–2011) Forward Modeling and Detection of the Potential Slip in Artificial Slope by GPR Zhenjun Zhang1 , Luo Ye1 , and Gao Lv1,2(B) 1 Mechanical Engineering College, Xi’an Shiyou University, Xi’an, China

2 Shaanxi Key Laboratory of Geotechnical and Underground Space Engineering, Xi’an University of Architecture and Technology, Xi’an, China Abstract. Human activities have a huge impact on natural slopes and can eas- ily cause natural disasters such as landslides. This leads to obvious environmental damage. Based on the method of Ground Penetrating Radar (GPR), the space posi- tion and filling properties of artificial high embankment slope slip planes were stud- ied. With the numerical simulation method, the transient reflection-transmission rule and characteristics of geological radar waves were studied. The imaging dif- ferences of geological radar were caused by the inhomogeneous distribution of relative dielectric constant and conductivity of artificial layers. GPR data was closely related to the relative dielectric constant of each artificial layer. The echo data had less affected by the relative dielectric constant of the deep artificial layer. The spatial position of the slip plane can be precisely deduced by the hierarchical numerical model. These conclusions of the artificial slope slip surface should be an important guiding significance for detection. Keywords: GPR · Artificial slope · Finite-different time-domain method 1 Introduction Artificial slope was a new artificial slope in construction. It’s influenced by human factors, also limited by the bearing capacity of the artificial foundation. And it is probably slipped during the process of construction or project operations [1, 2]. The surface detection technique of the non -disturbance has great importance to monitor potential slips of the fill slope if it was prior considered [3–5]. In this paper, the Ground Penetrating Radar (GPR) of non-destructively detecting can confirm the slips’ space position and fill shape timely and effectively was introduced. The GPR of detecting in slope’s engineering mainly has the following results: Sass et al. used the GPR to collect large date to study the Alps slope slips [6], which are caused by unconformity layer and stratified cuttings of persistent layers, they also analyzed the slip plane’s thickness and internal structure of the slope of the Swabian Alb Jurassic sedimentary rock field [7], the slope’s marl surface are soil clay and limestone. Zieliński A et al. used the GPR of detecting to analyze a certain Poland highway slope [8], and they proposed warning measures for potential dangers by analyzing the slope’s internal © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 221–228, 2023. https://doi.org/10.1007/978-981-19-3983-9_19 222 Z. Zhang et al. structural characteristics. Fankhauser K et al. used the method of the ground and airborne geological radar to analyze and determine the position of the bedrock of Switzerland’s steep slope [9], also the time and space distribution of water content of slope are analyzed with the method of conductivity in their analysis. CHEN Jigang et al. adopted a water injection system drilling TV system and the GPR to detect the depth of deep - mining and heavy - slanted seam floor, and digitally analyzed the number of floor cracks and borehole depth crack width and quantities before and the floor is mined [10]. Wang Yong-Qiang et al. used the GPR to detect the open mine slope and revealed the potential geological defects of the interior of the steep slope [11]. Li Daxin et al. speculated the causes of the slope’s collapsed by using the results of detecting Chongqing steel company material farm bank slope [12], which in virtue of high positioning accuracy and high resolution of the GPR. Recently, the research on the problem of sliding off the artificial slope of artificial fill slope by the positive inverse method is less studied. The influence laws of different compactness and water content of soil on the deep harmful geologic body was studied based on theoretical derivation numerical calculation and layered construction. The article also studied the spatial distribution of slope slip under laminar filling by Finite- Different Time-Domain and positive inverse method. 2 Impact of Human Engineering on the Environment Since the reform and opening up, China’s economic construction has made great achieve- ments, but the development model of high input, high consumption and high pollution has directly led to the severe situation of excessive resource and environmental costs of economic growth. Human beings are increasingly aware of the negative impact of engineering construction on the environment, thus investing a lot of human and financial resources in design, materials, construction, recycling, supervision and other aspects of continuous efforts to achieve green ecology, and even give up economic benefits for the environment in order to achieve sustainable development. Engineering construction should be synchronized with ecological environment construction planning, synchro- nized implementation, synchronized development, realize the unity of economic ben- efits, social benefits, ecological benefits, environmental benefits, in the protection of development, in the development of protection, tend to benefit and avoid harm, to avoid shortcomings, to promote the coordinated development of engineering construction and ecological environment, to achieve the harmony of man and nature. 3 The GPR Echo Model of the Layered Electromagnetic wave reflection can be generated in different dielectric layers, the GPR wave will generate reflection in each interface, while the GPR wave in the refractive part will also be discounted gradually, according to this principle, the multilayer geologic model of artificial high-fill is established, and the optimized 2d multilayer geological model as Fig. 1. Forward Modeling and Detection of the Potential Slip 223 Fig. 1. Geoelectric model of the multilayer fill. Supposing the layer of the filling is evenly distributed, the relative permittivity is ε1–εn, the thickness is h1–hn and the one-way trip time of each layer of GPR is t1–in. The thickness of each filling is defined as follows: h = h1 + h2 + h3 + · · · + hn (1) And the total detection time is defined as follows: t = 2 × (t1 + t2 + t3 + · · · + tn ) h1 hn (2) v1 = √cε = t 1 ; vn = √cεn = t n 1 Finally, two-way travel time equation of GPR wave is as follows t1 t2 t3 tn−1 tn h c = √ ε1 + √ ε2 + √ ε3 +···+ √ εn−1 + √ εn h = h1 + h2 + h3 + · · · + hn−1 + hn (3) t = 2 × (t1 + t2 + t3 + · · · + tn−1 + tn ) 4 Reflection Mechanism of Special Medium Interface The difference of reflection is generated by different dielectric constants can be further analyzed. Based on GPRMAX2D, this section adopts the Finite-Different Time-Domain to carry out a more specific image analysis of the reflection characteristics of differences of dielectric constant. In this section, the three kinds of material air water, and metals, which have obvious dielectric characteristics in the artificial filling layer are analyzed, and these three sub- stances can form reflection features in the artificial layer. Based on its characteristics, the geological model is established as Fig. 2. 224 Z. Zhang et al. Fig. 2. Test model of the special medium interface reflection. The basic setting of this model is as follows: the horizontal distance is 2.0 m, the detecting depth is 0.7 m, the unit grid is 0.0025 m × 0.0025 m, the boundary conditions are the perfect matching interface, the time depth is 40 ns, wavelet frequency is 900 MHz, the excitation source is the Ricker wavelet, the thickness of artificial is 0.35 m, reflector, and the special medium reflector is 0.35 m. The transmitting of GPR and receiving of the antenna in the surface of artificial scans from left to right, and then the reflection images can be drawn about the air-water and metal. At the same time, setting up a special medium reflector of pure soil as a contrast model, as shown in Fig. 3. Fig. 3. (a)–(d) Echo characteristics of (a) loess layer, (b) Air layer, (c) Water layer and (d) Metal layer, respectively. The images of different reflectors obtained by GPR are shown in Fig. 3, in which diagram (a) is a set of pure artificial without reflection, the diagram shows that, in the 1 ns–5 ns range, the waveform has very obvious change, this is the characteristics of the amplitude of Ricker wavelet. The whole waveform of 5 ns–40 ns is of uniformity, it indicates that there is no obvious reflection characteristic of geological radar wave inhomogeneous artificial. Diagram (b) shows that the GPR wave contacts with the air interface at a depth of 0.35 m, and then cause the waveform changes, the waveform and energy will invert when the geologic radar wave meets the air interface in the artificial. Diagram (c) shows the interface reflection of the GPR wave when it spreads from the artificial to the water at 0.35 m. Similarly, the water makes an obvious reflection of the GPR wave, about 99% of the waves generate reflection, the rest electromagnetic signals transmit and continue to spread, the deep image shows multiple reflected waves. Forward Modeling and Detection of the Potential Slip 225 Diagram (d) shows the reflection of the metal interface encountered by GPR waves. The reflection of metal on electromagnetic waves is close to 1, so the electromagnetic wave will generate complete reflection, and form the most obvious reflection images. As a kind of new idea and new method of geotechnical engineering, ground- penetrating radar is convenient and flexible in specific engineering, this paper combines the feature of relative permittivity properties of artificial and the feature image of analy- sis and back analysis, also the typical applications in the project of the GPR is the high fill slope slip detection. 5 Application in Zhen’an Slope 5.1 Overview of the Landfill Project of Artificial Slope The Zhen’an Pumped Storage Project is located in Pusatian Village, Yuehe Town, Zhen’an County, Shangluo City. The slope of the slag site is mainly Platform 1. The #1 construction platform is mainly located near the upper and lower storage link road pile number 4 + 800. The site covers an area of about 49,000 m2 in the middle and lower sections of the Red Cliff Gulch. The site is centered on the Red Cliff Gulch, with lush vegetation on both sides of the slope. The highest elevation of the site is 1110 m and the lowest elevation is 990 m, with a maximum height difference of 120 m. The slope of the site is about 40°–60°, with a gentle slope up and down. The bottom of the trench is about 3 m–5 m wide. Longitudinal slope drops of 28%. The ditch has perennial flow- ing water. The trench has lush vegetation, bedrock outcrops. No accumulation of loose material. 1# Construction platform affected by terrain. The site can only be spread along the Red Cliff Gulch and both banks. The site mainly outcrops marble. The depth of weak weathering is generally greater than 15 m–20 m, the rock body is more complete, and the development of fissures is common. Within 1 m of the surface layer, the rock body is fragmented and locally weathered into gravelly soil with plant roots interspersed, as shown in Fig. 4. 5.2 Detection and Analysis of Slope Slip Surface of Artificial Landfill According to the field situation, given the sliding position of the artificial high fill slope, the internal slip plane should be east-west, therefore the lines from east to west are respectively set in the edge and inner edge of the filling platform, and lines connecting the edge and inside are also stetted, to comprehensively understand the space position of the deep sliding surface of the filled foundation. The primary detection of the 10–20 m underground was detected by 100 MHz antennas. The main data are as follows: Diagram (a), (b) in Fig. 5 respectively correspond to line 1 and line 2. According to line 1 in diagram (a), 0 ns–40 ns indicates the surface of the filled soil, the working range is covered with Artificial, the GPR wave reflected even signals in this scope. The plus or minus the amplitude approximately, which means that the filling layer is compacted and average. Within the scope of 40 ns–120 ns, more obvious reflectors are appeared, according to the proposed rule, the relative dielectric constant of the filled soil at that point is significantly greater than the surface, and therefore, there is higher water content 226 Z. Zhang et al. Fig. 4. Construction platforms and on-site testing (a)Scanning image in a succession of GPR - line 1 (b) Scanning image in a succession of GPR - line 2 Fig. 5. Scanning image in a succession of GPR Forward Modeling and Detection of the Potential Slip 227 in the soil layer. The positive and negative wave reflections are more uniform in 120 ns– 240 ns, there is no obvious structural damage to the filling layer, and the water content is higher than the surface layer. At 240 ns–500 ns, the reflection signal is weak, and the interference signal is significantly increased. In this range, it entered into the bedrock body, and the influence of bedrock on electromagnetic wave attenuation is significant, it suggests that the water content in bedrock is high. According to line 2 in the diagram (b), 0 ns–40 ns indicates the surface of the filled soil, the working range is covered with artificial, the GPR wave reflected even signals in this scope, plus or minus the amplitude approximately, it shows that the filling layer is compacted and average. The reflectors are more obvious in the cope of 40 ns–120 ns, according to the proposed rule, the relative dielectric constant of the filled soil at that point is significantly greater than the surface relative dielectric constant, therefore, there is higher water content in the soil layer. The positive and negative wave reflections are more uniform in 120 ns–320 ns, there is no obvious structural damage to the filling layer, and the water content is higher than the surface layer. At 320 ns–500 ns, the reflection signal is weak, and the interference signal is significantly increased. In this range, it entered into the bedrock body, and the influence of bedrock on electromagnetic wave attenuation is significant, it suggests that the water content in bedrock is high. 6 Conclusions Based on the study of reflection and transmission mechanism of the GPR wave in the artificial layered soil, and the practical application of detection of the slope sliding surface, the conclusions are as follows: 1. The signal energy of the ground-penetrating radar is proportional to the square of the wave amplitude, so the energy size of the echo signal can also reflect the numerical of the wave amplitude. And the echo energy and wave amplitude are related to the relative permittivity of each layer medium. 2. With the increase of water content of fill layers, the multiple waves of the ground- penetrating radar gradually became obvious, and the position of multiple waves gradually extended to the time, the reason is that the speed of an electromagnetic wave in the media of higher relative dielectric constant increases gradually. When the filling layer has high conductivity, it has a strong attenuation effect on the electromagnetic wave, thus affecting the detection of the GPR wave on the deep filling. 3. The propagation wave amplitude of radar waves changes a little in a simple artificial sample. When radar waves reflect in the air interface, the amplitude of the wave is gradually shifted from positive to negative, this change is due to the shift of characteristics of the wave amplitude when it spreads from the artificial to the air. The artificial are kind of medium with high relative dielectric constant while the air is a lower kind of relative dielectric constant. The reflection of the water interface has the characteristics when the wave spreads from the medium of the low relative dielectric constant to the medium of high relative dielectric constant. The reflection on the metal interface is more obvious, and the reflection is almost twice as much as the reflection of the water interface. 228 Z. Zhang et al. Acknowledgment. Thanks the financial: Provincial College Students’ innovation and Entrepreneurship Project (s201910705073). Shaanxi Key Laboratory of Geotechnical and Under- ground Space Engineering (JT201901), Shaanxi Province Science Foundation for Youths (Grant No. 2018JQ5203). Shaanxi province housing and construction research and development projects (2020-K41). References 1. Hu, C.M., Mei, Y., Liu, Z.R.: Deformation mode and stability analysis of high sticking slope of collapsible losse. Chin. J. Rock Mech. Eng. 12, 2585–2592 (2012) 2. Ma, Y., Wang, J.D., Peng, S.J.: Deformation and failure mechanism of high sticking artificial slope. Chin. J. Geotech. Eng. 03, 518–528 (2016) 3. Zhu, C.H., Li, N.: Mesoscopic deformation mechanism of loses high-fill foundation based on soil electrical resistivity. Chin. J. Rock Mech. Eng. 03, 640–648 (2013) 4. Zhu, C.H., Li, N., Liu, M.Z.: Spatiotemporal laws of the post-construction settlement of the artificial-filled foundation of Lüliang airport. Chin. J. Geotech. Eng. 02, 293–301 (2013) 5. Su, L.H., Li, N., Lv, G.: Artificial soil water content fast nondestructive testing of airport surface. J. Xi’an Univ. Technol. 01, 40–44 (2015) 6. Sass, O., Krautblatter, M.: Debris flow-dominated and rockfall - dominated talus slopes: genetic models derived from GPR measurements. Geomorphology 86(1), 176–192 (2007) 7. Sass, O., Bell, R., Glade, T.: Comparison of GPR, 2D-resistivity, and traditional tech- niques for the subsurface exploration of the Öschingen landslide, Swabian Alb (Germany). Geomorphology 93(1), 89–103 (2008) 8. Zieliński, A., Mazurkiewicz, E., Łyskowski, M., et al.: Use of GPR method for investigation of the mass movements development on the basis of the landslide in Kałków. Roads and Bridges-Drogi i Mosty 15(1), 61–70 (2016) 9. Fankhauser, K., Guzman, D.R.L., Oggier, N., et al.: Seasonal response and characterization of a scree slope and active debris flow catchment using multiple geophysical techniques: the case of the meretschibach catchment Switzerland. EGU Gen. Assembly Conf. Abs. 17, 11833 (2015) 10. Chen, J.G., Xiong, Z.Q., Li, H.: Failure characteristics of the floor under pressure inclined and extra-thick coal seam in full-mechanized top coal caving faces. Chin. J. Rock Mech. Eng. S1, 3018–3023 (2016) 11. Wang, Y.Q., Cao, Z., Tan, Q.W.: GPR detection of open-pit landslide. J. China Coal Soc. 36(7), 1093–1097 (2011) 12. Li, D.X., Qi, M.S., Wang, C.L.: Georadar investigation for hidden danger in river dykes. Chin. J. Geol. Hazard Control 01, 21–25 (1996) Application of High Density Electrical Method in Karst Area Hong Qiang Zhang(B) , Zheng Bo Cao, and Wei Li Hebei Provincial Communications Planning, Design and Research Institute Co., Ltd., Shijiazhuang 050011, China

Abstract. When building bridges in limestone areas, karst development often affects the quality of piles and bridge safety. Karst development presents diver- sity, complexity and irregularity. If the development degree characteristics of karst cannot be found out, it will increase great difficulties and potential safety hazards for later construction. In the past, the development degree of karst cannot be com- pletely found out by drilling alone. This requires a combination of geophysical exploration and drilling. Using high-density electrical method, the karst develop- ment can be determined according to the inversion abnormal low resistance area. According to the geophysical results and combined with the drilling data, the characteristics of karst development degree can be fully displayed in space, and the design and construction work can be carried out on the basis of accurate and detailed geological data, so as to save the project cost, shorten the construction period and ensure the safety of bridge construction and later operation. Keywords: High density electrical method · Bridge survey · Karst 1 Introduction With the construction of more and more bridges in mountainous areas, a single drilling method cannot meet the requirements of builders. Especially in limestone area, karst development often affects the pile quality and bridge safety. If we can’t find out the development degree of karst, it will increase great difficulties and potential safety hazards for later construction. In this paper, the bridge survey in limestone area of a project is studied. The high-density electrical method combined with drilling can well find out the degree of karst development and provide accurate geological data for designers [1–3]. 2 Project Overview 2.1 Bridge Structure The upper part of a bridge is designed to adopt 16–30 m continuous T-beam, the lower part adopts column pier and column/ribbed slab abutment, and the foundation adopts pile foundation [3, 4]. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 229–233, 2023. https://doi.org/10.1007/978-981-19-3983-9_20 230 H. Q. Zhang et al. 2.2 Topography and Geomorphology The bridge site belongs to river terrace landform with slight topographic fluctuation. The surface elevation of the bridge site is 464.0−507.0 m. 2.3 Meteorological The project area has a warm temperate humid and semi humid continental monsoon climate with four distinct seasons, The general characteristics are: sudden cold and warm in spring (March−May), dry and lack of rain, mild climate and abundant rainfall in summer (June−August), sunny weather in autumn (September−November), large temperature difference, cold, dry and little snow in winter (December−February). The annual average temperature is 6.5 °C−10.3 °C. The data show that the minimum tem- perature in this area is minus 28.6 °C (January 15, 1981) and the maximum temperature is 36.6 °C (July 16, 1972). The precipitation in the year is unevenly distributed, mainly from June to September, accounting for about 80%−90% of the annual precipitation; the annual average rainfall is 727 mm, the maximum annual precipitation is 1122.6 mm, the maximum frozen soil depth is 1.2 m, and the annual average frost free period is about 135 days. The northwest monsoon prevails in winter and the southeast monsoon in summer. 2.4 Formation Lithology The surface layer of the bridge site area is covered with quaternary Holocene alluvial peb- ble layer and partially covered with quaternary upper Pleistocene proluvial and Deluvial gravel; The underlying bedrock is mainly middle Ordovician limestone and calcareous sandstone. The large chainage abutment is exposed as Yanshanian circumgyrate por- phyry syenite, and the Carboniferous mudstone and structural breccia zone are partially exposed. The engineering geological conditions are complex. 2.5 Geological Structure There is a small north-east transverse concealed normal fault 500 m south of the bridge site area. The overall strike of the fault is 23−36° NNE and the dip angle is 55−75°. The hanging wall of the fault is Sinian chert dolomite and dolomitic limestone, and the footwall is Ordovician Majiagou Formation limestone. In the route area, the southwest fault trend gradually changes from NNE to NE, and is concealed under the overburden. Through this survey, the fault is not a Holocene Active fault. Affected by the fault, structural breccia is developed in the Middle Ordovician (O2 ) limestone layer in the bridge site area, structural fractures are developed along the rock contact zone, water erosion marks on the contact surface are obvious, and dissolution is generally developed. 2.6 Hydrogeological Conditions Surface water. The surface water in the bridge site is xidaogou River, a tributary of Liuhe river. It is a seasonal river. The water volume changes with rainfall and has the characteristics of sudden rise and fall. It is a typical mountain river. Application of High Density Electrical Method in Karst Area 231 Groundwater. During the survey, the groundwater level in the bridge site area is 7.30−13.50 m, the groundwater is mainly Quaternary pore water, the atmospheric pre- cipitation infiltration and river lateral supply are the main supply modes of groundwater, and the discharge modes are mainly groundwater runoff, river discharge and manual min- ing. According to the water quality analysis results of the detailed engineering geological survey report of Chengping expressway, the chemical type of shallow groundwater in this area is mainly SO4-CaMg, which is slightly corrosive to the steel bars in concrete structures and reinforced concrete structures. It is recommended to take corresponding anti-corrosion measures. 3 High Density Resistivity Method 3.1 Working Methods Duk-4 high density resistivity measurement system produced by Chongqing Geologi- cal Instrument Factory is used for the high density resistivity method instrument. The system includes host dzd-8 multi-functional DC method instrument, mis-10c cascade electrode converter and distributed large line cable. The system has the characteristics of large storage capacity, accurate and fast measurement and convenient operation. Its main functions include automatic signal acquisition and automatic storage, The sys- tem is equipped with domestic advanced high-density electrical method forward and inverse interpretation software, which greatly improves the accuracy and efficiency of geophysical interpretation. High density resistivity method is to infer the formation change by measuring the resistivity change of the measured section and analyzing the geological drilling data. Different formation lithology has certain electrical differences, which can be inferred and interpreted according to electrical differences. The measurement and observation system adopts Wenner device, and the interpre- tation section is inverted trapezoid. The distance between the measuring points is 5 m. The field recorded data is replayed to the computer. After inspection and calibration, it is processed by professional processing software. The processing mainly includes section Fig. 1. Schematic diagram of high density resistivity method device. 232 H. Q. Zhang et al. resistivity contour drawing, forward and reverse interpretation, and color section drawing of forward and reverse interpretation. According to the resistivity contour map, forward and reverse interpretation resistivity section map, geological interpretation is carried out in combination with drilling data, and the geophysical interpretation geological profile is drawn. As shown in Fig. 1. 3.2 Technical Principle High density resistivity method is to infer the formation change by measuring the resis- tivity change of the measured section and analyzing the geological drilling data. Dif- ferent formation lithology has certain electrical differences, which can be inferred and interpreted according to electrical differences. 4 Workload Layout According to the 2016 edition of code for seismic design of buildings (gb50011–2010) and code for seismic design of Highway Engineering (jtgb02–2013), combined with the survey data, the geological structure of the project site is relatively stable and belongs to the general seismic section. The equivalent shear wave velocity of the site soil layer is 500 ≥ VSE > 250 m/s, the site soil type is medium hard soil, and the project site type is class II [4, 5]. 5 Abutment Stability Evaluation The small pile abutment is flat, with pebbles on the surface and bedrock under it. Good abutment stability. The slope of abutment with large chainage is about 37°, the surface is exposed as syenite porphyry, the rock stratum is stably distributed, and no adverse geology such as landslide and collapse is found. Good abutment stability. Pile foundation is adopted in abutment design, which can effectively increase abutment stability and reduce abutment slope sliding risk. If the gravity abutment is adopted, the sliding and overturning stability shall be checked. Slope stability shall be considered in abutment subgrade excavation and bridge head cone slope design. 6 Conclusion There is a small north-east transverse concealed normal fault 500 m south of the bridge site area. The overall strike of the fault is 23−36° NNE and the dip angle is 55−75°. The hanging wall of the fault is Sinian chert dolomite and dolomitic limestone, and the footwall is Ordovician Majiagou Formation limestone. In the route area, the southwest fault trend gradually changes from NNE to NE, and is concealed under the overburden. Through this survey, the fault is not a Holocene Active fault. Affected by the fault, structural breccia is developed in the Middle Ordovician (O2 ) limestone layer in the bridge site area, structural fractures are developed along the rock contact zone, water erosion marks on the contact surface are obvious, and dissolution is generally developed. Application of High Density Electrical Method in Karst Area 233 The results of high-density electrical method reveal that there is a closed low resis- tance anomaly within 37 m to the right of K7 + 285 ~ k7 + 285 and 31−54 m in depth. It is speculated that this is a karst cave development area or stratum rich in water. There is closed low resistivity anomaly within the range of 38 m to the right of K7 + 495 ~ k7 + 495 and the depth of 25−38 m. It is speculated that there is karst cave or stratum rich water. There is closed low resistivity anomaly within 25 m from the left of K7 + 555 to 30 m from the right of K7 + 555, and the depth is 25−65 m. It is speculated that this is a karst cave development area or stratum rich in water. References 1. Zhao, G., Chen, X., Tang, J.: New progress and development trend of earth electromagnetic method in China. Adv. Geophys. (4) (2007) 2. Yin, J., Ling, W., Lin, Y.: Application of high frequency magnetotelluric in deep tunnel detection. Water Conservancy Plan. Design, (9), 12–13 (2015) 3. Liu, H., Wang, H.: Application of comprehensive exploration technology in expressway along the river. Eng. Constr. (6), 803–805 (2009) 4. Code for seismic design of buildings (gb50011–2010), Edition 2016 5. Code for seismic design of Highway Engineering (jtgb02–2013) Green Building Construction Implementation Barriers in the Philippines a Hierarchical Model Cris Edward Monjardin(B) , Lovely Jasmin Dela Cruz, Ezekiel Esguerra, and Ondrea Nhika Dangzalan Mapua University Intramuros, Manila, Philippines

, {ljdelacruz,elesguerra, oncdangzalan}@mymail.mapua.edu.ph Abstract. The Philippines is one of the developing countries that attempt to prac- tice green building. Although with 300 + registered and certified green building projects acknowledged by LEED, green building is less practiced in the Philip- pine Construction Industry due to several barriers. Limited studies were conducted that discuss green building in the Philippines, especially the barriers to its imple- mentation. This research created hierarchical models of barriers to green building implementation in the Philippines by identifying, classifying, and ranking the existing barriers. The RRL looked into the technicalities of the green building certification, compiled the barriers identified by previous studies, and provided this research with its theoretical framework as the basis for the proposed method- ology. The objective was achieved using two (2) types of survey questionnaires, answered by the two (2) chosen respondents: multi-unit residential occupants and qualified construction experts with knowledge or experience with green build- ing. These questionnaires are put through statistical analysis which resulted in the research to identify and also rank the green building barriers under Social and Economic aspects of sustainability, significant in the Philippines. Finally, this research is geared towards assisting professionals in the construction industry, decision-makers, and policymakers in determining the best strategy for dealing with these challenges. Keywords: Green building · Building · Barriers · Hierarchical · Construction · Engineering 1 Introduction The concept of sustainable green building has been pioneered in the late 19th century and continues to be adopted throughout the 21st century by more nations. Green building’s global trend or status in each country may be distinguished by Green Building Rating Tools; also known as certification, utilized to assess and recognize buildings that meet certain green requirements or standards. From the Green Building Rating Tools, the number of green buildings which are qualified and have met such requirements and standards in each country is totaled and reflected. Globally, there is a variety of Green © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 234–242, 2023. https://doi.org/10.1007/978-981-19-3983-9_21 Green Building Construction Implementation Barriers 235 Building Rating Tools, although the major certifications such as LEED, BREEAM, Passive house, DGNB, and EDGE continue to be widely used [1]. Specifically, the LEED or Leadership in Energy and Environmental Design is the most recognized and utilized green building certification system among them. The countries: Mainland China, Canada, India, Brazil, and the Republic of Korea are the consistent countries and regions, outside the United States, ranked by LEED based on gross square meters of the green building projects [2]. In summary, the emergence of the green building movement brings benefits specifi- cally in the three pillars of sustainability: namely, environment, economic and social. Its adaptation and implementation are increasing in different countries and are currently a primary solution for mitigating negative impacts of the building and construction indus- try. However, wide-scale implementation of green buildings has not yet emerged globally, especially for developing countries, because of varying hindrances in each country, such as the degree of economic growth, geographical location, per capita capital, and many more. The research delved into the identification of the barriers of green building con- struction implementation in the Philippines based on the perception of possible building occupants and engineers and presented the results through a hierarchical model. Green Building in the Philippines. In the Philippine setting, green building practices were pioneered and have been slowly adopted throughout the years. Specifically, an organization named Philippine Green Building Council (PHILGBC) was established in 2007 to promote the building industry’s exchange of green building expertise to ensure a long-term sustainable environment. Parallel to LEED, a rating tool: Building for Ecolog- ically Responsive Design Excellence (BERDE), was also established by the organization alongside the Philippine government, to assess, measure, monitor, and certify the per- formance of local and national green building projects following environmental laws, regulations and mandatory standards [3]. However, BERDE holds very limited data on the projects they have certified, thus, it is better suggested to look up the green building projects which are LEED certified. According to LEED’s database, three hundred twenty-six (326) projects in the Philip- pines are registered for certification or certified. Two hundred thirty-two (232) of which are in the National Capital Region (NCR), which is the chosen scope of this research due to a large number of registered green building projects in the area. Out of the two hundred thirty-two (232) registered in LEED, only one hundred one (101) are certified. The remaining one hundred thirty (130) are not yet certified, some of which are still under construction or had been reviewed for certification, but their points did not reach the threshold to be certified. In addition, there are very few studies that investigate the causes for the slow development of green building in the Philippines. Regardless of the notable benefits for the environment and inhabitants, some aspects of green building construction make it less prevalent as the main method which hinders its steady progress. These so-called aspects may be referred to as barriers; serve as the hindrances preventing the implementation and acknowledgement of its potential as a better alternative to the conventional method of construction. Green Building Barriers. The review of related literature revealed the green building barriers that exist in the international setting. They are summarized and categorized in Table 1. 236 C. E. Monjardin et al. Table 1. Green building barriers and codes Code Barriers Social: awareness and education AE01 inadequate knowledge and information about green building concept [4] AE02 not enough research and case studies [4] AE03 Lack of building management role in supporting green building movement, [4] AE04 Lack of public awareness [4–6] AE05 Lack of database and information in GB [5, 6] Social: people and culture PC01 Burdensome implementation [4] PC02 lack of supportive atmospheres [4] PC03 Resistance to change [4] PC04 Negligence [4] PC05 Inconsistencies between formal regulations (e.g., existing federal, state and local legislation) and LEED [7, 8] PC06 Lack of government support [6, 9] PC07 Lack of interest from clients and market demand [5, 6] PC08 Lack of government incentives [5, 6] PC09 Lack of green building codes and regulation [5, 6] PC10 Lack of GB promotion by government [5] PC11 High degree of distrust of green building technology [5] Economic: financial F01 Higher costs for green building options than conventional ones. [4] F02 Risk of investment. [4, 6] F03 high cost of certification [4, 7–9] F04 Deficient financial support from the government and banks [4] F05 Scarcity of insurance solutions [7, 8] F06 Rental or resale value loss due to delay related to green construction procedures and conditions [7, 8] F07 Failure to use of financial incentives (tax/loan discounts, low financing rates) because of delays or lower certification levels than expected [7, 8] F08 High initial/investment cost [6, 9] F09 Long payback period [5, 9] F10 Higher cost of GBT products [5] F11 Lack of financing schemes [5, 6] (continued) Green Building Construction Implementation Barriers 237 Table 1. (continued) Code Barriers F12 High market prices and rental charges of green buildings resulting from green building technologies application [5, 6] Economic: management M01 Absence of further evaluation, [4] M02 Insufficient supervision from responsible parties, [4, 9] M03 Inadequate definition of project parties’ contractual roles and responsibilities, [7, 8] M04 Lack of importance attached to GBs by senior management, [5] M05 Lack of strategy to promote GB, [6] M06 Lack of design and construction team, [6] M07 Risks and uncertainties involved in adopting new technologies, [5] M08 Limited experience with the use of nontraditional procurement methods, [5] MO9 Resistance to change from the use of traditional construction methods, [5] M10 Conflict of interests among various stakeholders in adopting green building technologies, [5] M11 Adoption of green building technologies is time consuming and causes project delays, [5] Economic: technical T01 Lack of green construction experience and qualification, [5–8] T02 Contractors and subcontractors agreeing to standards that are not within their expertise and competence, [7, 8] T03 Lack of expertise in new products/technologies, [5–9] T04 Complex and rigid requirements involved in adopting green building technologies, [5] T05 Lack of local instituteUnavailability of local suppliers of GBTs and facilities for research and development (R&D) of GB, [5, 9] Economic: resource R01 Doubts about long-term viability and performance of new and untested products, materials and technologies, [7, 8] R02 Faulty performance of HVAC/electrical/plumbing systems and alternative water systems/alternative power generating equipment, [7, 8] R03 Failure to receive materials/products in a timely fashion causing delays, [7, 8] R04 Lack/unavailability of local suppliers of GBT products, [4, 5, 9] R05 Lack of green technologies, [6, 9] 238 C. E. Monjardin et al. 2 Methodology The methodology of this study is a quantitative research design based on the combined methodologies of three related studies by Wimala et al. [4], Chan et al. [5], and Dehdasht [10]. In Phase 1, the researchers collected information regarding green building and its rating system to have an understanding of the benefits green building construction offers to the society, environment, and the builder’s reputation. The information gathered is a huge component of the scope of the introduction of this paper and are greatly sourced from the websites of green building councils including WGBC, PHILGBC, BERDE, and LEED. Moving on to Phase 2, the researchers then identified the barriers of green building implementation existing in other countries such as Ghana, Malaysia, and Indonesia. The barriers identified are listed and categorized in Table 1 alongside their respective codes, which were used throughout the proceeding parts of this paper. In Phase 3, the preparation for the data gathering started in order to contextualize the found barriers in the Philippine setting. The barriers were summarized for the con- struction of the questionnaires. An initial questionnaire was made for the construction experts which contains 4-point Likert Scales to let them rate how strongly they agree or disagree if such a barrier exists or occurs in the country (4 = Strongly Agree, 3 = Agree, 2 = Disagree, 1 = Strongly Disagree). In Phase 4, the pilot testing of the questionnaire was conducted through Google Forms to 30 construction experts, which are engineers, architects, project managers, consultants, quantity surveyors and other professionals in the construction industry. The results of the pilot testing underwent a statistical analysis: Cronbach’s alpha, which will determine the reliability of the questionnaire. Once the questionnaire was considered reliable, the researchers proceeded to Phase 5 of this study after determining the sample size for both types of respondents. In Phase 5, the actual data gathering commenced with the qualified construction experts as respondents. The snowball sampling method was utilized via social media for the 50 participants. After which, the data gathered underwent the ANOVA Test to determine whether the categories have significant differences among the categories. The normalized values of the data gathered were also identified to determine which barriers are significant in the country and were included in the final questionnaire for the multi- unit residential occupants. In Phase 6, the questionnaire was then distributed in social media to reach 400 multi-unit residential occupants in NCR who are familiar about green building. 3 Results and Discussion Pilot Testing. The data from the pilot testing, which underwent Cronbach’s alpha, is equal to 0.869851. It obtained a value greater than 0.6 which interprets that the survey questions are internally reliable (or consistent). The last column is the value of Cron- bach’s alpha IF we delete that question. For each question, the Cronbach’s alpha is still greater than 0.6 thus, deleting any question to make the survey questionnaire reliable is not needed. Green Building Construction Implementation Barriers 239 Table 2. P-value of the categories according to Table 3. P-value of the categories accord- qualified experts ing to multi-unit residential building occu- pants Category P-value Category P-value Awareness and education 0.716485 People and culture 0 Awareness and education 0 Financial 0.000282 People and culture 0 Management 0.000007 Financial 0 Technical 0.426928 Management 0 Resource 0.070613 Technical 0.013755 Resource 0.000104 ANOVA Test. The ANOVA Test was done to determine the P-Values of the categories which dictates whether the barriers within that category have significant differences or none. An obtained P-value less than 0.05 signifies the difference is significant (Tables 2 and 3). Normalized Values. The critical barriers to the adoption of green building implemen- tation in the Philippines are defined as those with normalized values not less than 0.50. After normalizing the values, the results are as follows. Table 4. Significant barriers of each categories according to qualified experts Significant barriers AE PC F M T R AE03 PC05 F01 M01 T01 R01 AE05 PC06 F03 M02 T05 R04 PC08 F04 M03 PC10 F05 M04 F08 M05 F10 M07 F11 M08 F12 M09 M10 Table 4 shows what barriers are retained from the statistically analyzed data accumulated from the answers of the qualified experts in the Philippines. Table 5 shows what barriers are retained from the statistically analyzed data accu- mulated from the answers of the Multi-unit Residential Building Occupants in the Philippines. 240 C. E. Monjardin et al. Table 5. Significant barriers of each categories according to multi-unit residential building occupants Significant barriers AE PC F M T R AE04 PC02 F04 M02 T05 R04 PC03 F08 M03 R05 PC04 F11 M04 PC05 M05 PC06 M08 PC08 M09 PC010 M10 4 Conclusion The research concluded that even with the positive route of green buildings, barriers are still present to its implementation which are summarized and categorized in Table 1. Contextualizing the barriers and after implementing the ANOVA test, normalization, and Tukey HSD resulted in two (2) hierarchical models. The first hierarchical model shows the barriers according to construction experts. All of the barriers in the model are statistically identified as significant in the country. First, under the Social aspect; the barriers under the Awareness and Education category have no statistically significant difference, however, the ranking is as follows; (1) AE03, and (2) AE05. Under the People and Culture category, the barriers have a significant difference between them, and the ranking is as follows; (1) PC10, (2) PC08, (3) PC06, and (4) PC05. Next, under the Economic aspect; the barriers under the Financial category have a statistically significant difference, and the ranking is as follows; (1) F03, (2) F01 and F10, (3) F04 and F08, (4) F12, (5) F05, and (6) F11. Under the Management category, the barriers have a significant difference between them, and the ranking is as follows; (1) M05, (2) M09, (3) M08, (4) M04, (5) M10, (6) M01, (7) M07, (8) M02, and (9) M03. Under the Technical category, the barriers have no statistically significant difference, however, the ranking is as follows; (1) T05, and (2) T01. Lastly, under the Resource category, the barriers have no statistically significant difference, however, the ranking is as follows; (1) R04, and (2) R01. The second hierarchical model shows the barriers according to multi-unit residential occupants. Similar to Fig. 1, the barriers in the model are statistically identified as significant in the country. First, under the Social aspect; the barriers under the Awareness and Education category only one is significant, which is (1) AE04. Under the People and Culture category, the barriers have a significant difference between them, and the ranking is as follows; (1) PC10, (2) PC06, (3) PC08, (4) PC04, (5) PC03, (6) PC05 and (7) PC02. Next, under the Economic aspect; the barriers under the Financial category have a statistical significant difference, and the ranking is as follows; (1) F11, (2) F04, and (3) F08. Under the Management category, the barriers have a significant difference between them, and the ranking is follows; (1) M05, (2) M04, (3) M03, (4) M10, (5) M09, (6) Green Building Construction Implementation Barriers 241 M08, and (7) M02. Under the Technical category, there is only one significant barrier, which is (1) T05. Lastly, under the Resource category, the barriers have a significant difference, and their ranking is as follows; (1) R04, and (2) R05. In conclusion, there are barriers to green building construction implementation in the Philippines according to the perspectives of qualified construction experts and multi- unit residential occupants in NCR. The hierarchical models of the barriers provided an organized presentation, allowing future readers to easily identify according to the aspects of sustainability (social and economic), categories, and what is more significant among all barriers. Fig. 1. Hierarchical model of green building construction implementation barrier in the Philip- pines according to construction experts References 1. United Nations Environment Programme. Global Status Report for Buildings and Construc- tion: Towards a Zero-emission, Efficient and Resilient Buildings and Construction Sector. Nairobi (2020) 2. Stanley, S. U.S. green building council announces Top 10 countries and regions for LEED green building | U.S. green building council. Usgbc.org. https://www.usgbc.org/articles/ us-green-building-council-announces-top-10-countries-and-regions-leed-green-building. Accessed 04 Nov 2021 3. Culiao, R., Tae, S., Kim, R.: A review of the Philippine green building rating system, BERDE in comparison with G-SEED and LEED. Sbt-Durabi.org, 87–94 (2018) 4. Wimala, M., Akmalah, E., Sururi, M.R.: Breaking through the barriers to green building movement in Indonesia: Insights from building occupants. Energy Procedia 100, 469–474 (2016) 5. Chan, A.P., Darko, A., Olanipekun, A.O., Ameyaw, E.E.: Critical barriers to green building technologies adoption in developing countries: the case of Ghana. J. Clean. Prod. 172, 1067– 1079 (2018) 242 C. E. Monjardin et al. 6. Samari, M., Ghodrati, N., Esmaeilifar, R., Olfat, P., Shafiei, M.W.M.: The investigation of the barriers in developing green building in Malaysia. Mod. Appl. Sci. 7(2) (2013) 7. Gurgun, A.P., Arditi, D., Vilar, P.C.: Impacts of construction risks on costs in Leed-certified projects. J. Green Build. 11(4), 163–181 (2016) 8. Chiveralls, K., Zillante, G., Zuo, J., Wilson, L., Pullen, S.: Constructing corporate social responsibility: encouraging CSR through legislation and regulation. Paper presented at the COBRA 2011 - Proceedings of RICS Construction and Property Conference, pp. 1272–1284 (2011) 9. Shen, W., Tang, W., Siripanan, A., Lei, Z., Duffield, C., Hui, F.: Understanding the green technical capabilities and barriers to green buildings in developing countries: a case study of Thailand. Sustainability 10(10), 3585 (2018) 10. Dehdasht, G.: Barriers of sustainable lean construction implementation: hierarchical model (2018). https://doi.org/10.1007/978-3-030-02101-6_4 Urban Planning and Environmental Assessment Wind-Induced Dynamic Response Analysis of the Frame Lightning Rod Ran Liu1(B) , Li Song1,2 , Chenxing Cui1 , and Jinliang Liu1 1 School of Civil Engineering, Central South University, Changsha 410075, China

2 National Engineering Laboratory for High Speed Railway Construction, Changsha 410075, China Abstract. This paper simulated and analyzed the wind-induced dynamic response of a 500 kV substation’s typical frame lightning rod structure using the fluid-solid coupling numerical method. The analysis shows that the high-frequency vibration response of the upper members is significant, and it is greatly affected by multi- order vibration mode, the crosswind vortex-induced resonance is prone, which is unfavorable to the whole stress of the structure. The upper two sections of the lightning rod were optimized by applying spiral side plates and carried out the whole analysis. It is found that the maximum stress of the frame lightning rod occurred on the beam of the intersecting joint within the basic wind speed. With the increase of wind speed, the maximum stress position moves upward along the lightning rod and occurs at the bottom of the lightning rod sections. The structural optimization measures can effectively suppress the vortex-induced resonance of the lightning rod. The vortex-induced amplitude of the structure in the crosswind can be significantly reduced in the basic wind speed. So their joint action should be considered in the structural design and stress analysis. Keywords: Frame lighting rod · Scale model design · Structural optimization design · Wind-induced vibration · Fluid-solid coupling 1 Introduction The lightning rods are mostly installed on the frame to form a frame lightning rod [1]. The frame lightning rod structure is slender and long, determining that the wind load is its control load. With the improvement of social demand and technological progress, the lightning rod structure is gradually developing in a higher and lighter direction, which further enhances its sensitivity to wind load [2]. The frame lightning rod structure will sustain vibration under wind load and experience a large number of stress cycles during service, and its fatigue damage will be sustained accumulate and structural failure may occur [3]. In China, the collapse accidents of the frame lightning rod are not uncommon. As early as 1991, a 220 kV substation in Yuanjiang had a lightning rod breaking and collapse accident. Then, there were five collapse accidents of the 750 kV substation © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 245–257, 2023. https://doi.org/10.1007/978-981-19-3983-9_22 246 R. Liu et al. frame lightning rod from 2008 to 2015 [4–6]. The formal operation time of these broken substation frame lightning rods did not reach the structure’s design life and caused significant economic losses and social impacts. A 220 kV frame lightning rod of a 500 kV substation broke and collapsed in winter under a strong wind climate. The fracture position of the lightning rod occurred at the intersecting node between the bottom of the lightning rod and the beam [7], the fracture mode is torn from the beam’s cross-section. Therefore, it is essential to analyze the fracture cause, structural optimization design, and the wind-vibration response of the substation frame with lightning rods. After the accident, the department concerned conducted experimental research on the material chemical composition and mechanical properties of the broken part of the lightning rod, and the results met the requirements [6]. Chen Yiwen et al. [7] conducted modeling and simulation analysis on the structure and summarized the main reasons for the bottom fracture of the lightning rod: the code ignored the vortex-induced vibration of the frame lightning rod under wind load and its fatigue effect. The stress concentration of the steel tubular intersecting joint is prominent. The intersecting joint at the bottom of the lightning rod under the alternating wind load has an excellent fatigue impact, which causes the crack to expand continuously and leads to the lightning rod fracture accident. The plasticity and toughness of steel in the heat-affected zone decrease with repeated stress, and the fracture form changes from plastic fracture to brittle fracture. The research about the fracture causes of the frame lightning rod showed that wind load significantly impacts the fatigue of the lightning rod intersecting joint. Still, there are few studies on the quantitative analysis of its impact degree. In this paper, the fluid- solid coupling calculation method is used to simulate and analyze the wind-induced dynamic characteristics of the lightning rod and optimize the lightning rod structure. 2 Scale Model Design of the Frame Lightning Rod 2.1 The Whole Frame Lightning Rod Structure The research object of this paper is a 220 kV frame lightning rod of a 500 kV substation. The whole structure diagram is shown in Fig. 1. The material parameters are shown in Table 1. Studies had found that [7] there was apparent stress concentration at the intersecting joint of the middle-span (A-span) frame lightning rod under wind load. When the A-span lightning rod is in the whole structure, the variation law of stress and displacement at the connection points of each lightning rod section under different wind load conditions are the same as that in the single lightning rod so that the A-span lightning rod can be analyzed separately. The A-span single frame lightning rod is scaled according to the ratio of 1:10 to reduce the amount of calculation, and similar parameters are shown in Table 2. Wind-Induced Dynamic Response Analysis of the Frame Lightning Rod 247 Fig. 1. The schematic diagram of the whole frame lightning rod. Table 1. Material parameters [7] Steel model Elastic Poisson’s ratio Design strength Yield strength Mass density modulus Q235B 2.06 × 0.3 215 MPa 235 MPa 7850 kg/m3 105 Q345B 2.06 × 0.3 310 MPa 345 MPa 7850 kg/m3 105 Table 2. Similar parameters for the aeroelastic model design Similar parameters λL λM λU λEI λf Similarity ratio n n3 n0.5 n5 n−0.5 1:10 1:103 1: 3.16 1:105 3.16:1 2.2 Aeroelastic Model Design Due to the different bending stiffness of the bottom support of the A-span frame structure in each direction, the aeroelastic model adopts a rectangular steel tube structure. The calculation principle of the steel tube size is to keep the longitudinal and transverse flexibility ratio of the aeroelastic model consistent with the original structure, and the modal frequency meets the similarity ratio. The size is 20 mm × 40 mm × 1 mm, as shown in Fig. 2. There are five sections of the A-span lightning rod, the sizes from top to bottom are 48 × 3.5 × 1500, 89 × 5 × 2500, 140 × 4 × 2500, (325/220) × 6 × 4500, 325 × 10 × 4000, the diameter and length of the aeroelastic model circular tube can be calculated by the size similarity ratio λL , the thickness can be calculated according to the bending stiffness ratio λEI , the sizes of the aeroelastic model from top to bottom are 4.8 × 0.0858 × 150, 8.9 × 0.1284 × 250, 14 × 0.111 × 250, (32.5/22) × 0.1704 × 450, 32.5 × 0.2757 × 400. Because the thickness of each section of the 248 R. Liu et al. lightning rod is too small, it is not easy to converge in numerical calculation. So the thickness of each section is increased by three times. The material’s elastic modulus is reduced to 1/3 of the original to meet the bending stiffness scale ratio. After the size of the lightning rod aeroelastic model is determined, the additional counterweight required for each section is calculated according to the mass scale ratio, and the material density is adjusted to meet the mass scale ratio requirement. The size of the cross beam and lightning rod section of the prototype intersecting joint at the bottom of the lightning rod are 480 mm × 6 mm and 325 mm × 8 mm. To ensure the aeroelastic model’s intersecting joint and the prototype intersecting joint meet the scale ratio, the size of the cross beam is 48 mm × 0.207 mm. Fig. 2. The schematic diagram of the S-axis and W-axis direction of the aeroelastic model. 3 Fluid-Solid Coupling Analysis of the Aeroelastic Model The finite element analysis process of fluid-solid coupling of the frame lightning rod structure is shown in Fig. 3. The solution steps are as follows: (1) The finite element solid model and fluid domain model of the frame lightning rod are established by rhinoceros. Input the material parameters of each section of the frame lightning rod in the Engineering data module of the workbench. (2) Import the solid structure model into the Transient Structural module in the work- bench, then define the material parameters and steel pipe thickness of each section of the lightning rod, set the structural boundary conditions (fixed bottom surface of the support) and the solid coupling surface, and finally divide the structure mesh (see Fig. 4). (3) Import the fluid domain model into the HyperMesh software to divide the fluid domain 3D mesh, then import the mesh model into the fluid module in the work- bench, and set the boundary conditions and coupling surface of the fluid domain (see Fig. 5). (4) Set the fluid-solid coupling solution parameters through the SYTEM COUPLING solver, establish the fluid-solid coupling interface for data transmission of the fluid domain and solid structure, and use the spring smoothing method to update the data. In this paper, the weak coupling method is used to solve the fluid-solid coupling problem of the substation frame lightning rod, so the solution order is to calculate the fluid domain first and then the solid structure. Wind-Induced Dynamic Response Analysis of the Frame Lightning Rod 249 Start Establish the finite element Establish fluid domain model of the frame model and 3D mesh lightning rod (rhino) generation (HyperMesh) Import into the Transient Import the mesh model into Structural module in the Fluent module in Workbench Workbench Input the material Set the fluid-solid Define the boundary parameters of each coupling surfaces of conditions of the section of the frame solid structure and fluid domain (see lightning rod fluid domain Fig. 3) Engineering data respectively The data transmission of fluid-solid coupling interface is carried out through SYTEM COUPLING platform, and the data is updated by spring smoothing method In the weak coupling method, calculate the fluid domain first, and then calculate the solid structure End Fig. 3. The analysis process of the frame lightning rod structure 3.1 Finite Element Model of the Frame Lightning Rod and Fluid Domain In this paper, the finite element model of the frame lightning rod is established using the shell element, and the fluid domain model is established in the HyperMesh, as shown in Fig. 4–5. There are five boundary conditions in the fluid domain, including the inlet and outlet boundaries, the upper and lower boundaries, and the outer surface boundary of the lightning rod boundary. The fluid domain is defined based on the fifth section size of the lightning rod, the distances from its centroid to the inlet boundary and the upper boundary and lower boundary are the same, which are set to 10D, the distance from its centroid to the outlet boundary is set to 25D, and the height of the fluid domain is set to 2 m. The variation law of wind speed and turbulence characteristics (including turbulent kinetic energy and turbulent dissipation rate) at the inlet boundary with the height above the ground is programmed and input through UDF [8]. 250 R. Liu et al. Fig. 4. The aeroelastic model of the frame lightning rod. (a) Two-dimensional fluid domain (b) Fluid domain model Fig. 5. Fluid domain analysis model. The working environment of the frame lightning rod: basic wind pressure is ω0 = 0.35 kN/m2 , average altitude is z = 136.4 m, basic wind speed calculation formula: 2w0 2 × 0.35 v0 = = = 23.83 m/s (1) ρ 0.00125e−0.0001×136.4 where ρ = 0.00125e−0.0001z . The exponential law formula proposed by A.G. Davenport [9] for the average wind profile is widely used in civil engineering design. According to the geomorphic features of the substation, it belongs to the class B geomorphic area in China’s “Building Structure Load Code” [10]. The parameter value can refer to China’s load code [11]. The formula is as follows: U (z) = 23.83(z/10)0.15 , z < 350 m (2) 3.2 Model Validation The fluid-solid coupling calculation of the aeroelastic model is carried out. The position of the measurement point at the bottom of each section of the lightning rod is consistent with the test. The first two sections are 20 mm above the bottom joint, and the third section is 25 mm above the bottom joint. The last two sections are 40 mm above the bottom joint. When the wind speed is 5 m/s, and the wind angle is 0° (perpendicular to the S axis), Wind-Induced Dynamic Response Analysis of the Frame Lightning Rod 251 comparing the simulation calculation results with the test results, as shown in Table 3. It is found that the errors of W4 and W5 measuring points are 23.439% and 20.844%, and the other measuring points are less than 20%. Because of the test equipment errors and differences in the specifications of aeroelastic models, it is difficult to completely reproduce the test results in the numerical simulation, so the error is acceptable. Table 3. Mean square value of measuring point strain under 0° wind angle and 5 m/s wind speed (με). Test value Analog value Error (%) Test value Analog value Error (%) [12] [12] S1 3.394 2.993 19.401 W1 6.141 5.752 6.763 S2 3.133 2.857 14.81 W2 8.445 7.291 15.828 S3 3.232 3.613 2.278 W3 10.15 9.68 4.855 S4 5.602 4.703 15.67 W4 10.717 8.682 23.439 S5 11.321 9.274 17.525 W5 8.273 6.846 20.844 4 Wind-Induced Dynamic Response Analysis of the Frame Lightning Rod 4.1 Dynamic Response Analysis of the Aeroelastic Model At the local height of 10 m, the basic wind speed with a recurrence period of 50 years is 23.83 m/s, which corresponds to the wind speed of the aeroelastic model is 7.5 m/s. The strain time history of each measuring point of the prototype frame lightning rod under the basic wind speed is calculated using the fluid-solid coupling simulation method. And the strain power spectrum can be obtained by Fourier transform, as shown in Fig. 6. It can be seen that the vibration form of the first two sections of the lightning rod is the most complicated, and their frequency response includes the first four frequencies. The third section consists of the first, second, and fourth frequency components, and the last two sections are mainly first-order frequencies. Therefore, the high-frequency response of the lightning rod structure is primarily reflected in the upper members of the structure, and their vibration is jointly controlled by multi-order vibration modes, which is unfavorable to the structure. 252 R. Liu et al. 60 40 36 23.83m/s,W-2 (0°) 23.83m/s,W-3 (0°) 23.83m/s,W-1 (0°) 35 34 55 30 32 50 25 30 28 45 Strain (με) Strain(με) 20 Strain (με) 15 26 40 24 10 35 22 5 20 30 0 18 -5 25 0 5 10 15 20 0 5 10 15 20 0 5 10 15 20 Time (s) Time (s) Time (s) 1.2 23.83 m/s, W-1 (0°) 23.83 m/s, W-2 (0°) X:61.04 1.0 X:7.56 23.83 m/s, W-3 (0°) 2.0 X:61.04 Y:1.154 Y:0.95 Y:1.154 Strain power spectral density Strain power spectral density 1.0 Strain power spectral density 0.8 1.5 0.8 0.6 0.6 1.0 0.4 0.4 X:7.56 0.5 Y:0.20 0.2 X:17.94 0.2 X:39.06 X:61.04 X:17.94 Y:0.22 X:39.06 X:61.04 Y:0.19 X:61.16 Y:1.154 Y:0.08 Y:0.11 Y:1.154 Y:0.05 0.0 0.0 0.0 0 20 40 60 80 0 20 40 60 80 0 20 40 60 80 Frequency (Hz) Frequency (Hz) Frequency (Hz) Fig. 6. The fixed-point strain time history and the strain power spectrum of the lightning rod section before optimization (v = 23.83 m/s). 4.2 The Frame Lightning Rod Aeroelastic Model Optimization Design Through the wind vibration response analysis of the prototype frame lightning rod aeroe- lastic model, the vibration forms of the first and second lightning rod sections are more complicated, and vortex-excited resonance in the crosswind is prone to occur, which is unfavorable to the structure response. Therefore, the structural optimization measure adopted in this paper is to attach spiral side plates on the surface of the lightning rod to suppress the occurrence of vortex-excited resonance. The factors that affect the suppres- sion efficiency of the spiral side plate include the shape and the coverage rate of the side plate [13]. In this paper, the size of the spiral side plate is selected: the number of screw heads is 3, the screw pitch is 5D, the screw height is 0.25D, and the coverage rate is 50% (D is the outer diameter of the lightning rod tube). The optimized frame lightning rod finite element model is shown in Fig. 7. Fig. 7. Finite element scale model of the optimized frame lightning rod. Wind-Induced Dynamic Response Analysis of the Frame Lightning Rod 253 4.3 The Whole Analysis of the Optimized Frame Lightning Rod Aeroelastic Model Figure 8 shows the stress cloud picture of the aeroelastic model of the frame lightning rod before and after optimization under the wind speed of 5 m/s at the wind angle of 0°. It can be seen from the figure that the maximum stress of the aeroelastic model of the frame lightning rod before and after optimization occurs at the bottom of the second section of the lightning rod and on the beam respectively, the maximum stress is 94.547 Mpa and 28.043 Mpa. According to the similarity ratio between the simulated wind speed, and the actual wind speed and Eq. (3), the maximum stress of the frame lightning rod before and after optimization under 15.812 m/s actual wind speed can be calculated to be 283.641 Mpa and 84.129 Mpa, which does not exceed the yield strength of the material. The stress conversion relationship between the aeroelastic model and prototype structure: Ep εp Ep σp = σm = σm = 3σm (3) Em εm Em σp is the actual stress of the middle-span lightning rod in the whole prototype structure, and σm is the stress of the aeroelastic model. (a) The aeroelastic model before optimization (b) The optimized aeroelastic model Fig. 8. The whole stress cloud diagram of the aeroelastic model at the 20 s under 5 m/s wind speed. Analyze the influence of wind speed on the overall stress of the optimized frame lightning rod, the maximum stress, and its location under 3.162 m/s−31.623 m/s wind speeds and 0° wind angle are listed in table 4. Through comparison, it is found that the maximum structural stress increases with the increase of wind speed. When the wind speed is 31.623 m/s, the maximum stress value is 550.860 Mpa, which has exceeded the yield strength of the material. Within the basic wind speed, the maximum stress occurs on the beam. With the increase of wind speed, the location of the maximum stress moves upward along the frame lightning rod. For each lightning rod section, the most unfavorable position of stress occurs at the bottom of the rod. It shows that within the basic wind speed, the continuous cyclic vibration of the structure has the most significant 254 R. Liu et al. influence on the cumulative fatigue damage of the intersecting joint. The upper sections of the lightning rod are more sensitive to the increase in wind speed. Table 4. Maximum structural stress under each wind speed and 0° wind angle. Wind speed (m/s) Maximum Mises stress (Mpa) Maximum stress position 3.162 3.810 Beam 6.325 20.807 Beam 9.487 40.857 Beam 12.649 44.211 Beam 15.812 84.129 Beam 18.974 98.010 Beam 22.136 132.162 Beam 25.298 226.377 The bottom of the third section of the lightning rod 28.461 227.688 The bottom of the third section of the lightning rod 31.623 550.860 The bottom of the first section of the lightning rod 4.4 Analysis of the Frame Lightning Rod Aeroelastic Model Joints Since each lightning rod section’s most unfavorable stress position occurs at the bottom, the following mainly carried out joints analysis for the bottom of each lightning rod section. In this paper, the measuring points of the first four sections of the lightning rod aeroelastic model are arranged at 20 mm above the bottom of the lightning rod section, the measurement point of the fifth section of the lightning rod aeroelastic model is set at 40 mm above the centroid of the beam. At the height of the measuring point of each lightning rod section, two measuring points are set along the strong axis (S axis) and the weak axis (W axis). Figure 9 shows the comparison results of fixed-point strain time history and the strain power spectrum of each lightning rod section before and after optimization under 0° wind angle and 23.83 m/s basic wind speed. It is found that under the basic wind speed, the spiral side plate can significantly reduce the vibration amplitude of the fixed- point strain of the lightning rod. The fixed-point strain response of the second section is increased, and the other lightning rod sections are slightly reduced, indicating that the spiral side plate increases the downwind resistance of the second section of the lightning rod. Each lightning rod section is mainly dominated by the first-order frequency, it can be found that the spiral side plates reduce the frequency components in the frequency response of the upper sections of the frame lightning rod, which can effectively avoid high-frequency response. Wind-Induced Dynamic Response Analysis of the Frame Lightning Rod 255 50 0°,W1 (original structure) 0°,W2 (original structure) 0°,W3 (original structure) 45 60 0°,W1 (optimized structure) 0°,W2 (optimized structure) 60 0°,W3 (optimized structure) 40 35 50 30 25 50 Strain (με) Strain (με ) 20 40 Strain (με ) 15 10 40 30 5 0 -5 20 30 -10 -15 10 0 5 10 15 20 0 5 10 15 20 0 5 10 15 20 Time (s) Time (s) Time (s) 2.5 23.83 m/s, W-1 (0°) 0.8 23.83 m/s, W-2 (0°) 0.40 23.83 m/s, W-3 (0°) X:12.70 X:12.55 X:25.10 Y:2.24 0.7 Y:0.75 0.35 Y:0.38 2.0 Strain power spectral density Strain power spectral density Strain power spectral density 0.6 0.30 1.5 0.5 0.25 0.4 0.20 1.0 0.3 0.15 0.2 0.10 0.5 0.1 0.05 0.0 0.0 0.00 0 5 10 15 20 25 0 5 10 15 20 25 0 10 20 30 Frequency (Hz) Frequency (Hz) Frequency (Hz) Fig. 9. The fixed-point strain time history and the strain power spectrum of the lightning rod section before and after optimization (v = 23.83 m/s) Figure 10 shows the comparison results of each fixed-point strain response of cross- wind at 0° wind angle before and after structural optimization. Before the structural optimization, the second and fourth sections of the lightning rod structure have vortex- excited resonance at about 21 m/s wind speed. The third and fifth sections of the lightning rod structure have vortex-excited resonance at about 24 m/s wind speed. And the first section of the lightning rod has obvious vortex-excited resonance around the two wind speeds. There is no obvious vortex-excited resonance in each lightning rod section of the optimized structure, indicating that the structural optimization measure effectively suppressed the vortex-excited resonance of the structure. 60 120 Maximum absolute value of strain(με) 0°,S1 0°,S1 Maximum absolute value of strain(με) 0°,S2 0°,S2 100 0°,S3 0°,S3 0°,S4 40 0°,S4 80 0°,S5 0°,S5 60 20 40 20 0 0 0 5 10 15 20 25 30 0 5 10 15 20 25 30 35 Wind speed(m/s) Wind speed(m/s) (a) The aeroelastic model before optimization (b) The optimized aeroelastic model Fig. 10. The fixed-point crosswind strain responses of the aeroelastic model at 0° wind angle 256 R. Liu et al. 5 Conclusion In this paper, taking a 500 kV substation 220 kV frame lightning rod intersecting joint fracture accident as the engineering background, the optimization design is carried out of the frame lightning rod and analyzes its wind-induced vibration response based on the fluid-structure coupling numerical method. The main conclusions are as follows: (1) The whole analysis of the optimized aeroelastic model found that the greater the wind speed, the greater the structure’s stress. Within the basic wind speed, the maximum stress of the frame lightning rod occurs on the beam of the intersecting joint, indicating that the continuous cyclic vibration of the structure has the most significant impact on the cumulative fatigue damage of the intersecting joint. With the increase of wind speed, the maximum stress position moves upward along the lightning rod. When the wind speed reaches 28.461m/s, the maximum stress of the first section of the lightning rod at a 0° wind angle exceeds the yield strength of the material, and the stress of the structure is unsafe. (2) The vortex-excited resonance of the structure can be effectively suppressed by adding spiral side plates to the upper sections of the lightning rod. It can significantly reduce the vortex-excited amplitude of the structure of crosswind within the basic wind speed. Still, it will hurt harm the crosswind vibration of the structure under higher wind speeds. References 1. Sun, Z., Yu, C.: On installing the overhead lightning rod on the top of the terminal tower of 35 kV power transmission lines toward transformer substations. J. Hefei Univ. Technol. 25(1), 127–130 (2002) 2. Wang, S.: Studies of wind-induced vibration and wind-induced fatigue on high-rise structures. Zhejiang University (2005) 3. Repetto, M.P., Solari, G.: Wind-induced fatigue collapse of real slender structures. Eng. Struct. 32(12), 3888–3898 (2010) 4. Huang, B., Zou, B.: Analysis of lightning rod falling accident in 220 kV Yuanjiang substation. East Chin. Electric Power 30(8), 127–130 (1992) 5. Xu, X., Wu, G., Li, Y.: Cause analysis and Countermeasures of Substation lightning rod fracture. Ningxia Electric Power, (c00), 94–99 (2010) 6. Sun, T., Li, Y., Wen, D.: The effect on karman vortex street to the mechanical strength of the lightning rod of steel tube. Insulators Surge Arresters, 01(003), 142–146 (2017) 7. Chen, Y.: Wind-induced dynamic response analysis of substation framework with lightning rod. Zhengzhou University (2017) 8. Sun, X.: Simulation of data exchange of multi-physics coupling surfaces. Comput. Simul. 32(1), 23–28 (2015) 9. Zhang, H.: Non-oscillating and non-free-parameter dissipative finite difference scheme. Acta Aerodynamica Sin. 6(2), 143–165 (1988) 10. Ministry of Construction of the People’s Republic of China.: GB50009–2012. Building Structure Load Code. China Construction Industry Press, Beijing (2012) 11. Zdravkovich, M.M.: Review and classification of various aerodynamic and hydrodynamic means for suppressing vortex shedding. J. Wind Eng. Ind. Aerodyn. 7(2), 145–189 (1981) Wind-Induced Dynamic Response Analysis of the Frame Lightning Rod 257 12. Yi, Y.: Bearing Capacity Test Study and Fatigue Life Analysis of Planar Joints of Lightning Rod in Substation. Zhengzhou University (2018) 13. Zheng, T.: Study on the mechanism of spiral side plate for suppressing vortex-Induced vibration of ocean riser. Jiangsu University of Science and Technology (2013) Research on Green Renewal Strategy Based on LID Adaptation Toolkit Selection--The Example of the North Bay Section of Macau Inner Harbour Terminal Neighborhood Linsheng Huang1,2(B) 1 Faculty of Humanities and Arts, Macau University of Science and Technology, Avenida Wai Long, Taipa, Macau, China

2 Faculty of Civil Engineering, Putian University, Xueyuan Middle Street, Fujian, China Abstract. This paper focuses on the application of low impact development (LID) strategy in the adaptation of urban built environment and nature, pointing out that the sustainable regeneration of marginal historical areas should be highlighted in the practice of urban development from incremental large-scale development to stock quality improvement. This paper focuses on the specific problems faced by the gradually depolarized and fragmented historic space of the North Bay section of Macau’s Inner Harbour Terminal, with the operational model and technical app- roach of low-impact development, combining the summary and cross-examination of existing engineering cases, to construct an adaptive strategic toolkit for regional regeneration. The green regeneration for the North Bay section of the inner harbor pier is used as a framework to analyze its role in solving flooding problems, opti- mizing the spatial interface, and bringing social benefits, as well as its contribution to the sound development of the historic space of the Macao city port. Keywords: Low impact development · Inner harbor pier · Technical strategy · Toolkit · Practical significance 1 Background 1.1 Low Impact Development (LID) Model LID, or Low Impact Development, was first proposed by the United States in the 1990s from the perspective of ecology [1], mainly focusing on the design and bioretention of stormwater management sites, and then gradually involving planning and design, technical measures and maintenance management. Low Impact Development (LID) can be considered as a sustainable model of urban regeneration based on decentralized, micro-renewal and small-scale control measures under cost control to enhance urban adaptation to nature. The LID adaptation toolkit is one of the tools to select the best matching regeneration strategy for different areas. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 258–267, 2023. https://doi.org/10.1007/978-981-19-3983-9_23 Research on Green Renewal Strategy Based on LID Adaptation Toolkit 259 Fig. 1. Shanghai Sponge city Fig. 2. U.S. Green street diagram. Reference: http:// practice. Reference: https://baijia www.yuanlin8.com/landscape/7430.html hao.baidu.com/s?id=160066997 8082459030) Fig. 3. North Bay section of inner harbor pier The focus of LID research is different in here and abroad. Domestic research focuses on urban road landscape planning, road stormwater landscape design, urban stormwater flood control and drainage management, and the selection and layout of road stormwa- ter facilities, such as the urban environmental management and mitigation of internal disasters in the sponge city renovation in China guided by the LID concept (e.g. Fig. 1); foreign research focuses on both stormwater landscape design and management. In this way, according to the actual situation of each region, the management manual of stormwater runoff has been introduced [2], such as the U.S. “Michigan Low Impact Development Manual”, and the Australian Water Sensitive Urban Design (WSUD) (e.g. Fig. 2), which are used to guide the specific practice. 1.2 The North Bay Section of Macau Inner Harbour Terminal The Macau Inner Harbor Pier area is an early site of human activities in Macau, which has accumulated deep urban memory and witnessed the historical transformation of Macau changing from a small fishing village before the opening of the port to a modern city. The 16 existing pier buildings from the 1940s to 1980s in the North Bay section of the inner harbor retain the early reclaimed land pattern in terms of bayside space, road system, and ramp space (e.g. Fig. 3), with strong cultural presence and deep memory of the place. And it is a rare historical and cultural landscape of the fishing port that integrates port history, neighborhood environment and architectural culture. From the perspective of the current problems the neighborhood environment, the lack of detailed planning has led to the disorderly increase and decrease in the construction and renovation of urban infrastructure in the North Bay section of the Inner Harbor Pier in recent years, and lack of control and implementation standards for the elements, in the pier architecture and landscape a greater threat. With the gradual implementation of 260 L. Huang the dual-carbon goal, the lack of green facilities has further made the North Bay section of the inner harbor pier gradually become a marginal space of the city of Macau. At the same time, the development of a strategic toolkit based on the specific mode of LID application for realistic needs has become an urgent issue for the urban historic space of Macao. 2 Analysis of Current Problems and Case Studies 2.1 Problem ANALYSIS Flooding Problem. In recent years, global warming, sea level rise, and extreme weather conditions have been occurring frequently, testing Macau’s urban drainage, water stor- age, and flood prevention capabilities. In terms of relative elevation, the elevation of the ground level of the inner harbor pier is (+1.3 m−2.3 m) MSL, which is lower than the storm surge height (+1.95–3.78 m) MSL during typhoons, and the relatively low terrain increases the probability of flooding (e.g. Fig. 4). According to statistics, the North Bay section of the Inner Harbor Pier suffers from flooding to varying degrees throughout the year (e.g. Fig. 5), with flooding lasting up to 8 h, which seriously affects the use of the North Bay section of the Pier and the production and living of the residents in the city. Fig. 4. List of historical flood heights and tide heights (storm) in the North Bay section of Inner harbour pier. Hard Impact of Waterfront Barges. Since 1831, the inner harbor has undergone many large-scale artificial reclamation. Due to the lack of technology and concept, the early waterfront areas of the inner harbor terminals were mainly built with hard barges (e.g. Fig. 6). The non-ecological shoreline formed by the hard barge broke the original natural shoreline pattern, destroyed the river ecosystem in the pier area [3], and completely changed the ecological appearance of the original mangrove wetlands along the inner harbor. This anti-ecological reclamation is one of the main reasons for the extensive flooding and ecosystem problems in the inner harbor. Research on Green Renewal Strategy Based on LID Adaptation Toolkit 261 Fig. 5. Flooding Fig. 6. Current state of hard barges Fig. 7. Street bottom interface and mid-dimensional interface status. Single Multi-level Street Interface. The streetscape of the North Bay section of the Inner Harbor Pier is similar to that of other traditional streets, showing more of a stan- dardized street interface pattern, i.e., the ground floor interface of the street is mainly oriented by two two-way traffic lanes, with a high percentage of impervious material surfaces. The North Bay section of the street is narrow due to historical reasons, and the space of the ramp is occupied by hawkers and stores, which makes it even more con- strained. Motorized vehicles, non-motorized vehicles, and pedestrians are not effectively separated, resulting in a low sense of safety. There is a serious lack of greenery in the streets, and the surface is overly hardened, resulting in stagnation and low infiltration and storage of rainwater, making it impossible for surface water to infiltrate into the ground or drain out in time during heavy rains and astronomical tides, causing street flooding. At the same time, the chaotic flower cages and monotonous columns are not conducive to the control and enhancement of the overall character of the street community in the middle dimension of the building façade (e.g. Fig. 7). Fading Memory of the Place. After the 1980s, most of the pier buildings in the area lacked effective maintenance and repair, and make with the real problems of aging of the building body, disorganized internal space, deteriorated external form, and loss 262 L. Huang of pier functions. The whole area gradually lost the support of the main industry and degenerated into a passive space of the city, face severing the daily connection with the citizens’ life [4], leisure, and work, and losing the sense of co-frequent development and public historical identity with the city. 2.2 Case Study The existing problems of flooding, barging, street interface and cultural recognition of the site in the North Bay section of the Inner Harbor Pier are one of the objective and prominent problems faced by the Port Terminal area. In order to respect the history and culture of the port terminal and to improve the environment of the area, the LID strategy is a better way to be chosen through the perspective analysis of several cases. Shanghai Yangpu Riverfront Demonstration Site Renovation. The Yangpu river- front demonstration site in Shanghai was previously a low-lying, waterlogged area, which was the first foreign yarn factory in Shanghai since the construction of Yangpu Road by the public concession in 1869. The LID approach is used to preserve the geo- morphic state, using rain gardens, plant configurations, bioretention and other means to store precipitation and stagnate rainwater, taking into account both technical and aesthetic interests [5], and the urban industrial culture is continued (e.g. Fig. 8a). Transformation of Singapore Barge Terminal. The Singapore Barge Terminal is part of the historic district of Singapore, which was the center of the city’s early commercial and transportation port. By preserving, repairing, and faithfully restoring the historic buildings in the barge area [6], upgrading and transforming the space of the site, and designing the barge treatment and intimate platform, the barge terminal site is preserved and strengthened as a whole, and the transformation of its function to accommodate commercial tourism is promoted. It creates a historical urban open space against the background of urban skyscrapers, preserving the memory of the place and deepening cultural identity (e.g. Fig. 8b). Green Streets in Vermont, USA. Based on the LID strategy, different cities in the U.S. established green street manuals for local conditions, Vermont published the Vermont Green Street Guide on the basis of the manuals, proposing five levels of street green renewal, i.e., Level 1 increases the amount of vegetation on both sides of the street and reduces impervious surfaces, and Level 2 increases the amount of vegetation on both sides of the street and reduces impervious surfaces. Level 2 is a large number of landscaped areas and a wide tree canopy to retain rainfall, Level 3 is the installation of stormwater landscaping facilities, Level 4 is a combination of Level 2 and 3 features proposed to be integrated with multiple access modes, and Level 5 is the management and renovation of the overall scope of the street. The effectiveness, means, cost, and complexity of the project increase step by step, and in view of the built environment and actual urban traffic conditions, the first, second, and third level of street renovation is generally the focus [7] (e.g. Fig. 8c). Research on Green Renewal Strategy Based on LID Adaptation Toolkit 263 a Shanghai b Singapore c U.S. Navy Pier Philadelphia Yangpu Riverfront Barge Terminal Update Fig. 8. Three case diagrams. The above cases all take into account the scale, status, capital and other constraints of the region, combine cultural, landscape, and facility elements with parallel character- istics, and respect regional attributes to promote regeneration practices. To summarize the specific application of different strategies of LID presented in the cases (e.g. Fig. 9), it is clear that they are functionally effective but also have limitations in use. For the North Bay section of the Inner Harbor Pier, the above case studies will serve as a useful toolkit to guide the selection of green regeneration strategies for the North Bay section in terms of development intensity, capital investment, regional space, and effectiveness. 3 Strategy Selection Basis and Toolkit Construction 3.1 Basis for Strategy Selection The choice of specific strategies for LID varies from region to region due to different urban problems, policy context, and ideology. To summarize the above case studies and actual situations, the main approaches and practices of LID include bioretention areas, cisterns and rain barrels, grass pond filter strips, green roofs (photovoltaic panels), site design, permeable paving, road and parking lot design, vegetated pockets, and infiltration wells. However, not every strategy is suitable for the regeneration of the North Bay section of the Macau Ferry Terminal, facing the main considerations being: First, The built-up area is relatively stable and has a traffic function. The area has been built for many years and can be spatially divided into a riverbank area, a road traffic area, and a pedestrian area of the rides, etc. The overall environment of the area is relatively stable and is not suitable for large-scale planning and renovation; Second The area has many historical elements and is not suitable for high-intensity development from the perspective of heritage conservation and utilization [8] (e.g. Fig. 9); Third, The space in the area is compact and there are no patches of unused vacant land. The area is spatially compact and there are no patches of idle vacant land, which makes it unsuitable for regeneration with high land use requirements; Fourth, Some of the fishery functions are still present in the area, so the needs of industrial development and the appropriateness of the strategy should be taken into consideration. In addition, given that the North Bay section is located in the marginal spatial area of the city of Macau, the current social concern is low and the inclination of capital is limited, so the high cost of development investment is not appropriate. 264 L. Huang 3.2 Construction of the LID Strategy Toolkit Based on the above case study, we further analyze the implementation effects of LID strategy in terms of functional role, economic cost and landscape appreciation, and based on the realistic needs and implementation limitations of the North Bay section of the Inner Harbor Pier, we establish a matching analysis table for the green regeneration strategy of the North Bay section under the principles of minimum intervention, minimum cost and maximum effect. The regeneration effect on the street area is strong and prominent, and the vegetation paved area is stronger and has average function ((e.g. Fig. 10), and this is used to construct the LID adaptation strategy toolkit for the North Bay section of the Inner Harbour Terminal. Fig. 9. List of historic buildings Fig. 10. LID strategy matching analysis Fig. 11. Green renewal strategy for the North Bay section of inner harbour terminal based on LID adaptation toolkit. Bioretention. The North Bay section of the Inner Harbor Terminal can be constructed as a bioretention area by combining the underutilized space formed by the spacing between Research on Green Renewal Strategy Based on LID Adaptation Toolkit 265 the terminal buildings and the stepped greenery to soften the hard embankment. Soil, plants and microorganisms can be used to treat stormwater, so as to alleviate surface runoff [9], and at the same time enhance the ornamental landscape of the waterfront. Low-Impact Site Design. Low-impact site design is a green regeneration strategy that has minimal impact on the historic streetscape and features of the North Bay section. From the bottom interface, as in the Yangpu River demonstration section and Vermont green streets, the integrated street site consideration, the unused riverfront walkway is simply transformed into a landscape grass pond planting surface or modular grass pond, the elevation is lowered to below the road surface, while both sides of the carriageway and the curb hard interface can be considered in combination with gravel paved drainage ditches to increase the drainage surface, thus using sunken grass pond, permeable curb interface, etc. Through scientific planning, the rainwater infiltrated by grass ponds and ditches will form a network through underground water collection structures or drainage systems to effectively mitigate surface runoff and reduce flooding caused by heavy rainfall. From the middle dimension, the vertical greening transformation can be carried out by using the skin of the new building façade and the vertical surface of the building such as the columns of the rider building, which are incompatible with the pier building, to intercept rainwater in multiple directions, reduce the probability of flooding and optimize the overall landscape interface of the street, weakening the huge contrast between the old and new buildings along the street surface. Permeable Paving Blocks. At present, the North Bay section of the inner harbor pier is paved with cement, whether it is for the riding building or the pedestrian walkway along the river. For example, in the old port of Marseille, France6 , from the city land- scape beautification and maintenance of the historical environment, the selection of local characteristics of permeable paving blocks to update the hardened walkway, forming a linear permeable area of the street, slowing down the speed of runoff into the under- ground drainage network and less likely to cause long-term accumulation of water on the ground [10], but also play a role in enhancing the historical landscape of the port. Vegetated Pockets. Given the spatial constraints of the North Bay section, for the exist- ing site, long and shallow vegetated pockets or vegetated ditches can be created along the riverwalk to mitigate surface runoff and facilitate stormwater infiltration. From these analysis, it can be seen that different strategic choices of LID adaptation toolkits will inevitably have different effects on the application targets. In light of the reality of the North Bay neighborhood, a flexible approach to building a strategic toolkit is undoubtedly the best guarantee for the neighborhood’s vitality (e.g. Fig. 11). 4 Practical Implications of LID Adaptation Toolkit 4.1 Enhance the social value of the neighborhood through green renewal The revitalization of the North Bay section of the Inner Harbour Pier must combine the protection of the historical environment with the protection of the landscape environment, 266 L. Huang through the selection and intervention of the LID adaptation toolkit strategy, the aquatic ecological system of the North Bay section of the neighborhood is complemented, from softening the hard barges, optimizing the spatial interface of roads and building facades, and reorganizing the ecological system and the industrial environment. By implanting the cultural elements of the pier, the enhancement of the street environment will definitely increase the interactive relationship with public life, reconstruct the ecological system and industrial landscape of the street, and strengthen the cultural memory of the street. 4.2 Green Renewal of Infrastructures to Promote Diversified Economic Models Based on the LID adaptive toolkit, the original monotonous pedestrian space is broken up and a sustainable cultural slow walking system such as a riverfront walkway and a pedestrian walkway with a riding building is constructed to maintain the diversity of landscape corridors. In order to restore the fishing port style of the North Bay section of the Inner Harbor Pier, upgrade and adjust the industrial attitude along the shoreline, and gradually form a sustainable urban waterfront leisure district with tourism and service industries as the core under the premise of environmental improvement, and integrate the culture of fishery and pier to create a waterfront leisure place for citizens and enhance the attractiveness of the district. Summary The LID adaptive toolkit strategy is not a unidirectional choice, but a multidimensional possibility. Based on the above analysis, it can be seen that the LID adaptation toolkit for the North Bay section of the Inner Harbour Pier, which is based on bioretention, low-impact site design, permeable paving blocks and vegetation pockets, can meet the current needs and desires of the neighborhood enhancement and achieve the goals of regional environmental, social and economic development, as well as provide a useful reference for the green regeneration practice of marginal urban historic spaces. References 1. Sun, Z.P.: Xiamen,s infrastructure based on the concept of “low impact concept” of retrofit sutdy, Anhui University of Construction (2021) 2. Zhi, X.f.: Research on public facilities design of urban waterfront space under the interaction design concept, Harbin Institute of Technology, pp. 8–11 (2016) 3. Deng, Y.: Renovation and adaptive-reuse for the waterfront historical industrial estate: analysis on the renewal along Singapore River. Mod. Urban Res. 8, 25–32 (2008) 4. Sun, M., Li, Z.Y.: Research on regeneration design strategy of waterfront industrial heritage in urban inner fringe case study on Philadelphia navy yard. Urban Architect. (8), 26–29 (2017) 5. Zhang, M., Zhang, J., Qin, S.: Media of landscape-four stepped narrations from the construc- tion of rainwater garden at Yangpu waterfront public space, Chin. Landscape Architect. (7), 49–54 (2021) 6. Liu, S.Y., Yang, F.: Adaptive rouse of historic buildings/neighborhoods in the mind frame of sustainable development. Hist. Building Conserv. 2, 4–11 (2009) 7. Fan, W.: Study on rainwater landscape facility design of green street—a case of Vermont in the USA. Urbanism Architect. 18(17), 148–150 (2021) Research on Green Renewal Strategy Based on LID Adaptation Toolkit 267 8. Lou, J.K., Lin, A.L.: Research on street green space landscape preference based on narrative preference method proceedings of the annual china urban planning conference, pp. 26–29 (2021) 9. Wang, Y.C.: Research on the prevention and control of waterlogging in soutuern china based on low impact development model. Water Conservancy Sci. Technol. Econ. 08(4), 17–20 (2021) 10. Wang, X.H.: Study on the Renewal Planning of Dalian Port Area Under the Perspective of Historical Heritag, Harbin Institute of Technology (2019) The Impact of Urbanisation on Catchment Discharge in Malaysia: A Case Study on Sungai Selangor Catchment Mayuran Jayatharan(B) , Andreas Aditya Hermawan , Amin Talei , and Izni Zahidi Discipline of Civil Engineering, School of Engineering, Monash University Malaysia, 47500 Bandar Sunway, Selangor Darul Ehsan, Malaysia

Abstract. Malaysia is one of the countries that experience a tropical climate with an abundant amount of rainfall. However, this large amount of rainfall paired up with rampant development has increased the severity and frequency of floods occurring in the nation. This research investigates the utilisation of a hydrological model to assess the effects of urbanisation on the river flow regime. The software used to develop the hydrological model for this research is the MIKE 11 Nedbør- Afstrømnings-Model or better known as MIKE 11 NAM. Urban development is captured by developing runoff coefficients with land use and land cover (LULC) data. The applicability of Landsat-8 satellite images in determining LULC is also explored within this research. In addition, future development scenarios are devel- oped based on the rate of urbanisation within the Sungai Selangor catchment. The results show that the MIKE 11 NAM software captured the river flow regime of the study site relatively well, with overall performance criteria score for the coefficient of determination, r2 = 0.7 and Nash- Sutcliffe efficiency, CE = 0.6. Landsat-8 satellite images can estimate the rate of urbanisation within the study area. The runoff coefficients developed from the LULC data improved the overall performance criteria score. Lastly, the increase of developed areas in the expanse of forest area caused a sharp increase in the value of the runoff coefficient. Keywords: Urbanisation · Hydrological modelling · Tropical flood · MIKE 11 NAM; Land use · Landsat-8 1 Introduction The rapid rate of urbanisation has caused an increase in the severity and frequency of river floods occurring globally. In addition, the increase in population density and rampant development in flood-prone areas has intensified the flood tolls. Thus, it is essential to assess the effects of urban development on flood severity. Studies in the past suggest that the type of land use and land cover (LULC) significantly influences flood peaks and flood volume [1, 2]. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 268–280, 2023. https://doi.org/10.1007/978-981-19-3983-9_24 The Impact of Urbanisation on Catchment Discharge 269 The increase in developed areas tends to lead towards higher impervious surfaces. These impervious surfaces contribute to larger discharge peaks and volume due to rainfall being unable to infiltrate the soil. Thus, the quantification of impervious land becomes vital in determining the influences of LULC on urban development. Based on the paper done by Chabaeva, Civco [3], there are three main approaches for quantifying the amount of impervious cover. These methods consist of (1) interpretive detection, (2) spectral pattern recognition, and (3) statistical and mathematical modelling. Among these three methods, interpretive detection produces the most accurate results, although at the same time also the most time-consuming. In recent years, remote sensing techniques have been popular for researchers to quantify and classify land use within an area of interest (AOI). Landsat images have been at the forefront as the primary data source for reliable remote-sensing images for LULC analysis [4–7]. Landsat images are moderate resolutions satellite images provided by the United States Geological Survey (USGS), freely accessible to the public domain. Landsat-8 is the latest satellite launched into space in 2013 and takes approximately 725 images per day. Users can access these images as soon as 8 h after acquisition [8]. Therefore, Landsat images are an ideal choice for researchers to determine long-term land-use changes within a particular AOI. The current or past status of LULC is important for hydrologists as it allows the determination of runoff coefficient within an AOI. Runoff coefficient, in simple terms, can be defined as the ratio of runoff to rainfall [9]. It is predominantly used as a parameter in hydrological modelling, where it represents the amount of runoff generation within a catchment area. In the past, the runoff coefficient has been used to assess the influence of land-use types on runoff generation. It was found that land use type had a significant influence on runoff coefficient compared to other factors such as topography, vegetation cover, soil type, and moisture content of the soil [10–12]. Runoff coefficient is an essential parameter within hydrological models [13]. In the past, researchers have tried to establish a proper methodology in assessing the effects of LULC on river flood regimes. The paper done by Salvadore, Bronders [14] states that there is currently no universal methodology for accurately simulating the urban water system. However, the article suggests that remote sensing data and measur- able parameters within hydrological models might be the step to reduce uncertainties. Thus, this research aims to assess the relationship between urbanisation and river flood discharge by (1) investigating the applicability and reliability of utilising Landsat Images and a hydrological model in determining runoff coefficient, (2) assessing the effects of urbanisation towards river flood discharge in future development plans. 2 Study Site and Data Used The study area chosen for this research is the Sungai Selangor River basin situated within the west coast of Peninsular Malaysia in the state of Selangor. This area has a relatively humid climate with slight variation in temperature throughout the year [15]. The catchment area of the river basin is 1955 km2 and length of the main river is 110 km with several tributaries. The main river experiences an average discharge of 57 m3 /s 270 M. Jayatharan et al. throughout the year with seasonal rainfall variations causing the flow to vary between 23 m3 /s and 122 m3 /s [16]. Historical records suggest that occasional overspills from the Northeast monsoon cause large-scale heavy rainstorms in December with a mean monthly rainfall ranging between 200 mm to 260 mm. The dry months from January to February alongside June to August have a mean monthly rainfall range from 120 mm to 150 mm [16]. The LULC within the study area is obtained from the Department of Town and Coun- try Planning Malaysia (JPBD). The upstream region of the basin consists of a mountain- ous area with forested foothills. The middle region comprises agricultural plantations comprising rubber trees and oil palm. The downstream region consists mainly of low- lying peat swamp forests. Furthermore, developed areas make up 47% of the study area. The developed areas consist of industrial, commercial, residential, and general utilities infrastructures. The remaining 53% consists of water bodies, reserved forest area, and bare land. In addition, a total of 19 telemetric rainfall stations within the catchment area were considered for this research. Hydrological data of 15-min resolution is provided by each station and was obtained from the Hydrology Unit of the Department of Irrigation and Drainage Malaysia (DID). The data range considered is from the year 2013–2018. 3 Methodology 3.1 Development of the Hydrological Model The hydrological model was developed for the downstream streamflow station of the study area by using the MIKE 11 NAM software. The MIKE 11 NAM model is a conceptual model developed by the Danish Hydraulic Institute [17]. An event-based rainfall-runoff (R-R) modelling approach was adopted to tackle the several data gaps within the dataset used. Flood events were used for calibrating and validating the hydro- logical model.. The total catchment area for the downstream flow station is 1186 km2 , shown in Fig. 1. A total of 55 rainfall-runoff events were considered to develop the hydrological model, where 45 events were considered for calibration and 10 for valida- tion. This research adopts several performance criteria to assess the model performance, including the coefficient of determination (r2 ), Nash – Sutcliffe coefficient of efficiency (CE) and relative peak error (RPE). Table 1 below summarises the criteria scores used to evaluate the flood events based on past research in assessing hydrological models [18–20]. The Impact of Urbanisation on Catchment Discharge 271 3.2 LULC Classification Landsat-8 images were used to determine and classify the land use within the catchment area. Landsat-8 images were obtained from the USGS earth explorer website (https:// earthexplorer.usgs.gov/), with a total of 12 images spanning from 2015 to 2021. Landsat images with a cloud cover of below 30% were considered before further inspection was conducted for land use classification. The classification was done using ArcGIS software, and the Supervised classification method was adopted. Land use types were classified into four main types: developed area, water bodies, forest area, and agricultural land. An accuracy assessment was done to each Landsat-8 image considered for the classification. This assessment was done by reusing the chosen Landsat-8 image as ground truth survey points were not available for this research. The results obtained from the land use classification done with the Landsat-8 images were then compared with official land use data obtained from JPBD. 3.3 Development and Validation of Runoff Coefficient The areas obtained from the official land use data were used to develop the runoff coeffi- cient for the catchment area. The runoff coefficient was generated using the impervious coefficients within the local stormwater management guideline, MSMA [21]. Flood events were used to validate the runoff coefficient derived from the local guideline. The model parameter that represents imperviousness was tweaked to assess its performance. The flood hydrograph generated using the runoff coefficient derived from the official land use data were compared with the simulated hydrograph obtained from the validated hydrological model. The performance criteria values obtained from the simulated hydro- graphs determine the capabilities of utilising land use data to derive runoff coefficients as input for hydrological models. Table 1. Performance rating table [18–20] Model performance criteria Performance rating Poor Fair Fairly good Good Excellent Coefficient of determination (r 2 ) <0.5 >0.5 >0.6 >0.75 >0.9 Nash-sutcliffe coefficient of efficiency (CE) <0.5 >0.5 >0.65 >0.7 >0.8 Relative peak error (RPE) >0.7 <0.6 <0.5 <0.3 <0.1 272 M. Jayatharan et al. Fig. 1. Catchment area considered for hydrological modelling 3.4 Development of Future Scenarios The developed runoff coefficients were then used to assess future development scenarios within the catchment area. Two future scenarios were developed to be compared with the government plans for the AOI in 2035 called the “Rancangan Struktur Negeri 2035” or RSN 2035 [22]. In this research, two land-use scenarios were developed to compare with the RSN 2035. In Scenario (1), trendlines were projected to 2035 based on the official land use data. Scenario (2) is done using the values obtained from Scenario (1) but assuming that the forest area is fully cleared to make way for the developed area. The results obtained from this research and the findings of the government policies were assessed and discussed. 4 Results and Discussion 4.1 Calibration of the MIKE 11 NAM Model The calibration was done using the auto-calibration feature within the software. The range of parameters adopted for calibration and the final averaged calibrated parameters is shown in Table 2. Figure 2 shows the box plot for all the model performance criteria values obtained for the calibration of the hydrological model. The median value for r2 and CE is within the excellent range, whereas RPE is within the good range. There is only a single outlier for r2 and CE, respectively. The Impact of Urbanisation on Catchment Discharge 273 Table 2. MIKE 11 NAM calibration range and averaged calibrated parameters (Range adopted from Doulgeris, Georgiou [23]). Parameter definition Abbreviation Lower bound Upper bound Averaged parameters Maximum water content Umax (mm) 10 20 11.14 in surface storage Maximum water content Lmax (mm) 50 300 173.46 in root zone storage Overland flow runoff CQOF (-) 0.1 1 0.40 coefficient Time constant for CKIF (h) 500 1000 914.81 routing interflow Time constant for CK1 (h) 3 48 24.25 routing overland flow Root zone threshold TOF (-) 0 0.7 0.35 value for overland flow Root zone threshold TIF (-) 0 0.7 0.32 value for interflow Time constant for CK2 (h) 3 48 14.58 routing overland flow Root zone threshold TG (-) 0 0.7 0.40 value for GW recharge Time constant for CKBF (h) 1000 4000 1347.42 routing baseflow Lower baseflow, Cqlow (-) 0 100 86.68 recharge to lower reservoir Time constant for Cklow (hour) 1000 30000 24405.67 routing lower baseflow Baseflow BF (m3 /s) Manual calibration 43.44 Fig. 2. Box plots of the performance measures for the calibration stage. The dot points are outliers. 274 M. Jayatharan et al. 4.2 Validation of the MIKE 11 NAM Model Figure 3 shows the box plot for the model performance for the validation of the hydro- logical model. The median value for r2 and RPE is within the good range, whereas CE is within the fairly good range (Refer to Table 1). Thus, it is deemed that the calibra- tion of the hydrological model was successful in capturing the river flow regime of the catchment area. Fig. 3. Box plots of performance measures for the validation stage. 4.3 Landsat Image Classification The accuracy assessment for this research considers the kappa coefficient as an indicator to determine the reliability of the classification. A total of 60 ground truth survey points were considered for each Landsat satellite image. The average value obtained for the Kappa coefficient is 0.50, which is regarded as moderate by standards set by Cohen [24]. This could be primarily due to human error upon manual inspection of the satellite image. This is verified by the low user’s accuracy for the classification of the various type of land use. Figure 4 shows the area comparison between Landsat-8 satellite images and official use data. In terms of agriculture and developed areas, the values obtained from the satellite images and official data have an average percentage difference of 12%. The difference in developed areas is caused mainly by the clouded areas getting misclassified into developed land. However, there is a large discrepancy when examining forest areas and water bodies classification. The average percentage difference for forest area and water bodies is 20% and 90%, respectively. It was found that most of the cloud cover was located in the mountainous forest areas within the study site. This could have caused the increase in developed areas and the decrease of forest area in the Supervised classification process. Cloud cover plays a major role in hindering the performance of satellite images classification [25, 26].In addition, the colour spectrum between agriculture and forest land could have been another factor in this. The oil palm plantation areas and forest land seem to overlap in certain classification areas. Miettinen, Gaveau [27] and Oon, Mohd Shafri [28] reported similar discrepancies when classifying oil palm plantations and tropical forest areas within Malaysia. These factors could have caused the difference in values for the forest cover. The Impact of Urbanisation on Catchment Discharge 275 There is a significant fluctuation over the years for the area of the water bodies. This could be due to the satellite images taken during the monsoon seasons, which sees some areas being inundated due to floods. Furthermore, there are areas where aquaculture or fish farming is practised within the study site. This practice could cause some parts of the satellite image to be classified as water bodies as farmers could periodically empty the tanks. Water bodies Landsat Developed land Landsat Official 400 Official 40 -28.4% -151.0% -7.7% 300 0.2% Area (km2) 30 -44.9% -72.7% Area (km2) 200 20 10 100 0 0 2016 2018 2020 2016 2018 2020 Forest Landsat Agriculture Landsat Official Official 600 12.8% 13.7% 600 -23.92% 32.0% -13.07% 2.59% Area (km2) Area (km2) 400 400 200 200 0 0 2016 2018 2020 2016 2018 2020 Fig. 4. Comparison between land use classification of Landsat-8 satellite images and official land use data Based on the results obtained, it can be said that there is a potential of utilising Landsat-8 images in determining LULC within a study area. Landsat-8 satellite images could provide an initial estimate of the types of land use present within an AOI, especially in the developed area, agricultural land, and forest area. However, it is highly dependent on the availability of satellite images within the USGS database and the percentage of cloud cover present within an AOI. 4.4 Runoff Coefficient Figure 5 shows the obtained runoff coefficients based on the MSMA guidelines. There is a steady inclination of the runoff coefficient over the years. This is due to the increase of developed areas and the decline of the forest area, which causes an increase in the imperviousness area within the catchment. The model parameter CQOF is identified as the model representation of the impervious factor. The runoff coefficient was then plugged into the model parameter CQOF whilst maintaining the other calibrated model 276 M. Jayatharan et al. parameters. The simulation results are shown in Fig. 6 and Table 3. It can be seen that there is an improvement in the hydrographs in terms of simulating peak discharge and the overall profile of the hydrograph. There is a vast improvement in terms of r2 , CE, and RPE when comparing the simulated hydrograph from the calibrated parameters and the hydrograph, which utilises the runoff coefficients developed from the official land use data. Official RC values 0.80 0.77 RSN 2035 RC value Runoff Coefficient (RC) 0.75 Scenario (1) RC value Scenario (2) RC value 0.70 0.63 0.65 0.60 0.61 0.55 0.50 2015 2020 2025 2030 2035 Year Fig. 5. Runoff coefficients developed from the official land use data and the projected coefficients for Scenario (1), Scenario (2), and RSN 2035. (a) Event 1 Observed (b) Event 2 Observed 200 Simulated 250 Simulated Runoff Coefficient Runoff Coefficient 200 Discharge (m3/s) Discharge (m3/s) 150 150 100 100 50 50 0 0 0 5000 10000 15000 0 5000 10000 15000 20000 Time (min) Time (min) Fig. 6. (a) and (b): Hydrograph events used to assess the model performance criteria when utilising the runoff coefficient obtained from the MSMA guidelines The Impact of Urbanisation on Catchment Discharge 277 Table 3. (a) and (b): Parameter alterations and model performance based on runoff coefficients derived from MSMA guidelines Model impervious parameter (CQOF) Year Event Name Simulated Runoff coefficient 2016 Event 1 0.4 0.55 2018 Event 2 0.4 0.56 Model performance criteria Simulated Runoff coefficient Event name r2 CE RPE r2 CE RPE Event 1 0.74 0.71 0.22 0.76 0.61 0.22 Event 2 0.78 0.69 0.31 0.76 0.71 0.16 4.5 Development of Future Scenarios The projected land-use area for both scenarios is shown in Table 4. In Scenario (1), trendlines were projected to 2035 based on the official land use data. Figure 7 shows the trendlines developed for Scenario (1). A power relationship was considered for the agriculture and water bodies. Water bodies such as lakes and wetlands generally do not decline over time and maintain over the years. Furthermore, there is a clear intent by the government to keep the existing agricultural land within the catchment area as the place is considered the rice bowl for the Selangor state. Developed land sees a big jump from 2019 to 2020, and thus an exponential relationship was considered. The forest area shows an evident linear decline in terms of the forest area despite the government reports saying there is an intent to maintain the reserved forest areas. Thus, a linear relationship was considered for the forest area. The projected land-use area shown in Table 4, where the total area obtained for the projection to 2035 is equivalent to the whole catchment area. Table 4. Projected areas for the year 2035 RSN 2035 Scenario (1) Scenario (2) Water bodies (km2 ) 15.98 10.58 10.58 Developed land (km2 ) 357.62 405.57 700.49 Forest (km2 ) 413.53 294.93 0 Agriculture (km2 ) 398.88 474.92 474.92 Total area (km2 ) 1186 1186 1186 Runoff coefficient 0.61 0.63 0.77 278 M. Jayatharan et al. Scenario (2) is done using the values obtained from Scenario (1) but assuming that the forest area is fully cleared to make way for the developed area. This is expected in the state of Selangor as there is a rapidly increasing trend of population growth. The main driving force of this migration is primarily due to the opportunities provided by urban areas, such as employment and education. Thus, this prompts the government to develop new settlements to accommodate this migration, and the forest area provides the leeway for this development [29, 30]. The runoff coefficient obtained for RSN 2035 and Scenario 1 is 0.62 and 0.63, almost the same. However, Scenario 2 has a significant increase in runoff coefficient with a value of 0.77. This proves that the rate of urbanisation has a significant impact on the runoff coefficient of the catchment area. Thus, utmost care must be taken into consideration for future development plans within the region. 440 466 466 443 500 500 471 471 460 600 437 600 Agricultural Area (km2) Forest Area 400 400 ( km2) 200 y = 438.57x0.0269 y = -10.885x + 513.08 200 R² = 0.2859 R² = 0.8803 0 0 2016 2017 2018 2019 2020 2016 2017 2018 2019 2020 Year Year 233 233 235 235 270 20 16 14 13 13 13 300 Water bodies land (km2) Developed 200 10 (km2) y = 219.73e0.0307x y = 15.669x-0.117 100 R² = 0.5875 R² = 0.8269 0 0 2016 2017 2018 2019 2020 2016 2017 2018 2019 2020 Year Year Fig. 7. Official Land use data with trendlines used for projection in Scenario (1) 5 Conclusion The summarised findings of this research are: 1. The consideration of land use to develop runoff coefficients for the development of hydrological models increases the performance of the calibrated model. The method conveyed in this paper would benefit water practitioners in calibrating hydrological models which initially provide poor performance rating scores. 2. Landsat images can be used to fill in gaps for determining the rate of urbanisation when there is a lack of high-resolution data available. However, some discrepancies should be taken into consideration before implementing them. Thus, utmost care must be taken when using the Landsat-8 satellite images. 3. The development of future scenarios based on current land use data and government policies provide hydrologist with an initial indication of the potential effects of urbanisation towards discharge within the watershed. The Impact of Urbanisation on Catchment Discharge 279 4. The increase in the developed area drastically increases the value of the runoff coefficient, and thus some measures must be taken to curb this. The implementa- tion of nature-based solutions would be a potential avenue in reducing the overall imperviousness of the catchment area. 6 Future Recommendations There are parts of this research that require fine-tuning to validate further the methods proposed throughout this paper. Firstly, a more extensive dataset should be considered, especially in developing trendlines for future projection and determining the applica- bility of Landsat-8 satellite images for land use classification. If data and funds are available, the researcher should consider higher resolution satellite images for the land use classification in replacing Landsat-8 satellite images. Ground truth survey should be conducted within the study area to provide a more reliable accuracy assessment. Lastly, various other methods utilised to obtain runoff coefficients should be considered and compared with the methods proposed within this paper. Acknowledgement. This research was funded by the RECONECT grant provided by European Commission, Horizon 2020 Innovative Actions, with the project code of ENG/RECONECT/08– 2018/010. In addition, the authors would like to thank DHI for providing the MIKE software and DID for the hydrological data used in this research. References 1. Tollan, A.: Land-use change and floods: what do we need most, research or management? Water Sci. Technol. 45(8), 183–190 (2002) 2. Saghafian, B., et al.: Flood intensification due to changes in land use. Water Resour. Manage 22(8), 1051–1067 (2008) 3. Chabaeva, A., Civco, D.L., Hurd, J.D.: Assessment of impervious surface estimation techniques. J. Hydrol. Eng. 14(4), 377–387 (2009) 4. Deng, Z., et al.: Land use/land cover classification using time series Landsat 8 images in a heavily urbanized area. Adv. Space Res. 63(7), 2144–2154 (2019) 5. ED Chaves, M., CA Picoli, M., Sanches, I.D.: Recent applications of landsat 8/OLI and Sentinel-2/MSI for land use and land cover mapping: a systematic review. Remote Sens. 12(18), 3062 (2020) 6. Hussein, K., et al.: Land use/land cover change along the eastern coast of the UAE and its impact on flooding risk. Geomat. Nat. Haz. Risk 11(1), 112–130 (2020) 7. Umukiza, E., et al.: Impacts of land use and land cover changes on peakdischarge and flow volume in kakia and esamburmbur sub-catchments of Narok town, Kenya. Hydrology 8(2), 82 (2021) 8. Loveland, T.R., Irons, J.R.: Landsat 8: the plans, the reality, and the legacy. Remote Sens. Environ. 185, 1–6 (2016) 9. Weng, Q.: Modeling urban growth effects on surface runoff with the integration of remote sensing and GIS. Environ. Manage. 28(6), 737–748 (2001). https://doi.org/10.1007/s00267 0010258 280 M. Jayatharan et al. 10. Sriwongsitanon, N., Taesombat, W.: Effects of land cover on runoff coefficient. J. Hydrol. 410(3), 226–238 (2011) 11. Mahmoud, S.H.: Investigation of rainfall–runoff modeling for Egypt by using remote sensing and GIS integration. CATENA 120, 111–121 (2014) 12. Wang, G., et al.: Using the SWAT model to assess impacts of land use changes on runoff generation in headwaters. Hydrol. Process. 28(3), 1032–1042 (2014) 13. Shadmehri Toosi, A., et al.: River basin-scale flood hazard assessment using a modified multi-criteria decision analysis approach: a case study. J. Hydrol. 574, 660–671 (2019) 14. Salvadore, E., Bronders, J., Batelaan, O.: Hydrological modelling of urbanized catchments: a review and future directions. J. Hydrol. 529(P1), 62–81 (2015) 15. Seyam, M., Othman, F.: The influence of accurate lag time estimation on the performance of stream flow data-driven based models. Water Resour. Manag. 28(9), 2583–2597 (2014). https://doi.org/10.1007/s11269-014-0628-9 16. Nahrim: Final Report: Study of the Impact of Climate Change on the Hydrologic Regime and Water Resources of Peninsular Malaysia (2006) 17. DHI, Mike ZERO - User Guide. MIKE by DHI (2017) 18. Talei, A., Chua, L.H.C.: Influence of lag time on event-based rainfall-runoff modeling using the data driven approach. J. Hydrol. 438–439, 223–233 (2012) 19. Ajmal, M., Waseem, M., Ahn, J.-H., Kim, T.-W.: Improved runoff estimation using event- based rainfall-runoff models. Water Resour. Manage. 29(6), 1995–2010 (2015). https://doi. org/10.1007/s11269-015-0924-z 20. Hossain, S., Hewa, G.A., Wella-Hewage, S.: A comparison of continuous and event-based rainfall-runoff (RR) modelling using EPA-SWMM. Water (Switzerland), 11(3) (2019) 21. DID, Urban Stormwater Management Manual for Malaysia, 2nd (Edn.) (2012) 22. JPBD, Rancangan Struktur Negeri Selangor 2035 (2015) 23. Doulgeris, C., et al.: Ecosystem approach to water resources management using the MIKE 11 modeling system in the Strymonas River and Lake Kerkini. J. Environ. Manage. 94(1), 132–143 (2012) 24. Cohen, J.: A coefficient of agreement for nominal scales. Educ. Psychol. Measur. 20(1), 37–46 (1960) 25. How Jin Aik, D., Ismail, M.H., Muharam, F.M.: Land Use/Land cover changes and the relationship with land surface temperature using landsat and MODIS imageries in cameron highlands, Malaysia. Land 9(10), 372 (2020) 26. Karakizi, C., et al.: Detailed land cover mapping from multitemporal landsat-8 data of different cloud cover. Remote Sens. 10(8), 1214 (2018) 27. Miettinen, J., Gaveau, D.L.A., Liew, S.C.: Comparison of visual and automated oil palm mapping in Borneo. Int. J. Remote Sens. 40(21), 8174–8185 (2019) 28. Oon, A., et al.: Discriminating between large-scale oil palm plantations and smallholdings on tropical peatlands using vegetation indices and supervised classification of LANDSAT-8. Int. J. Remote Sens. 40(19), 7312–7328 (2019) 29. Yaakob, U., Masron, T., Masami, F.: Ninety years of urbanisation in Malaysia: a geographical investigation of its trends and characteristics. J. Ritsumeikan Soc. Sci. Humanit. 4(3), 79–101 (2010) 30. Aisyah, A., et al.: Deforestation analysis in Selangor, Malaysia between 1989 and 2011. J. Trop. For. Sci. 27(1), 3–12 (2015) Can an Environmentally Sustainable Construction be Affordable? Paulo Mendonça(B) and Clara Vieira School of Architecture, Art and Design, Lab2PT, University of Minho, Azurém Campus, 4800-058 Guimarães, Portugal

Abstract. The development strategy based on unlimited economic growth has been increasing extraction and consumption of natural resources and is responsible for pollutant emissions and wastes, on such a way that we risk of rapidly depleting planet earth. And although there is an awareness on optimizing available economic resources and an effort to reduce the environmental impact of production, what is certain is that pollutant emissions have not been reduced at a global level, quite the opposite. The construction industry is one of the sectors with great responsibility for greenhouse effect gas emissions. This work intends to demonstrate that, by avoiding the use of new and industrialized materials, replacing them as much as possible with reused, natural and locally available ones, such as adobe and wood is possible to reduce the environmental impact, but still being economically feasible. As a case study, a single-family housing project developed according to these principles is presented. Keywords: Sustainable construction · Reuse · Local materials · Adobe · Wood Frame 1 Introduction to the Case Study Environmental policies in construction sector have been mainly focused on the adoption of energy efficiency measures in the use of buildings that, in the long term, improve the building’s behavior and, consequently, increase comfort, provide less maintenance cost and less impact on environment. However, the option for the use of more sustainable materials in the construction of buildings, in terms of their production, has been a strategy relegated to a second plane. The material itself implies an associated impact, and it is noteworthy that architecture, in its physical state, is an agglomeration of materials. Thus, the judiciously selection of materials to be used has a significant influence on the impact of buildings in all phases of its life cycle. All design stages that precede construction influence, albeit indirectly, the environ- mental impact of buildings. When the building is correctly designed in relation to its climatic environment, it ends up reducing the environmental impact related with its use. During the construction phase, it is the selection of the materials that has most significant consequences on its environmental impact. At the time of its selection some aspects must © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 281–288, 2023. https://doi.org/10.1007/978-981-19-3983-9_25 282 P. Mendonça and C. Vieira be taken into account: the amount of material to be used; its nature (because the closer to its natural state, the smaller its processing, and, consequently, the lesser its impact); prefer reused materials (thus avoiding the pollution generated by the production of new materials and also avoiding the production of solid waste, destined for garbage). These were the premises for defining the Case Study solution presented and analyzed in this research. The case study is the design of a single-family house described in detail in [1] and showed in Fig. 1, which aims to counteract a whole set of trends in consumer architecture: it minimizes itself in its area, seeks to optimize natural resources, and, above all, it has as a basic principle to optimize the selection of materials. The design options, thought out from the initial stage and which ended up influencing the materials to be selected, were based on four strategies in order to guarantee the minimum impact of the building: Less; More Natural; More Reuse; Less Transport, More Local. Fig. 1. Isometric view and vertical section of the case study house (adapted from [1]). 2 Environmental Impact Analysis of the Case Study The materials used in the exterior walls and structure on housing buildings in Portugal are generally heavyweight [2]. Concrete and brick are used in the conventional construction system that allows high thermal inertia and acoustic insulation. Following a previous study by Mendonça [3], it could be concluded that brick accounts for almost 40% of the Embodied Energy (EE) and consequently also of the Embodied Carbon (EC) of the Can an Environmentally Sustainable Construction be Affordable? 283 conventional building construction, and especially on exterior walls. Thus, for obtaining more environmentally sustainable building solutions, the consideration of materials with less EE and EC than brick and concrete should be pondered. The use of more lightweight solutions, even when associated to conventional heavyweight solutions in what can be called as mixed weight solutions, has proved to present less embodied energy and be at least equally efficient in terms of functional performance [3]. Combining the traditional materials–Earth and Wood, allows to obtain a Mixed Weight solution, such as the one here proposed and analyzed in comparison with conventional construction (Concrete and Brick) and with Lightweight building systems (Light Steel Frame and Wood Frame solutions). Concrete has been associated with significant environmental impacts of construction and, even so, its use remains widespread and characteristic of conventional construc- tion [4]. In conventional concrete and brick system, the components are almost always permanently fixed, composing an inseparable unit, which causes components with short useful life to condition components with longer useful life. A fundamental principle for efficient reuse of building components is the differentiation between the building com- ponents. Systems that pretend to be easily disassembled should allow dry assembling, with components prepared to be mechanically fixed and not glued in a permanent and irreversible way. Modular buildings start to be common in Portugal, mainly for single family houses. The most common materials used in this type of buildings are timber and steel [5]. As exterior cladding materials, these buildings preferably use wooden solutions (strips, agglomerated or plywood panels), metal (profiled sheets and sandwich panels), in ventilated façades, in addition to the ETICS solution which is not very suitable for disassembly and transport. Oliveira et al. [4] studied the viability of four constructive systems used in South European (France, Greece, Italy, Portugal, Spain) in terms of economic and environ- mental costs, considering the cost of materials, shipping cost and labor cost. From this study, were selected three of the constructive systems analyzed for a current Portuguese typology of a single family housing dwelling: Conventional System (CS) made with non-structural hollow brick and structural reinforced concrete frames; Light Steel Fram- ing (LSF) and Wood Frame (WF). These were compared with the case study system, called as A + W, based on Adobe walls and WF roof. All systems were analyzed without foundations, as these were assumed to be equal for all the solutions. The analyzed solutions were defined to have in common the same heat transfer coefficient for opaque horizontal elements 0,26 W/m2. ºC and vertical opaque elements 0,30 W/m2.ºC. These coefficients respond to the required demands for the more severe climatic zone as mentioned in the Portuguese thermal regulation [6]. In the present study the environmental parameter considered was the Embodied Carbon (EC). In this study was considered just the envelope and dividing walls as well as the roof. The envelope walls and roof present similar properties in terms of U value in all the solutions and are adequate to Portuguese thermal regulations [6]. The foundations were not considered, as these were assumed to be equal for all the solutions and depending on the type of soil. All other common specifications, such as technical installations, were not evaluated as these would be equal in all compared solutions. The methodology 284 P. Mendonça and C. Vieira considered in the weight evaluation of the CS, LSF and WF solutions were described in detail by Oliveira et al. [7]. By the analysis of Fig. 2, it can be concluded that A + W Case Study solution using adobe in the exterior walls and wood in the roof allows a reduction of 86% in Embodied Carbon of materials production and transport, in relation to the conventional hollow brick solution, and still present 50% of reduction in relation with the second best solution, the Light Wood Frame. Regarding all the environmental aspects considered, the exterior wall system using natural material come out as more sustainable than the Conventional Solution as well as the Light Steel Frame and Wood Frame solutions, considering both the EC of materials production and transport. Fig. 2. Materials production and transport embodied carbon of the analysed solutions. 3 Economic Cost Analysis of the Case Study Economic cost is one of the most important factors to take into account when designing. When comparing the cost of each material, for example, wood tends to loose, because the cost exceeds that of concrete [7]. But is this cost real? Although the final price is one of the main factors for those who invest in real estate, it is important to realize that the cost of materials will always depend on the market in which it operates, and that is how the lower price of concrete is justified. Otherwise why would it make sense for a Can an Environmentally Sustainable Construction be Affordable? 285 material with more manufacturing process to be cheaper? It is always the adhesion that determines the market price. However, there is always a way to ensure that the price of construction does not rise: equilibrium. If wood takes an initial high price, let us use wood only in the quantity needed and don’t forget that in the end of its useful life, the selective dismantling of a wood building is much cheaper and less impacting than the dismantling of a concrete building. And let’s try to reduce the price of materials whenever possible and show a good compromise between environmental impact and cost. By the analysis of Fig. 3, it can be concluded that A + W Case Study solution is 51% more expensive than the Conventional Heavyweight solution and 10% more expensive than the Light Wood Frame solution, however it is 13% cheaper than the Light Steel Frame solution. Despite not having been accounted for in this study, since the reference solutions did not include these components in the comparative analysis, the reuse of coatings and frames, as forecasted in this proposal (see Fig. 4) would significantly reduce the cost of these components, by 30%. Fig. 3. Construction cost of the analysed solutions. In the analysis of the solutions studied, considering its EC, the most polluting solution is the conventional system (CS), if we include transport in the analysis, since the LSF solution, even though it is the one with the greatest EC in its production, it becomes slightly less polluting than the CS if transport is included, as it is much lighter. The WF and A + W solutions are the least polluting. With regard to economic cost, the conventional solution turns out to be the cheapest, however, as it requires more transport, adding the associated costs, it ends up becoming more expensive, depending on the 286 P. Mendonça and C. Vieira distance at which the materials have to be purchased, being considered an average of 200 km in Portuguese reality [7]. The WF is the second cheapest solution, as it doesn’t need a lot of transport (like the LSF) it ends up gaining even more advantage in its price. The LSF solution is the most expensive solution due to the high cost of steel. The A + W proposal is slightly more expensive than the WF one. Fig. 4. Reused window frames selected to compose the south façade of the case study house (adapted from [1]). 4 Conclusions Although the reduced environmental impact of the Case Study solution was a somewhat predictable conclusion, due to the fact that Adobe and Wood are used, materials close Can an Environmentally Sustainable Construction be Affordable? 287 to their natural state; the fact that it could be an affordable solution, still less expensive than that of the Wood Frame was not that predictable. In fact, there is always a tendency for the most current practice to be the most accessible due to the demand-supply law, however, what is not normally expected is that an intensive and unconventional labour solution may still have a competitive cost. As mentioned above, the case study intended to make a compromise between envi- ronmental and economic costs, and hence the balance between Adobe structural walls together with a Wooden structure. This balance turns out to be very visible in this analy- sis: as the Wooden structure is found, for the most part, in the roofing solution, we realize that its increased cost in comparison to Conventional construction in concrete and hol- low brick is due to the Wooden roof solution. The double-wall adobe walls present a low cost of construction, comparing it with the other wall solutions, it turns to be the cheapest, however, the Adobe + Wood solution ends up being the second most expensive essentially due to the cost of the roofing solution. This research aimed to answer the question whether it is possible to have a low envi- ronmental impact solution with affordable economic cost. In the Case Study House here presented, only the pollution generated by the transport of materials were considered, but not the economic cost. It was pre-defined that only local materials would be used in the Case Study, with a maximum distance of 50 km from the site. The building material used in more quantity, Adobe, would be manufactured in the site, manually and by hand, using the land extracted during the foundations execution phase. Although the economic costs of transport were not accounted for, as this is usually already included in the final cost of the product, if it could be independently accounted it would only benefit the Case Study solution. Acknowledgments. This work has support of Lab2PT - Landscapes, Heritage and Territory laboratory and FCTMEC and when applicable of the FEDER co-financing, in the aim of the new PT2020 partnership agreement. References 1. Vieira, C.: O projeto de uma casa humilde: num contexto de sensibilidade ambiental. Guimarães: Escola de Arquitetura da Universidade do Minho, outubro de 2020. http://hdl. handle.net/1822/69246. Accessed 05 July 2021 2. Mendonça, P., Couto, J.P., Reis, A.P.: Economic and construction analysis of lightweight membranes in housing in temperate climates. Environ. Eng. Manag. J. 10(10), 1485–1493 (2011) 3. Mendonça, P.: Living under a second skin–strategies for environmental impact reduction for solar passive constructions in temperate climates (in Portuguese). Ph.D. thesis in Civil Engineering. University of Minho (2005) 4. Oliveira, F., Mendonça, P., Couto, J.P., Camões, A.: Environmental impact and comparative economic analysis among different building constructive systems used in Portugal. WSEAS, energy, environmental and structural engineering series n. 29 (2014). ISSN 978-960-474-391-9 5. Couto, J.P., Mendonça, P.: Deconstruction Roles in the Construction and Demolition Waste Management in Portugal-From Design to Site Management; Waste Management/Book Chapter 16, Sunil Kumar (Ed.) from National Environmental Engineering Research Institute (NEERI)–India, INTECH, pp. 301–322 (2011). ISBN 978-953-307-469-6 288 P. Mendonça and C. Vieira 6. REH-Portuguese Regulation for the Energy Performance of Housing Building, Decree-Law 118/2013 of 20th August, Portugal 7. Oliveira, F., Mendonça, P. Couto, J.P., Camões, A., Silva, E.: Comparative environmental and economic analysis of South European building constructive systems. WSEAS Trans. Env. Devel. 10, 509–528 (2014). E-ISSN: 2224-3496 Climate Change and Sustainable Campus Planning: A Review of Michigan Universities’ Climate-Related Plans AyseOzcan Buckley1(B) and Eric J. Strauss2 1 Faculty of Economics and Administrative Sciences, Political Science and Public Administration, Giresun University, 28200 Giresun, Turkey

,

2 SPDC Urban and Regional Planning, Michigan State University, East Lansing, MI 48823, USA

Abstract. This paper explores the link between climate change and sustainability within the framework of higher education, with the aim of investigating the funda- mental motivations for climate planning and sustainable development in selected six Michigan Universities . Regarding this, UNFCCC principles and Sustainable Development Goals (SDGs) constitute the criteria of the study. Within the context of the UNFCCC principles and sustainability the text-based analysis is carried out in six main categories: (1) Climate change (2) Greenhouse gas emissions, (3) Sustainability, (4) Renewable energy, (5) Responsibility covering education, com- munity and awareness, and (6) International documents or agreements. Briefly, this analysis may provide a useful lens for a climate-related campus planning and then guide for developing some policy recommendations. Keywords: Campus climate planning · Climate action plans (CAPs) · Climate-related plans (CRPs) · Sustainability · Sustainable development · UNFCCC · Michigan 1 Introduction Universities are the most important educational institutions and drivers that contribute to the spread of sustainable and ecological practices and adaptation of climate planning thanks to their transformative and pioneering role in society. Sustainable university or sustainable campus regarding climate change planning practices has become a rapidly increasing trend around the world, in particular in Michigan Universities. Hence, “de- veloping sustainable and ecological campus practices in climate change context” is to create places that are environmentally friendly, minimize energy savings, have an effec- tive waste management, use nature-friendly products and materials, and finally contribute to sustainable development. Universities conducting research and educational activities in the field of climate change also play an important role in creating awareness, developing policies, testing © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 289–312, 2023. https://doi.org/10.1007/978-981-19-3983-9_26 290 A. Buckley and E. J. Strauss different policy options and sometimes identifying them. Despite the prevalence of aes- thetic landscape concerns of campus planning all over the world, since 2000s there is a consensus in academia to suggest that climate change objectives and sustainability are frequently a guiding focus of the campus planning process. There are many ways for universities to support further policy integration between adaptation, sustainable development and GHGs emissions reduction, owing in part to the common themes, scopes and objectives of three global agendas [1]. Universities can have a direct contribution to address climate change through reducing carbon emissions of travel and energy usage interventions (e.g. renewable energy, waste-to-energy, energy efficiency) [2]. It is worth noting, however, that Michigan CAPs (Climate Action Plans) have recently placed more attention in campus planning efforts taking into consideration climate change impacts. University campuses might be taken as a model for cities to reduce greenhouse gas emissions and to reach sustainable development goals because they are smaller in scale and structurally like cities. Based on this fact, in this study, climate change perception of Michigan universities is examined within the United Nations Framework Convention on Climate Change (UNFCCC) [3] principles and Sustainable Development Goals (SDGs). Thus, the study discusses campus climate planning by analyzing the climate-related plans (CRPs) (e.g. sustainability plan and energy plan) or climate action plans (CAPs) of six selected Michigan Universities. In this context, the selected six Universities’ Plans for climate change (Michigan State University, University of Michigan, Wayne State University, Grand Valley State University, Western Michigan University, and Kalamazoo College) are the focus of the paper. The study aims to explore the level of response of these selected plans to the principles of the UNFCCC and SDGs through content analysis including text-based evaluation. Additionally, the research question points out that how these plans can develop climate-sensitive campus planning in cities and what are the roles of universities in promoting sustainability in cities. 2 Methodology and Approach The core question of the study is to analyze how the UN SDGs are being integrated into university climate planning. The data of this study were gathered through primary sources, like the information provided by the university climate action plans or similar plans and websites of selected universities and their own guidelines. Literature research was done by using Web of Science [4], Scopus and Google Scholar search engines. To find previous publications and research we used the climate- related keywords such as “climate change”, “climate change planning or climate action planning”, “green campus”, “sustainability and university”, “sustainable campus.” After initial search, snowball method was followed to explore other research. Additional research was done in Web of Science by using the keyword “sustainable campus” in order to specify number of academic studies. There is no single definition or set of criteria for the concepts of sustainable campus and climate-friendly or green campus in the literature for universities. Each institution sets its own goals for a sustainable campus and has its own motivational responses to climate change. Climate Change and Sustainable Campus Planning 291 It has been determined that green campus projects and academic studies have increased rapidly since 2007 within the framework of climate change and sustainabil- ity. According to the results of the Web of Science scanning with the key concepts of “sustainable campus, climate change planning, campus climate planning, green cam- pus, eco-campus, climate friendly campus, university climate planning….” the number of publications contains climate-related keywords (29,835 results for all years since 1980) in the last 42 years, and it supports the theory we mentioned here (since 1980). The publications containing the keywords “green campus (1374 results), university cli- mate planning (1016 results), sustainable campus (2163 results), eco-campus (30)” have been increased especially since 2005. For the keyword “campus climate planning” the publications intensified after 2012. In addition, in recent years, green campus activities and other university projects have been classified in different ranking systems1 . The ranking topics cover from research, education and academic reputation to environmental performance. Within this regard, the UI GreenMetric ranking is an important initiative that makes an international assessment of the sustainability of universities2 [5, 6]. We firstly prefer to analyze university CAPs, sustainability plans or energy plans to discuss campus climate planning. It is hard to evaluate and systematically measure the implementation of SDGs and climate planning awareness for each university. Content analysis is going to reveal the core components and policies in each plan. Taking into consideration research limitations and discussions, we point out some other research questions as follows: • RQ1. What are the institutional factors behind preparing or developing University Climate Action Plan or similar one? • RQ2. What is the motivation to develop policies against climate change and how do they engage these policies into sustainable goals (What is the motivation for climate responses?) • RQ2. How do they influence Universities actions institutionally? • RQ3. Do or will University CAPs tend to look increasingly similar in their sustainability politics, programs and practices due to evaluations? • RQ4. What are the benefits and disadvantages of having a CAP or similar plan to support SDGs? Is it important to sustainability performance? • RQ5. What are the implications and lessons learnt for sustainability practices on campus? 1 Prestigious rankings are as follows: Academic Ranking of World Universities (ARWU); Times Higher Education World University Ranking (THE); Quacquarelli Symonds (QS); THE Impact; Sustainability Tracking, Assessment and Rating System (STARS) or IU GreenMetric World University Ranking. 2 UI GreenMetric was created in 2010 by the University of Indonesia, and it consists of six main categories which are Setting & Infrastructure (15%), Energy & Climate(21%), Waste (18%), Water (10%), Transportation (18%) and Education (18%) [7]. 292 A. Buckley and E. J. Strauss 3 Climate Change and Sustainability In this part, we discuss the UNFCCC principles and Sustainable Development Goals for UN 2030 Agenda within the context of climate planning. 3.1 UNFCCC Principles and Its Importance for University Climate Planning The UNFCCC is structurally styled as a framework convention, like the Vienna Con- vention for the Protection of the Ozone Layer and the Convention. As knowledge and consensus grow within this framework, the agreement is supplemented by a series of protocols and amendments imposing progressively more specific and more stringent obligations on the treaty parties” [8]. The UNFCCC principles that have influenced, and continue to shape international climate negotiations are summarized below [3, 9]. 3.1.1 Equity Principle (Article 3.1) The principles of the Convention, under Article 3, in particular the principles in Arti- cle 3.1 which reads as follows: “The Parties should protect the climate system for the benefit of present and future generations of humankind, on the basis of equity and in accordance with their common but differentiated responsibilities and respective capa- bilities. Accordingly, the developed country Parties should take the lead in combating climate change and the adverse effects thereof.” 3.1.2 Common But Differentiated Responsibilities (Article 3.1 and 3.2) This means that while developing country parties are expected to contribute to climate mitigation, because of superior capacity to undertake mitigation and greater contribution to the problem of climate change as a result of historical emissions, developed countries are expected to “take the lead in combating climate change and the adverse effects thereof.” 3.1.3 Precautionary Principle (Article 3.3) This means that the Parties should take precautionary measures to anticipate, prevent or minimize the causes of climate change and mitigate its adverse effects. To achieve this, such policies and measures should take into account different socio-economic contexts, be comprehensive, cover all relevant sources, sinks and reservoirs of greenhouse gases and adaptation, and comprise all economic sectors. 3.1.4 The Right and Obligation to Support Sustainable Development (Article 3.4) The countries or Parties have a right to, and should, promote sustainable development. Policies and measures should be integrated with national development programs, regard- ing that economic development is essential for adopting measures to address climate change. Climate Change and Sustainable Campus Planning 293 Additionally, under the UNFCCC, all parties are required to submit national GHG inventories, and developed country parties are required to submit more detailed descrip- tions of mitigation policies and projections of the projected impact of these policies on GHG emissions. This responsibility also reveals the importance of the data and evaluations developed by the climate plans or climate-related plans of the universities. 3.2 Sustainable Development Goals and University Climate Planning Studies on sustainability have started since the Stockholm Declaration in 1972, and higher education institutions have been involved in these studies since 1990 [10]. The 1972 Stockholm Declaration also drew attention to the relationship between Higher Edu- cation Institutions (HEIs) and the environment and campus planning was reconstructed as a separate political area of environmental management in this regard. Since 1990, universities have taken various initiatives related to sustainability, formed unions, soci- eties, and signed declarations and conditions. With the Talloires Declaration signed in 1990 in France, it has become an important step for HEIs to focusing on environmental problems [11]. The Rio + 20 Declaration in 2012 has five scopes for universities; “Teaching sus- tainable development concepts, promoting research on sustainable development issues, greening campuses, supporting sustainability efforts, and promoting and incorporating international cooperation” [12, 13]. After the Rio Declaration, SDG targets were set in 2015. The United Nations Decade of Education for Sustainable Development (2005–2014) emphasizes the importance of education to increase world sustainability [14]. It has aimed at integrating the principles and practices of sustainable development into all aspects of education and learning, to encourage changes in knowledge, values and atti- tudes with the vision of enabling a more sustainable and just society for all [15]. The overall objective was “to integrate the principles, values and practices of sustainable development into all aspects of education and learning” and create “a more sustainable future in terms of environmental integrity, economic viability and just society for present and future generations.” [16, 17]. There is now an increased awareness at the international policy level that education has a crucial role to the advancement of sustainable development and climate change planning, with many countries committed to continuing to work to SDGs at the national and local levels. Especially for the last 20 years, the majority of university campuses in Europe and North America have been developing CAPs, sustainability plans, energy plans and similar institutional and legal instruments as part of the development of environmen- tal policies and UN sustainable development goals. In parallel, they are trying to be involved in the “sustainable and green campus” approach through various research pro- grams, courses and seminars. Today, university administrations continue to strengthen the ground that will reduce concerns about climate change mitigation and campus sustainability’ [13]. Sustainable development goals for UN 2030 Agenda consisting of 17 interrelated topics cover all aspects of sustainable development, and universities have an important role to achieve almost half of the targets. In this regard, we can ask these questions to 294 A. Buckley and E. J. Strauss look into university roles in responding the SDGs and climate change impacts: How academia may support these 17 goals and what methods academia should use to reach those? How SDGs can be applied within university climate planning context? The SDGs, which consist of 169 targets and 6014 actions, are important for devel- oping adaptation policies to climate change as a whole. Looking at the goals within the framework of the role of universities in climate planning, it can be said that two of 17 goals directly form a legal and political baseline for climate planning. These are Goal 13 and Goal 11. Goal 13’s main point is to take an urgent action to combat climate change and its impacts. Under this goal five targets are identified for SDGs action areas. All those targets are directly related to university climate planning and giving the some respon- sibility to universities. For example, integrating climate change measures into national policies, strategies and planning is one of the targets (see target 13.2) that draws atten- tion to the climate planning at all levels. This target 13.2 clearly emphasizes the impor- tance of education, awareness-raising and human and institutional capacity on climate change mitigation and adaptation policies. 13.b can also be evaluated within this scope. It stresses to promote mechanism for raising capacity for effective climate-related planning and management, especially in least developed countries and Small Island developing States. Shortly, it is said that long-term climate strategies, national adaptation plans and adaptation communication are indicators for all these targets. Goal 11’s main point is about sustainable cities and communities, and its focus is to make cities and human settlements inclusive, safe, resilient and sustainable. There are ten targets identified under this goal. Providing access to safe, affordable and sustainable transport systems and green and public spaces for all (target 11.2 and 11.7), enhancing integrated and sustainable human settlement planning and management (target 11.3) and reducing the adverse per capita environmental impact of cities (target 11.6) directly guide universities in terms of action areas in climate planning. In addition, target 11.b states that it is important to increase the number of cities and human settlements adopting and implementing integrated policies and plans towards inclusion, resource efficiency, mitigation and adaptation to climate change. The SDGs Agenda sets a universal education program with the adoption of Sus- tainable Development Goal 4, which aims to ‘ensure inclusive and equitable quality education and promote lifelong learning opportunities for all’ [18]. Under Goal 4, it is widely recognized that one of the most ambitious, interesting and challenging targets is target 4.7, which aims to: “By 2030, ensure that all learners acquire the knowledge and skills needed to promote sustainable development, including, among others, through Education for Sustainable Development and sustainable lifestyles, human rights, gender equality, promotion of a culture of peace and nonviolence, global citizenship and appre- ciation of cultural diversity and of culture’s contribution to sustainable development” [19]. Other SDGs and its prominent targets are briefly summarized in terms of university climate planning in Table 1 below. Climate Change and Sustainable Campus Planning 295 Table 1. How university climate planning is related to other sustainable development goals Goals Targets Goal 1. No Poverty (7 targets) Target 1.4 directly states that everyone, in particular the poor and the vulnerable, has equal rights to economic resources, as well as access to basic services, ownership and control over land and other forms of property, inheritance, natural resources Finally, it can be interpreted education is critical to lifting people out of poverty Goal 2. Zero Hunger (8 targets) In general, each of the 8 targets is important in terms of climate change. Various sub-targets have been determined on topics such as food production, protection of the agricultural sector, food safety and protection of the right to access to food In particular, target 2.4 points out the importance of ensure sustainable food production systems and implement resilient agricultural practices that help maintain ecosystems, that strengthen capacity for adaptation to climate change, extreme weather, drought, flooding and other disasters and that progressively improve land and soil quality Goal 3. Good Health and Well-being Universities play a key role in helping (13 targets) societies move towards healthy lifestyles and well-being and understanding the planet. Under Target 3.9, the prevention of diseases and deaths caused by environmental pollution and chemicals is particularly emphasized Goal 5. Gender Equality Within the 9 targets, Universities can develop (9 targets) policies that undertake reforms to give women equal rights to economic resources, as well as access to ownership and control over land and other forms of property, financial services, inheritance and natural resources Goal 6. Clean Water and Sanitation (8 targets) Universities have responsibilities to increase skills and the capacity to use natural resources more sustainably and to promote hygiene for everyone. All those can be seen as part of climate planning (continued) 296 A. Buckley and E. J. Strauss Table 1. (continued) Goals Targets Goal 7. Affordable and Clean Energy University programs can promote better (5 targets) energy conservation and uptake of renewable energy source. It has also a crucial role to transform economy based on renewable resources by research and education Goal 8. Decent Work and Economic Growth There is a direct correlation among such areas (12 targets) as economic structure, entrepreneurship, job market skills and levels of education regarding climate change. Universities are responsible for stimulating green economy policies, green jobs and sustainability in this regard Goal 9. Industry, Innovation and Infrastructure It is necessary to develop the capacity required (8 targets) to build more resilient infrastructure and more sustainable industrialization within the scope of climate change. Universities as a small town or city can achieve these targets in their own campuses Goal 10. Reduced Inequalities Universities can make a proven difference on (10 targets) the economic and social systems to reduce inequalities and ensure equality. Climate change is one of the factors effecting or deepening inequalities and so, these two issues must be considered together Goal 12. Ensure sustainable consumption and Universities can make a critical difference to production patterns production patterns (e.g. with regard to the (11 targets) circular economy) and to consumer understanding of more sustainably produced goods and prevention of waste. (see the targets 12.2, 12.3, 12.4, 12.7, 12.8) Goal 14. Life Below Water Universities are important in developing (10 targets) awareness of the marine environment and building for proactive consensus regarding wise and sustainable use by research programs. (see the target 14.a) Goal 15. Life on Land Universities can increase skills and capacity to (12 targets) underpin sustainable livelihoods and to conserve natural resources and biodiversity, particularly in threatened environments by having specific research and training programs Goal 16. Peace, Justice and Strong Institutions Universities can promote peaceful and (12 targets) inclusive societies for sustainable development. Social learning is vital to facilitate and ensure participative, inclusive and just societies, as well as social coherence (continued) Climate Change and Sustainable Campus Planning 297 Table 1. (continued) Goals Targets Goal 17. Partnerships for the Goals Universities can strength the means of (19 targets) implementation and revitalize the global partnership for sustainable development by their transformative role. Regarding this, universities also builds capacity to understand and promote sustainable development policies and practices to work together at all levels 4 The Content Analysis of Selected Michigan Universities’ Climate-Related Plans (CRPs) We present an analysis of selected six Michigan universities which develop a climate- related plan (CAPs, Sustainability Plan, Energy Plan or similar one). Plans and their objectives are examined according to the following two documents: (1) UNFCCC Principles, (2) UN Sustainable Development Goals for 2030 Agenda. The samples constituting the scope of the study were randomly selected, and espe- cially were taken into account the universities with older climate plans. These universities are as follows: Michigan State University, University of Michigan, Grand Valley State University, Western Michigan University, Wayne State University and Kalamazoo Col- lege. There are a large number of key components that have been identified to reveal the contents of the plans. The Climate-Related Plans (CRPs or Climate Action Plans-CAPs) of six Universities were evaluated by reading through the basic concepts, policies and strategies determined according to the two main policy documents (UNFCCC and UN SDGs) we mentioned above. In this sense, the content of the Plans was also subjected to a text mining survey. Through the process of text mining analysis of CRPs/CAPs it is possible to identify the most important issues the plans focused. Selected climate-related plans provided from their own website mainly response to questions relating to the status of their climate change targets within campuses. Responses of the Plans to climate change indicate that the plans have been formally compiled through a university board, council or similar institutions. Based on the responses from these Plans, it was determined that most often climate-related plans cover, or are planned to contain: GHG emissions reductions at all areas, renewable energy investments, developing education and research programs, campus planning regarding sustainability, engaging with communities at all levels, and infrastructure investment and/or design requirements; monitoring and evaluation and/or investigation of co-benefit opportunities with emissions reduction as future goal. 4.1 Status and Components The research conceptualizes the university as having four principal modalities on the axis of the relationship between the UNFCCC principles and the UN SDGs: 1. Education & Research, 2. Community Outreach & Sustainability, 3. Campus Operations & Planning, 298 A. Buckley and E. J. Strauss and 4. Goals for future generations. All plans are examined by the following criteria to identify their status and components (see Table 2). Table 2. Criteria for the analysis of the selected plans Education & research Community outreach Campus planning & Goals for future & sustainability operations • Global Climate • Economic, political, • Climate change • GHGs emissions Change studies cultural spheres, mitigation and reduction • Climate education & • Service delivery, adaptation practices • Climate Adaptation Research (academic • Collaborations (in (Clean transportation, policies opportunities, between public energy efficiency, • Establishing a courses, grants, institutions, private water efficiency, culture based on publications etc.…) sector and NGOs) waste management, sustainability • Interdisciplinary • Awareness studies, nature conservation studies • Community projects, and remediation, • Environmental law • Access opportunities pollution prevention • Renewable energy to disadvantaged etc.) studies groups • Green buildings & • Public debate • Justice & equality green infrastructure • Disabled people The evaluated plans are listed as follows: (1) Michigan State University Energy Transition Plan, (2) Grand Valley State University Climate Action Plan, (3) University of Michigan President’s Commission on Carbon Neutrality Report, (4) Western Michigan University Climate Action Plan, (5) Wayne State University Sustainability Strategic Plan and (6) Kalamazoo College Sustainability and Climate Action Plan. 1. Michigan State University Energy Transition Plan (2012, reviewed 2017) MSU has a very clear and progressive approach to the SDGs and climate change. The Office of Sustainability is responsible for developing policies, strategies and plans within the context of SDGs. The Office describes its mission as a part of UN Sustainable Development Goals. The policy framework of the MSU is based on four components (4Cs): (1) Cam- pus Planning which draws attention to creating a smart campus for sustainable future, (2) Curriculum (Research & Education) covering investments in sustainable schol- arship, cooperative education and research innovation, (3) Building a Community of global leadership in sustainability and climate change, (4) Establishing a Culture in sustainability [20]. MSU has ten sustainability-related plans and guides which cover energy, natural areas preservation, food, storm-water management, campus mobility and planning and well- head protection program. We specifically analyze the Energy Transition Plan regarding the research focus. The Plan is reviewed every five years, and so we analyzed the recent Climate Change and Sustainable Campus Planning 299 document with its first version. The first plan was prepared in 2012 as an Energy Transi- tion Plan. The updated plan in August 2017 is called Sustainability Report that explains Energy Transition Plan Five-Year Review [21]. In addition, the first plan includes a strat- egy document called Next Generation Energy Strategies prepared two years before the Energy Transition Plan. MSU mainly focuses on developing renewable energy strategies within the sus- tainability context. In this regard, MSU has made significant strategic investment in sustainability especially since 1999. Sustainability Committee established (precursor to Office of Campus Sustainability) in 1999. Another unit related to sustainability, which is Energy Transition Steering Committee, was created in 2011 to consider energy options and future advancements [22]. The Sustainability Report which is the five-year review of the Plan is briefly explain- ing the practices, strategies and operations within the five-year period of the University. It is also providing a summary of the steps MSU is taking over the next five years to continue energy progress through sustainability engagement with the campus commu- nity [21]. The core components of the Energy Plan are fully adopted in this Report (such as stakeholder engagement in energy and sustainability future, investments in energy technologies, develop sustainability across the university, advance sustainability in infrastructure). In connection with these goals, there are the following campus sus- tainability initiatives: Anaerobic digester (ADREC-Anaerobic Digestion Research and Education Center), LEED certification for all new construction and solar carports. The core goal of the MSU Energy Transition Plan consisting of 186 pages is to trans- form MSU to 100% renewable energy. In this regard, the Plan’s content is explained by three goals: Improving the physical environment, investing in sustainable energy research and development, become an educational leader in sustainable energy. In addition, the Plan also targets to develop a reliable energy system. Reliability, capacity, environment, health and cost are the variables of energy planning the University planned. The Plan briefly has a comprehensive energy planning and tries to set an integrated approach to all campus sectors. 2. University of Michigan President’s Commission on Carbon Neutrality Report (March, 2021) The U-M has long and deep historical roots of leadership and innovation in envi- ronmental issues, especially in the environmental justice context. Early green efforts on campus started in 1970s. U-M begins sponsoring vanpools to facilitate employee com- muting and institutes a recycling program. In 1970, the University of Michigan (U-M) held the nation’s first “Environmental Teach-In,” which developed a model for the events of the first Earth Day celebration, and in the 1980s, it helped to pioneer the environmen- tal justice discipline. U-M began accounting for its GHGs emissions in the early 2000s, and in 2011, U-M established its current GHGs reduction goal (25% below 2006 levels by 2025) for the Ann Arbor campus [23]. 300 A. Buckley and E. J. Strauss In 2011, the university’s sustainability initiative was launched, and this initiative com- mits the campus to have a series of ambitious goals, including climate-related policies. In the following period, in February 2019, the University establishes the President’s Com- mission on Carbon Neutrality (PCCN) to develop recommendations for how to achieve carbon neutrality for U-M [24]. The President’s Commission on Carbon Neutrality is charged with recommending strategies for U-M to achieve net-zero emissions and submits its final report and recom- mendations to university leadership. The final report which was published in May 2021 includes a set of 50 proposed actions to achieve net-zero emissions university-wide, and this report is the subject of the analysis of the paper. Additionally, University of Michi- gan strengthened its commitment to sustainability with the adoption of six long-term Sustainability Goals in September 2011. The goals were identified through a Campus Sustainability Integrated Assessment process (CSIA) which was an intensive two-year project led by the Graham Sustainability Institute [25]. The U-M is evaluated within the scope carbon neutrality3 goal which includes car- bon neutrality targets and sustainable investing strategies. The University is committed achieving carbon neutrality across the Ann Arbor, Flint, and Dearborn campuses. Looking into the Carbon Neutrality Report details it is seen that there are four core components: Emissions reductions at all levels, developing a campus planning policy, giving a significant importance to climate-related research and university programs and engaging with all stakeholders and communities on campus and off-campus. The Report is quite detailed document which starts with the explanations about the Commission (PCCN) structure and process overview. Not taking into account the two first parts it can be seen that the Report explains emission reduction targets within four policy frameworks. The first two parts are introducing the Letter from the Commission and Guiding Principles. Four policy frameworks are discussing the following topics: (1) Process Overview, (2) Organization and Culture, (3) Neutrality Goals and Emis- sions Mitigation Strategies, (4) Next Steps. This paper’s focus is especially on the last three frameworks. The principles guiding these four policy frameworks are listed as follows: Carbon Neutral, Sustainability, Equity and Justice, Scalable and Transferable, Regional Community Involvement, U-M Community Participation and Accountability, and financially taking responsibility. The first one, Process Overview, is a summary of the Commission’s works and the pol- icy frameworks. The core components of the Report regarding this section are community engagement, environmental justice, carbon accounting, and financial analysis. The second, Organization and Culture, is a section identifying four sub-areas which are leadership structures, campus planning, research and education, external collaboration. 3 Carbon neutrality, at a global level, is balance between emitting carbon and absorbing carbon from the atmosphere in carbon sinks. At the level of an institution, carbon neutrality means that all quantifiable GHGs attributable to that institution’s activities are eliminated or offset by investments in carbon credits or sequestration projects [23]. Climate Change and Sustainable Campus Planning 301 The third, Neutrality Goals and Emissions Mitigation Strategies, is a detailed section of the Report describing all targets for carbon neutrality. It covers all recommendations under three scopes which are about emissions reduction strategies. The fourth section is a one-page summary of the targets. The third section provides recommendations for carbon neutrality goals across all three GHGs scopes and associated strategies for how those goals can be achieved. For emission Scopes 1 and 2, it addresses how U-M can transform its energy infrastructure, operate its buildings more efficiently, source 100% renewable electricity, and transition to a carbon-free vehicle fleet. It also describes numerous strategies to mitigate and improve accounting for various Scope 3 emissions, such as those related to commuting, university travel, food, waste, leased space, and building materials. Briefly, it guides carbon offsetting strategies that can be employed until all emissions are fully eliminated [23]. The University commits to carbon neutrality (inclusive of offsets) for Scope 1 emissions across all three campuses by 2025. Scope 2 emissions are associated with purchased electricity. In the context of Scope 1 and 2, the University is responsible for converting heating and cooling infrastructure from natural gas-fueled system, ensuring that all electricity clearly map to carbon-free generation sources, reducing energy demand and conservation behaviors among campus community members, and fully decarbonizing the university’s transit system. Scope 3 GHG emissions are all of the upstream and downstream GHGs related to U-M activities excluding Scope 2 emissions, such as commuting, university travel, and purchased goods. What makes this Report and U-M’s climate approach unique and distinguishes it from other universities’ plans is the clear emphasis on environmental justice and its inclusion as a separate policy area. There is a holistic and progressive approach to the issues analyzed in the Report. In this regard, U-M expects the University’s climate plan will emphasize opportunities to engage the whole community and also alumni, donors, patients and visitors. The Report accepts that there must be high-level and systemic institutional challenge for the cultural shift which is necessary to achieve the carbon neutral goals and sustainability in near future. The Report gives a significant emphasis engaging the broad spectrum of research, scholarship, and educational opportunities in providing solutions to the institutional and global challenges leading to carbon neutrality. Remarkable research areas include renewable energy generation; bio-sequestration; carbon capture, utilization, and storage; energy efficiency; electricity grid design and management; waste-to-energy systems; mobility; life cycle analysis; carbon pricing policies; and the social impacts of climate change. In this context strategy recommendations are based on expanding and prioritizing carbon neutrality curriculum, training, and literacy programs for all members of the U-M community across all three campuses. Also investing in institutional structures must be another component of the carbon-neutrality-focused. 3. Grand Valley State University Climate Action Plan (2018) The Report was compiled by the Grand Valley State University (GVSU) Climate Action Committee. The initial CAP was submitted to the ACUPCC in January 2010. An updated plan was accepted in January 2012. Moving forward this document is the second update, and additional updates are available every two years. 302 A. Buckley and E. J. Strauss The Plan consisting of 46 pages starts with the title of GVSU’s Greenhouse Gas Inventory. Within this part, the report has brief explanations about campus emissions, scopes for GHG accounting and reporting purposes, inventory methodologies and GVSU’s carbon footprint. The second part of the plan is about GVSU’s Climate Action Methodologies. These are explained under the following headings and in three different scopes according to the emission source type. Scope 1 covers On Campus Stationary Sources, University Fleet Vehicles Refrigerants and Chemicals and Agriculture Sources. Scope 2 gives a short explanation about purchased energy. Air travel and commuting are introduced within Scope 3. The CAP is the legal tool showing attention to document its commitment to the environment, and to showcase its continued efforts to promote a sustainable lifestyle for its faculty, staff, students, and visitors [26]. The other policies/topics discussed in the GVSU CAP are as follows: (3) Summary of Inventory, (4) Greenhouse Gas Mitigation Phases, (5) Greenhouse Gas Mitigation Strate- gies (including strategies to affect student behavior, student activities, energy reduction projects for campus offices and classrooms, mitigation strategies for transportation, mit- igation strategies for campus dining), (6) Climate Education and Research (including the subtitles of academic opportunities, climate related courses, grants, and publications). GVSU CAP’s focus is to mitigate GHGs emission and to develop a sustainable society on campus by producing knowledge and research. 4. Western Michigan University Climate Action Plan (2012) WMU has an unexpectedly deep historical roots showing concern for the environ- ment and sustainability. In 1972, an Environmental Studies Program was first established at WMU. This is an interdisciplinary program that offers a “coordinate major” and a minor option open to all undergraduate students. In the 1980s, WMU developed an inno- vative strategy for funding energy conservation projects, which has been recognized by the Sustainable Endowments Institute as the first Quasi-Green Revolving Fund in the nation. Over the period between 1992 and 1997 WMU made a very significant reduction in campus CO2 emissions by converting the central power plant from coal to natural gas cogeneration [27]. In 2000, the Campus Sustainability Assessment Project, the first national effort to review and evaluate best practice in campus sustainability assessment, took place. WMU’s first greenhouse gas inventory emerged from a classroom project in 2002 and a full greenhouse gas inventory for the 2000–2001 academic years was completed. WMU has also the Strategic Sustainability Initiatives Report which was prepared in 2009. The Report is intended to set the stage for comprehensive, collaborative, campus- wide planning on sustainability [28]. It is seen that there are the following four pri- oritized strategic sustainability initiatives determined by the Report: Governance & Administrative, Operations & In Reach, Community Service & Outreach, Curriculum. Climate Change and Sustainable Campus Planning 303 All these studies can be considered as important steps in developing the University’s climate planning and sustainability policies. After 2011, WMU was focused on reduction of GHG emissions. In 2012, WMU Climate Action Plan was declared in 2012. The Plan’s focus is on developing mitigation strategies to reduce energy-related emissions. According to the Plan explanations, GHGs inventories have been performed using the protocol suggested by the Clean Air Cool Planet Campus Carbon Calculator. 2065 is determined as WMU’s target date for climate neutrality. In this regard, the major part of this Plan outlines specific strategies, if employed, will reduce net GHGs emissions to zero by 2065 [27]. The Plan also includes policies for research related to climate change, efforts to integrate climate change into the broad curriculum, and ideas for reaching out to and involving the local community in these efforts. Overall, the Plan is built on a conservative approach in recommending specific strategies to reduce GHG emissions regarding the achieving the ACUPCC agreement.4 Under the title of Recommended Strategies for Accomplishing Reductions, the CAP identifies the following ten different areas which are the sources of emissions: Gen- eral energy demand reduction, stationary combustion (power plant), purchased electric- ity, alternative energy supplies-distributed, commuting, campus vehicles, and air travel including study abroad, solid waste and wastewater, negative contributions to GHG emissions, trajectory for GHG reductions. The Plan discusses educational efforts and research efforts to climate change under separate headings. With regard to climate change, the plan additionally identifies research areas during the next few decades. Especially these areas are related to energy manage- ment and conservation in different sectors such as buildings, manufacturing processed transportation. The WMU CAP shortly emphasizes that they are overly keen on making the zero carbon target for climate change and sustainability part of the official curriculum. 5. Wayne State University Sustainability Strategic Plan (2017–2022) The first WSU Sustainability Plan outlining a variation of environmental actions to lead a green campus was developed in 2007. In coordination with this plan, the Environ- mental Initiatives Committee was created to oversee the completion of green programs and initiatives both on and around campus. Four years later in 2011, the Office of Cam- pus Sustainability was officially established to continue the push for all-encompassing, sustainable initiatives as our campus community continued to evolve. WSU has been developing green policies especially since 2002. For example, first recycling program was developed for Wayne State main campus at the same year. 4 The ACUPCC (The American College & University Presidents’ Climate Commitment) is a high-visibility effort and also an active partnership among colleges and universities to accelerate the education, research, and community engagement needed to slow and stop human-induced climate change while setting an example by eliminating net greenhouse gas emissions. Nearly 700 institutions, representing 6 million students, have signed the ACUPCC to date, and are making their GHG Inventories, Climate Action Plans, and Progress Reports publicly available online via the ACUPCC Reporting System: www.acupcc.org/reportingsystem [29]. 304 A. Buckley and E. J. Strauss The updated strategic plan (2017–2022) for the Office of Campus Sustainability represents a framework that guides environmental initiatives on campus. The document identifies the vision and mission of the Wayne State Strategic Plan encompassing energy conservation, waste reduction, transportation, and food systems based on sustainability [30]. The Plan describes the WSU sustainability vision and mission as follows: [30]. The institution’s guiding values are identified by a holistic approach that stresses making campus community to work together across all disciplines, cultures and objec- tives. There are five guiding values identified in the Plan: Collaboration, Integrity, Inno- vation, Excellence, Diversity and Inclusion. One of the remarkable points in the Plan is on the dissemination of environmental literacy. WSU is responsible for promoting environmental literacy to the campus community in all aspects of campus life regarding guiding values. The Plan is a very brief document and explains the vision and mission with the following six goals that encompass policies for climate change and sustainability: Zero waste, GHG emissions (and Carbon Footprint), Alternative transportation, Sustainable food, Funding, Academics, Research and Engagement. Within the context of these goals the University commits to reduce its greenhouse gas emissions by 25% below 2014 average levels by 2022. Therefore, it is identified as a need to establish a multi-pronged plan to achieve efficiency in utility service use, energy innovation, energy conservation measures, alternative energy and reducing the University’s carbon footprint [30]. The Plan also provides an evaluation for the implementation of all these targets in 1, 3 and 5 year periods. In particular, campus-wide waste reduction and GHG reduction are the prominent policies of the Plan. 6. Kalamazoo College Sustainability and Climate Action Plan (2010, updated 2020) The Kalamazoo College Sustainability and Climate Action Plan is a frame designed to envision, measure, and reduce GHGs and related impacts of Kalamazoo College on the climate. It was written by committee and signed in January 2010, and it was created within a 28-month project that began as a student-led initiative following the signing of the American College & University Presidents Climate Commitment in 2007 [31]. The Plan is also a very brief document as WSU Plan mentioned above, but it discusses the issues more satisfactorily. The Plan starts with Kalamazoo College’s Commitment for climate change and sustainability. In the second part, under the title of the Nature of Plan, it is seen that the content of the Plan adopts a universal perspective on the sustainability and protection of the planet and environmental issues. Climate Change and Sustainable Campus Planning 305 There are three core policies which are described as aggressive but achievable sus- tainability goals and strategies: Greenhouse Gas Reduction, Resource Conservation, and Academics and Civic Engagement. These three policies are explained by select- ing specific objectives in areas in which Kalamazoo can achieve real and measurable progress. The plan contains interim goals for 2015 but the overarching goal is a reduction of GHGs emissions from campus operations to 25% below 2008 levels by the year 2020, and complete climate neutrality by 2050. In order to track whether the College has reached its goals, the plan requires yearly tracking of our carbon data [32]. No carbon reports were produced from 2014 to 2018. The first Carbon Report was prepared in 2018, and it will serve as the baseline for the future revision of the current Climate Action Plan which will make new goals for the next five years. This 2018– 2019 Carbon report was generated by the current Sustainability Interns for Facilities Management. The components of all the selected six plans are summarized in Table 3 below. Table 3. The components of the selected universities’ plans University Analyzed plan Components Michigan State University Energy Transition Plan Core Goal- (Jan 2012, 186 pages) transition MSU to 100% renewable energy • Improve the physical environment the target for GHGs emissions reduction 2025 - 55% 2030 - 65% • Invest in sustainable energy research and development • Become an educational leader in sustainable energy • Develop reliable energy systems University of Michigan The President’s Core Goal-Mitigation Commission on Carbon Carbon Neutrality and Neutral Report emissions mitigation (March 2021, 208 pages) • Carbon accounting Sustainability Goals • Community engagement (website review) • Environmental justice • Campus planning • Research and education • External collaboration (continued) 306 A. Buckley and E. J. Strauss Table 3. (continued) University Analyzed plan Components Grand Valley State University Climate Action Plan Core Goal-Mitigation (2018, 46 pages) Reduction GHG emissions • CAP outlines steps to achieve carbon neutrality by 2043 • CAP is intended to clarify the path toward carbon neutrality • Energy reduction projects for campus • Transportation • Climate education and research Wayne State University Sustainability Strategic Core Goal-Sustainability Plan (2017–2022) (40 Reduction GHG emissions, pages) • Zero waste • Reducing GHGs emissions • Alternative transportation • Sustainable food • Developing a green fund for sustainability • Academics, Research & Engagement Western Michigan University WMU Climate Action Plan Core Goal-Mitigation (2012) (30 pages) Reduction GHG emissions, • WMU’s target date for climate neutrality is 2065, • Education & Research • Community Outreach Kalamazoo College Kalamazoo College Core Goal-Mitigation Sustainability and Climate Reduction GHG emissions, Change Plan • Greenhouse Gas Reduction, (2010, 31 pages) • Resource Conservation, and 2018–2019 Carbon Report • Academics and Civic (website review) Engagement All these components of the Plans are closely related to the following UN SDGs Agenda for 2030 and its targets: Goal 13. Climate Action, Goal 11. Sustainable Cities & Communities, Goal 4. Quality Education, Goal 6. Clean Water and Sanitation, Goal 7. Affordable & Clean Energy, Goal 3. Good Health and Well-being, Goal 9. Industry, Innovation & Infrastructure, Goal 12. Responsible Consumption & Production, Goal 16. Peace, Justice & Strong Institutions. Climate Change and Sustainable Campus Planning 307 4.2 A Conceptual Density Review The Plans are also analyzed on the key issues identifying in Table 4 below regarding text- based analysis. To measure the content of CAPs we used the term frequency technique for each plan, and we attempted to reveal a conceptual density map with the results obtained from the key policy terms and their closest sub-terms. For example, the frequency of the term ‘greenhouse gas emissions’ was determined by searching three different sub-terms such as greenhouse gas emissions, greenhouse and GHGs. The simplest form of term frequency technique we use here is to count the number of uses of each key term in each plan. Thus, the conceptual perception of the CAPs/CRPs on the policy can also be seen. Through the text-based analysis, it is possible to identify the most frequent terms or the key policy components cited in the CRPs/CAPs. According to the classification of this study, the most frequent terms are climate, greenhouse gas emissions (GHGs), sustain- ability, renewable energy followed by university, research and education, responsibility (including the terms of education, community, awareness) and international documents (agreements). Table 4. Key policy terms of the analysis 1 Climate Change • Climate, Climate Change, Climate Planning, Campus Climate Planning, Campus Planning, University Climate Planning 2 Greenhouse Gas Emissions/GHGs • Greenhouse Gas Emissions, Greenhouse, GHGs 3 Sustainability • Sustainability, Sustainable, Sustainable Future, Sustainable Campus, Campus Sustainability, Sustainable Development, Sustainable Development Goals, SDGs, Sustainable University 4 Renewable Energy • Renewable Energy, renewable 5 Responsibility: Education, Community & Awareness • Education, Research, Community, Society, Awareness, University, Justice, Environmental Justice 6 International Documents/Agreements • Agreement, IPCC, United Nations, UNFCCC, Climate Change Framework There are different text-based evaluation techniques that can be applied for examining the text patterns and their mining process [33]. The techniques that are mainly addressed and used in the literature are: (1) Document classification, (2) Information retrieval (key- word search/querying and indexing), (3) Document clustering (steps/policies/phrases clustering), (4) Natural language processing (spelling correction, grammatical parsing, and word sense disambiguation), (5) link analysis (relationship extraction), and (6) web mining (site analysis) [34–40]. 308 A. Buckley and E. J. Strauss In this study we used the information retrieval technique, and the techniques of document clustering and information extraction to analyze the selected Plans. We used term frequency within the information retrieval. The term frequency shows how often a term, word or concept occurs in a document. It also indicates the significance of a determined term (or concept) within the overall document. Here, we focus on the 6 (six) areas corresponding to the actions addressed in the selected plans. Each area is evaluated by examining the CAPs/CRPs in searching the frequency of key policy terms. According to the classification of this study, the most frequent key terms are sustainability, greenhouse gas emissions (GHGs) and responsi- bility terms followed by climate change, renewable energy and international documents (agreements). Each Plan illustrates the need for more responsibility that signifies the importance of education, research and awareness to climate change and sustainability policies (see Fig. 1 below). As seen in Table 4 above, the key policy term of the first category is linked to the following six sub-terms: Climate, Climate Change, Climate Planning, Campus Climate Planning, Campus Planning and University Climate Planning. For this category, Fig. 1 shows that all six selected Universities’ Plans have a comprehensive approach to reduce GHGs emissions within the context of campus energy transformation and sustainability when comparing other categories. The U-M Carbon Neutral Report is the plan that has the highest term frequency value of the category. The number of uses of this term in the plan is determined to be 129. The second is GVSU (87), the third is WMU (63). These plans are followed by MSU (34) and Kalamazoo College (32). No climate change or related key terms have been identified in the WSU Sustainability Strategic Plan. 1000 800 MSU 600 400 GVSU 200 WSU 0 U-M U-M MSU WMU WSU Kalamazoo Fig. 1. Comparative Evaluation of Policies of Selected Universities’ Climate-Related Plans (Michigan) The second key policy term is greenhouse gas emissions which includes the sub- terms greenhouse gas emissions, greenhouse and GHG. It presents a few correlations relating to climate change, sustainable development, renewable energy, green infras- tructure, campus transportation, research, education and awareness. MSU keeps its first Climate Change and Sustainable Campus Planning 309 position for this category as the plan having the highest frequency value which is 177 uses. The MSU Plan is followed by U-M (110), WMU (110), GVSU (98), Kalamazoo College (91), and WSU (21). Sustainability is the third policy key term and includes the following sub-terms: sus- tainable, sustainability, sustainable future, sustainable campus, campus sustainability, sustainable development, sustainable development goals, SDGs and sustainable univer- sity. For this category U-M ranks first with a total of 665 uses. These terms are used 169 times in the WSU Plan and 150 times in WMU Plan. These are followed by MSU (112), Kalamazoo College (80) and GVSU (66). The key term for the fourth category is renewable energy including two sub-terms renewable energy and renewable. MSU Energy Transition Plan has the highest number (295) among the selected plans. MSU Plan’s focus is almost to develop a reliable energy system on campus. This is followed by U-M (84), GVSU (27), Kalamazoo College (20), WMU (6) and WSU (6). The key policy term for the fifth category is responsibility which includes the sub- terms education, research, community, society, awareness, university, justice, environ- mental justice. It is also noticeable that the U-M Plan ranks first place for this policy, with the highest term frequency value which is determined to be 977 uses. The second is MSU Plan (526), the third is WSU (164), WMU (159), and these plans are followed by GVSU (105) and Kalamazoo College (48), respectively. International documents or agreements are the key policy term of the sixth cate- gory. In this category, the analysis is based on five sub-terms: Agreement, IPCC, United Nations, UNFCCC, and Climate Change Framework. These terms have been chosen by considering the leading actors in the international system regarding climate change and sustainability. The terms are not commonly used in each plan, so the key terms and sub- terms of this category are the lowest term frequency value. For example, in the WSU Plan, the terms have not been determined. This may be a result of approaches based mainly on campus sustainability. U-M still maintains its rank for this category with the highest value which is determined 44. The other CRPs are listed for the term frequency value as follows: WMU (6), MSU (5), GVSU (4), Kalamazoo College (2). 5 Conclusion Understanding the roles of universities by examining CAPs/CRPs is important to deter- mine the possibilities of developing climate planning focused on the sustainability of the campus. The intersection of sustainability and higher education in the United States has deep roots dating back to the 1972 United Nations Conference on the Human Envi- ronment which issued the Stockholm Declaration. With the evolution of the field came recognition that higher education institutions were both contributing to the problem and well-positioned to develop solutions. Climate-related plans are developed by most uni- versities to support sustainability and mitigate GHGs since the beginning of 2000s, and still some universities are developing new ones. The paper conceptualized the university as having four principal modalities on the axis of the relationship between the UNFCCC principles and the UN SDGs: 1. Edu- cation & Research 2. Community Outreach & Sustainability, 3. Campus Operations & 310 A. Buckley and E. J. Strauss Planning, 4. Goals for future generations. All plans are examined by these modalities to identify their status and components. We present an analysis of selected six Michigan universities which develop a climate- related plan (CAPs, Sustainability Plan, Energy Plan or similar one). The evaluated plans are listed as follows: (1) Michigan State University Energy Transition Plan, (2) Grand Valley State University Climate Action Plan, (3) University of Michigan President’s Commission on Carbon Neutrality Report, (4) Western Michigan University Climate Action Plan, (5) Wayne State Sustainability Strategic Plan and (6) Kalamazoo College Sustainability and Climate Action Plan. Within the context of the UNFCCC principles and sustainability the text-based anal- ysis is carried out in six categories covering climate change, GHGs, sustainability, renew- able energy, responsibility level and international agreements. With this analyze selected Michigan Universities’ plans were evaluated regarding the link between climate plan- ning and sustainability. The Plans were also analyzed on the key policy terms identifying climate-related concepts regarding text-based analysis. To measure the content of CAPs we used the term frequency technique for each plan, and we attempted to reveal a con- ceptual density map with the results obtained from the key policy terms and their closest sub-terms. Thus, this paper tried to reveal the relationship between climate change and sustainability in selected six Michigan Universities’ Plans. Responses of the Plans to climate change and sustainability indicate that the plans have been formally compiled through a university board, council or similar institu- tions. Based on the responses from these Plans, it was determined that most often climate-related plans cover, or are planned to contain: GHG emissions reductions at all areas, renewable energy investments, developing education and research programs, campus planning regarding sustainability, engaging with communities at all levels, and infrastructure investment and/or design requirements; monitoring and evaluation and/or investigation of co-benefit opportunities with emissions reduction as future goal. References 1. Wals, A.E., Blewitt, J.: Third-wave sustainability in higher education: some (inter)national trends and developments. In: Jones, P., Selby, D., Sterling, S. (eds.) Sustainability Education: Perspectives and Practices Across Higher Education, pp. 55–74. Earthscan, New York (2010) 2. Alghamdi, A., et al.: Benchmarking of water, energy, and carbon flows in academic buildings: a fuzzy clustering approach. Sustainability 12(11), 4422 (2020) 3. UNFCCC (United Nations Framework Convention on Climate Change). UN Climate Convention. https://unfccc.int/resource/docs/convkp/conveng.pdf. Accessed 12 Dec 2021 4. Web of science - WoS (2022). https://apps.webofknowledge.com/summary.do?pro duct=WOS&doc=1&qid=1&SID=E34HkWtST9pX7RhnmQD&search_mode=Advanc edSearch&update_back2search_link_param=yes 5. Bullock, G., Wilder, N.: The comprehensiveness of competing higher education sustainability assessments. Int. J. Sustain. High. Educ. 17(3), 282–304 (2016) 6. Disterheft, A., et al.: Environmental Management Systems (EMS) implementation pro- cesses and practices in European higher education institutions–top-down versus participatory approaches. J. Clean. Prod. 31, 80–90 (2012) 7. UI Green Metric. UI Green Metric Guidelines 2021. https://greenmetric.ui.ac.id/publications/ guidelines/2021. Accessed 24 Jan 2022 Climate Change and Sustainable Campus Planning 311 8. Hunter, R., et al.: Environmental, health, wellbeing, social and equity effects of urban green space interventions: a metanarrative evidence synthesis. Environ. Int. 130, 104923 (2019). https://doi.org/10.1016/j.envint.2019.104923 9. Kuh, K.F.: The law of climate change mitigation: an overview. In: Encyclopedia of the Anthropocene, vol. 2, pp. 505–510 (2018). https://doi.org/10.1016/B978-0-12-809665-9.100 27-8 10. Lozano, R., Lukman, R., Lozano, F.J., Huisingh, D., Lambrechts, W.: Declarations for sus- tainability in higher education: becoming better leaders, through addressing the university system. J. Clean. Prod. 48, 10–19 (2013) 11. Koç, H.E.: Environmental Sustainability of University Campuses: A Practical Assessment Tool. Master of Science Middle East Technical University (2014) 12. UN Rio Declaration. https://legal.un.org/avl/historicarchives.html. Accessed 14 Oct 2021 13. Tan, H., Chen, S., Shi, Q., Wang, L.: Development of green campus in China. J. Clean. Prod. 64, 646–653 (2014) 14. Wals, A.E.J.: Sustainability in higher education in the context of the UN DESD: a review of learning and institutionalization processed. J. Clean. Prod. 62, 8–15. https://www.sciencedi rect.com/science/article/pii/S0959652613003880. Accessed 4 Feb 2022 15. UNESCO: Shaping the Future We Want: UN Decade of Education for Sustainable Develop- ment (2005–2014) Final Report. https://sustainabledevelopment.un.org/content/documents/ 1682Shaping%20the%20future%20we%20want.pdf. Accessed 8 Jan 2022 16. Alshuwaikhat, H.M., Abubakar, I.: An integrated approach to achieving campus sustainability: assessment of the current campus environmental management practices. J. Clean. Prod. 16, 1777–1785 (2008) 17. Arroyo, P.A.: New taxonomy for examining the multi-role of campus sustainability assess- ments in organizational change. J. Clean. Prod. 140(Part.3), 1763–1774 (2017) 18. UN 2030 Agenda for Sustainable Development and its 17 Sustainable Development Goals. www.un.org/sustainabledevelopment. Accessed 22 Feb 2022 19. UNESCO. Issues and trends in Education for Sustainable Development. In: Leicht, A., Heiss, J., Byun, W.J. (eds.). UNESCO, Paris, France. https://unesdoc.unesco.org/ark:/48223/pf0000 261445. Accessed 4 Mar 2022 20. MSU (Michigan State University). “The Office for Sustainability” (About). https://sustainab ility.msu.edu/about/index.html. Accessed 20 Feb 2022 21. MSU Sustainability Report. https://sustainability.msu.edu/. Accessed 12 May 2021 22. MSU Energy Transition Plan. https://ipf.msu.edu/energy-transition-plan. Accessed 12 Nov 2021 23. U-M Carbon Neutrality Report (PCCN Report). https://president.umich.edu/initiatives-and- focus-areas/environment-sustainability-and-carbon-neutrality/. Accessed 1 Feb 2022 24. U-M (University of Michigan): Planet Blue Campus. https://planetblue.umich.edu/campus/ about/historic-milestones/. Accessed 20 Feb 2022 25. The U-M Sustainability Goals. https://ocs.umich.edu/sustainability-goals/. Accessed 12 Jan 2022 26. GVSU (Grand Valley State University) Climate Action Plan. https://www.gvsu.edu/sustainab ilityguide/sustainability-issues-energy-10.htm. Accessed 14 Nov 2022 27. WMU Climate Action Plan. https://wmich.edu/sustainability/reports-publications/office-sus tainability-publications. Accessed 10 Feb 2022 28. WMU (Western Michigan University). PUSC Strategic Sustainability Report. https://wmich. edu/sustainability/reports-publications/office-sustainability-publications. Accessed 30 Jan 2022 29. UN SDGs (Sustainable Development Goals): Sustainable Development Goals Partnerships Platform. https://sustainabledevelopment.un.org/partnership/?p=2375. Accessed 2 Jan 2022 312 A. Buckley and E. J. Strauss 30. WSU (Wayne State University) (2017–2022): WSU Sustainability Strategic Plan (About). https://sustainability.wayne.edu/about/strategic-plan. Accessed 14 Dec 2021 31. KC-Kalamazoo College Carbon Report. https://sustainability.kzoo.edu/kalamazoo-college- 2018-2019-carbon-report/. Accessed 22 Feb 2022 32. Kalamazoo College Sustainability and Climate Action Plan. https://sustainability.kzoo.edu/ initiatives/climate-action-plan/. Accessed 14 Oct 2021 33. Gupta, V., Lehal, G.S.: A survey of text mining techniques and applications. J. Emerg. Technol. Web Intell. 1(1), 60–76 (2009) 34. Talib, R., Hanif, M.K., Ayesha, S., Fatima, F.: Text mining: techniques, applications and issues. Int. J. Adv. Comput. Sci. Appl. 7(11), 414–418 (2016). https://doi.org/10.14569/IJA CSA.2016.071153 35. He, W.: Examining students’ online interaction in a live video streaming environment using data mining and text mining. Comput. Hum. Behav. 29(1), 90–102 (2013) 36. Sagayam, R., Srinivasan, S., Roshni, S.: A survey of text mining: retrieval, extraction and indexing techniques. Int. J. Comput. Eng. Res. 2(5), 1443–1446 (2012) 37. Dang, D.S., Ahmad, P.H.: A review of text mining techniques associated with various application areas. Int. J. Sci. Res. (IJSR) 4(2), 2461–2466 (2015) 38. Weiss, S.M., Indurkhya, N., Zhang, T., Damerau, F.: Text Mining: Predictive Methods for Analyzing Unstructured Information. Springer, New York (2005). https://doi.org/10.1007/ 978-0-387-34555-0 39. Agrawal, R., Batra, M.: A detailed study on text mining techniques. Int. J. Soft Comput. Eng. (IJSCE) 2(6), 2231–2307 (2013) 40. Fan, W., Wallace, L., Rich, S., Zhang, Z.: Tapping the power of text mining. Commun. ACM 49(9), 76–82 (2006) Synthesis of a Novel pH-Responsive Emulsifier Based on Dynamic Covalent Bond and Its Application in Reversible Oil-Based Drilling Fluids Guoshuai Wang1,2 , Guancheng Jiang1,2(B) , Jun Yang1,2 , Yinbo He1,2 , and Yue Fu1,2 1 College of Petroleum Engineering, China University of Petroleum (Beijing), Changping District, Beijing 102249, China

2 State Key Laboratory of Petroleum Resources and Prospecting, China University of Petroleum (Beijing), Changping District, Beijing 102249, China Abstract. Owing to their advantages in the preparation of stimuli-responsive emulsifiers and emulsions, dynamic covalent bonds have recently attracted great interest. Removal of traditional emulsifier-stabilized oil-based drilling fluids (OBDFs) is difficult after drilling operations. To address this issue, a pH-responsive emulsifier (AT–FA) based on dynamic imine bonding between 1-aminotetradecane (AT) and 4-formylbenzoic acid (FA) for reversible OBDFs is prepared. Fourier transform infrared spectroscopy (FT-IR) and proton nuclear magnetic resonance (1 H NMR) confirm the presence of dynamic covalent imine bonds and the molecu- lar structure of AT–FA and characterize them. AT–FA could effectively reduce the oil–water interfacial tension at alkaline pH (7.6), at which AT–FA is a long-chain alkyl carboxylate-type anionic emulsifier that forms highly stable water-in-oil microemulsions. However, at acidic pH (4), its dynamic imine bond is cleaved and converted to a cationic emulsifier in the form of amine salts to stabilize oil-in- water (O/W) microemulsions. More importantly, this phase inversion behavior is reversible. Finally, reversible OBDFs are successfully developed using AT–FA as a pH-responsive emulsifier, which exhibited reasonable rheological and filtration properties and excellent emulsification stability. Since it is converted to a water- soluble O/W drilling fluid at pH 4, its removal rate is as high as 98.46%. This pH-responsive emulsifier based on dynamic imine bonds can not only ensure the emulsification stability of OBDFs, but also facilitate the later removal of OBDFs, showing great application potential in the drilling industry. Keywords: Dynamic covalent bond · pH-responsive emulsifier · Oil-based drilling fluids · Oily mud removal 1 Introduction Owing to their excellent temperature resistance, lubricity, inhibition, and anti-pollution ability, oil-based drilling fluids (OBDFs) have become an important technical means © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 313–323, 2023. https://doi.org/10.1007/978-981-19-3983-9_27 314 G. Wang et al. for drilling deep and ultra-deep wells, highly deviated wells, shale horizontal wells, and complex formations [1]. In the petroleum drilling industry, OBDFs are in high demand due to the continuous exploration and development of unconventional fields such as ultra-deep, deep water, and shale in recent years [2]. However, after the completion of drilling operations, removing the emulsified mud from the surfaces of the wellbore, casing, and drilling cuttings is very difficult, which will seriously affect the cementing quality, increase post-treatment costs, and even cause ecological pollution [3]. Over the past decades, to solve the above problem, researchers have mainly focused on devel- oping additional technical means, such as chemical washes, solvent extraction, thermal desorption, and microbial degradation [4–6]. However, these methods have disadvan- tages such as poor universality, complex operation, unsatisfactory efficiency, and higher processing cost [7]. OBDFs are essentially water-in-oil emulsions stabilized by various emulsifiers and their excellent emulsion stability and non-aqueous nature are the keys to their remarkable drilling performance. These properties are also the root cause of the difficulty in removing OBDFs [3, 8]. Inspired by the mechanism of intelligent regulation of emulsion type by stimuli-responsive emulsifiers, some researchers have used pH-responsive emulsifiers instead of traditional emulsifiers to develop reversible OBDFs to solve the removal problems [9, 10]. Reversible OBDFs can be switched, if required, between water-in- oil (W/O) and oil-in-water (O/W) emulsion drilling fluids. They have the excellent performance of conventional OBDFs in the drilling stage and the easy cleaning advantage of water-based drilling fluids in the cementing and completion stage. When compared to other removal techniques, this pH-triggered emulsion phase inversion strategy is cheaper and highly efficient in solving removal problems. The pH-responsive emulsifiers currently used in reversible OBDFs are mainly tertiary amine surfactants, which are seldom available, their preparation process is complex, and their synthesis cost is high [11, 12], which seriously affects the development and application of reversible OBDFs. Dynamic covalent properties of imine bonds provide a facile route to develop pH- responsive emulsifiers due to the simple preparation method, wide availability of syn- thetic raw materials, and fast response characteristics [13]. Dynamic imine bonds refer to the spontaneous formation of imine bonds between primary amino molecules and aldehyde molecules under an alkaline environment, but rapid cleavage under an acidic environment [14]. This approach is mainly applied in the preparation of pH-responsive O/W emulsions. The study of pH-responsive W/O type OBDFs based on dynamic imine bonds has not been documented yet. In this study, we report a novel pH-responsive emul- sifier (AT–FA) based on dynamic imine bonds and use it to prepare reversible OBDFs to solve the disadvantages of traditional OBDFs. We also investigate its interfacial activity, emulsifying ability, stimuli-responsive mechanism, and application effect. 2 Materials and Methods 2.1 Materials 1-aminotetradecane (AT, 96%), 4-formylbenzoic acid (FA, 97%), calcium oxide (CaO, 98%), Span-80 (99%), and calcium chloride (CaCl2 , 98%) were purchased from Anhui Zesheng Technology Co., Ltd (China). Sodium hydroxide (NaOH, 99%), ethanol Synthesis of a Novel pH-Responsive Emulsifier 315 (99.9%), and acetic acid (99%) were obtained from Shanghai Aladdin Biochemical Technology Co., Ltd (China). 5# White oil, barite (weighting agent), organoclay (tack- ifier), styrene butyl acrylate copolymer (SBAC, filtrate reducer), and lecithin (wetting agent) were kindly offered by Beijing Shida Bocheng Technology Co., Ltd (China). 2.2 Synthesis of AT–FA AT–FA was synthesized using the Schiff base reaction [11]. AT and FA were added to a beaker containing ethanol at a molar ratio of 1:1 and reacted for 1 h under magnetic stirring at 1700 rpm. The product was purified by a rotary evaporator. 2.3 Characterization of AT–FA A Nicolet IS 10 Fourier transform infrared spectroscopy (FT-IR) analyzer (Thermo Fisher Scientific, USA) was used to identify the functional groups of AT–FA, and the FT- IR spectra were recorded in the wavenumber range 400–4000 cm−1 . A Bruker AVANCE III 600 M nuclear magnetic resonance (NMR) apparatus (Bruker Instruments, Germany) was used to characterize the molecular structure of AT–FA. The sample was diluted in CD3 OD and its 1 H chemical shifts were recorded. The interfacial tensions (IFTs) of emulsifiers at the mineral oil–water interface were measured at 70 °C by an OCA25 tensiometer (DataPhysics Instruments, Germany) using the pendant drop method. 2.4 Preparation and Characterization of Emulsions Different mass fractions (wt%, based on the total volume of oil and water) of AT–FA were added to 210 mL of 5# white oil and stirred vigorously at 8000 rpm for 20 min for the dispersion and dissolution of AT–FA. Subsequently, 1.75 g of CaO and 140 mL of deionized water were added and stirred at 10000 rpm for 20 min to obtain the emulsions. The pH value of the emulsion samples was adjusted using 40 wt% acetic acid solu- tions and NaOH and was measured by a PB-10 digital pH meter (Shanghai Nuoxuan Scientific Instrument Co., Ltd., China). The demulsification voltage (ES) of emulsions was determined by a DWY-2 electrical stability tester (Tongchun Analytical Instruments, China). The microscopic morphology of emulsions was observed by a DM4 M optical microscope (Leica, Germany), the particle size distributions were calculated by Nano Measurer software. The conductivity of the emulsions was determined by an AR8011 conductivity test pen (Shandong Annai Automation Instrument Co., Ltd. China). 2.5 Determination of Rheological and Filtration Properties The rheological properties of the drilling fluid samples were determined by a ZNN- D6B six-speed viscometer (Tongchun Analytical Instruments, China). The rheological parameters such as the apparent viscosity (AV), plastic viscosity (PV), and yield point (YP) were calculated based on the American Petroleum Institute (API) test guidelines. The high-temperature and high-pressure filtration volumes (FLHTHP ) of the drilling fluid samples were measured by a GGS42-2A filtration meter (Tongchun Analytical Instru- ments, China) at 120 °C and 3.5 MPa. The aging experiments of the drilling fluid samples were carried out in a BGRL-5 roller oven (Tongchun Analytical Instruments, China). 316 G. Wang et al. 2.6 Removal Rate Tests The removal rate of emulsified mud was measured using a six-speed viscometer at 50 °C. First, a clean, constant-weight stainless steel sleeve was immersed in the drilling fluid sample for 1 h. Then, the sleeve was suspended for 2 h, and the unadhered emulsified mud was removed by gravity. Next, the sleeve was mounted on the viscometer and washed with aqueous solutions of different pH values for 10 min at 300 rpm. After washing, the removal rate of emulsified mud is calculated using Eq. 1. w1 − w2 Removal rate = × 100% (1) w1 − w0 where W 0 is the weight of the clean sleeve, W 1 is the weight of the sleeve adhered with emulsified mud, and W 2 is the weight of the sleeve after washing. 3 Results and Discussion 3.1 Characterization of AT-FA Figure 1a shows the FT-IR spectra of FA, in which the broad peak at 1675 cm−1 was attributed to the stretching vibration of the carbonyl bond (C = O) in the carboxyl group and aldehyde group (−CHO) [13]. For AT, the double peaks at 3331 and 3255 cm−1 were ascribed to the stretching vibration of the primary amine group (−NH2 ) [15]. For AT–FA, the C=O peak narrowed, the –NH2 peak flattened, and a new peak appeared at 1644 cm−1 , indicating that FA and AT were involved in the Schiff base reaction and formed a dynamic imine bond (–CH=N–) [16]. Figure 1b shows the 1 H NMR spectra of AT–FA, in which the proton signal (8.42 ppm) in the –CH=N– bond is observed [11], and the molecular structure of AT–FA is also given. These results proved the formation of the dynamic imine bond between AT and FA. 3.2 Evaluation of Interface Activity of AT–FA The presence of lipophilic alkane chains and hydrophilic carboxyl groups made the small-molecule AT–FA amphiphilic, so it forms a directional adsorption layer at the oil– water interface. However, for emulsifiers to form stable emulsions, excellent interfacial activity is a necessary attribute [13]. Figure 2 shows the IFTs of AT, FA, and AT–FA at the mineral oil–water interface measured at alkaline pH (7.6). The IFT of AT–FA decreased rapidly with increasing concentration and has the largest decrease among the three. For instance, at 0.4 wt%, the IFTs of FA, AT, and AT–FA are 22.15, 8.18, and 0.12 mN/m, respectively. AT–FA has the highest interfacial activity. Furthermore, these results support the idea that the interfacial activity originated from dynamic imine bonding products AT-FA rather than AT or FA. Synthesis of a Novel pH-Responsive Emulsifier 317 (a) FA AT AT-FA C=O 1675 −NH2 3331䠈㻟㻞㻡㻡 −CH=N− 1644 4000 3500 3000 2500 2000 1500 1000 500 Wavenumber (cm-1) (b) f f f f ff d f b c g f f ff e f a b c e -CH=N- g a bc d 12 10 8 6 4 2 0 δ (ppm) Fig. 1. FT-IR spectra (a) of AT, FA, and AT-FA and 1 H NMR spectra (b) of AT–FA. 3.3 Evaluation of Emulsifying Ability Table 1 presents the ES values, emulsification efficiency, and average droplet size of the W/O emulsions emulsified by different concentrations of AT–FA at pH 7.6, while the corresponding microstructure observed by optical microscopy and particle size distribu- tion are shown in Fig. 3. It can be seen from Table 1 that the emulsifying ability of AT–FA is strongly dependent on concentration. With increasing concentration of AT–FA, the ES value increases rapidly and the average droplet size decreases sharply, indicating the gradual enhancement of the emulsion stability. At 5 wt%, it has the highest ES value (1031 V) and the smallest average droplet size (2.05 µm). The emulsification efficiency is always 100%, and no phase separation is observed after 48 h at any concentration. These 318 G. Wang et al. 25 FA 22.15 AT 21.12 20 Interfacial tension (mN/m) AT-FA 18.26 15.36 15 14.3 12.42 11.9 10 9.43 8.18 5.23 5 0.37 0.27 0.12 0 0 0.1 0.2 0.3 0.4 Concentration (wt%) Fig. 2. IFTs of FA, AT, and AT–FA at the mineral oil–water interface. results indicate that AT–FA has excellent emulsifying ability and possesses the potential as an emulsifier for OBDFs. This is because AT–FA forms divalent metal soap with calcium oxide (CaO), a higher fatty acid calcium, reducing the repulsive force between polar head groups. Hence, AT–FA is densely arranged at the oil–water interface, form- ing a densely packed emulsifier layer, thereby giving the emulsion high anti-coalescence stability [2]. Table 1. Emulsification properties of different concentrations of AT–FA Sample ES (V) Emulsification efficiency (%) Average droplet size (µm) 1 wt% AT–FA 98 100 8.42 2 wt% AT–FA 241 100 5.37 3 wt% AT–FA 613 100 3.86 4 wt% AT–FA 976 100 2.71 5 wt% AT–FA 1031 100 2.05 Synthesis of a Novel pH-Responsive Emulsifier 319 Fig. 3. Optical microscopic images, particle size distribution, and appearance photographs of the W/O emulsions at various concentrations of AT–FA. Optical micrographs and appearance photographs were collected after 48 h of storage. 3.4 pH-Responsive W/O Emulsions with Reversible Phase Inversion Behavior Due to the presence of dynamic covalent bonds (imine bonds), the molecular structure of AT–FA undergoes reversible changes under acid–base stimulation and hence AT–FA- stabilized emulsions achieve reversible phase inversion. To investigate the pH-responsive behavior, a W/O emulsion stabilized by 5 wt% AT–FA was selected. As shown in Table 2, when the pH was adjusted to 4 by adding acetic acid aqueous solution, the ES value of the original W/O emulsion decreased to 0 V, the conductivity value increased to 75.9 µS/cm, and the emulsion droplets became water-soluble. At the same time, regular micron-sized droplets (1–11 µm) were observed in the optical micrograph (Fig. 4). These changes indicate the formation of an O/W emulsion. NaOH was then added to the acid-triggered O/W emulsion to adjust the pH back to 7.6. After stirring for 20 min, the ES value of the emulsion returned to 791 V, the conductivity value returned to 0, and the emulsion droplets became water-insoluble, indicating the reformation of a W/O emulsion. As expected, similar droplets and particle size distributions to the original W/O emulsions were observed (Fig. 4). These similar parameters indicate that the stability of the alkaline-triggered W/O emulsion is close to the original W/O emulsion. Therefore, the AT–FA-stabilized W/O emulsions exhibit reversible phase inversion behavior, and AT–FA can be used to formulate reversible OBDFs. 3.5 Application in Reversible OBDFs We successfully developed reversible OBDFs using AT–FA as the pH-responsive emul- sifier, and Table 3 shows their basic properties. There is no significant change in the values of AV, PV, YP, and ES of reversible OBDFs before and after aging at 120 °C, which indicates that the drilling fluid has good thermal stability. The ES value of the aged drilling fluid is as high as 675 V, but its FLHTHP value is only 2.0 mL, which indicates its high emulsification stability and excellent filtration control ability. More importantly, 320 G. Wang et al. Table 2. ES values, conductivity values, water solubility, and emulsion type of emulsions stabilized by 5 wt% AT–FA at different pH pH ES (V) Conductivity (µS/cm) Water solubility Emulsion type 7.6 (Original state) 1031 0 Insoluble W/O 4 (Acidic-triggered) 0 75.9 Soluble O/W 7.6 (Alkaline-triggered) 791 0 Insoluble W/O Fig. 4. Optical micrograph images (48 h after stimulation) and particle size distributions of emulsions stabilized by 5 wt% AT–FA at different pH. the drilling fluid is changed to O/W type under acid stimulation (pH 4) and returned to W/O type under alkali stimulation (pH 8). In addition, the performance parameters of the drilling fluid before and after the phase inversion meet the actual drilling operation. The cleaning properties of reversible OBDFs based on the pH-responsive emulsifier AT–FA are further evaluated by removal rate tests. For comparison, we fabricate conven- tional OBDFs using the traditional emulsifier Span-80, with other additives remaining unchanged. As shown in Fig. 5, at acidic pH (4), the conventional OBDFs adhered to the sleeve was difficult to clean, and the removal rate was only 29.90%. However, reversible OBDFs were easily desorbed, and the removal rate was as high as 98.46%. These results demonstrated that the reversible OBDFs based on the pH-responsive emulsifier AT–FA not only possess reasonable rheological and filtration properties and excellent emulsion stability, but also have the advantage of easy removal. It showed promising application prospects in drilling engineering. 3.6 Analysis of the pH-Responsive Mechanism Figure 6 shows the 1 H NMR spectra of AT–FA before and after acid–base stimulation, revealing its pH-responsive mechanism. We infer from Fig. 6 that when AT–FA is treated with acetic acid, the –CH=N– bond peak at 8.42 ppm disappeared, whereas the −CHO group peak appeared at 10.06 ppm, which suggests that the dynamic imine bond in AT– FA broke and dissociated into separate FA and AT. Interestingly, after the mixture reacted with acetic acid was treated with NaOH again, the –CH=N– bond peak at 8.42 ppm reappeared, while the −CHO group peak disappeared at 10.06 ppm, which indicates that the formation and cleavage of dynamic imine bonds are reversible. A reversible phase inversion mechanism is proposed from the above results. Under an alkaline environment (pH 7.6), AT–FA is an anionic surfactant similar to higher Synthesis of a Novel pH-Responsive Emulsifier 321 Table 3. Rheological, emulsification, and filtration properties of reversible OBDFs. Aging time AV PV YP pH Drilling fluid type ES FLHTHP (h) (mPa·s) (mPa·s) (Pa) (V) (mL) 0 53.5 44 9.5 8 W/O 683 / 16 56 46 10 8 W/O 675 2.0 39.5 34 5.5 4 O/W 0 9.2 60.5 50 10.5 8 W/O 515 3.0 Note: The formula of reversible OBDFs: 210 mL 5# white oil + 5 wt% AT-FA + 0.5 wt% CaO + 1 wt% organoclay + 2 wt% SBAC + 1 wt% lecithin + 140 mL brine (25 wt% CaCl2 ) + barite, the density of drilling fluid is 1.4 g/cm3 Fig. 5. Cleaning effects at pH 4 of reversible OBDFs stabilized by AT–FA (a) and conventional OBDFs stabilized by Span-80 (b). AT-FA AT-FA/H+ AT-FA/H+/OH- -CH=N- -CHO 12 10 8 6 4 2 0 δ (ppm) Fig. 6. 1 H NMR spectra of AT–FA before and after acid–base stimulation. fatty acids, which can form divalent metal soaps with calcium ions [17]. Due to its excellent interfacial activity and strong lipophilicity of divalent metal soaps, the emul- sifier layer protrudes toward the oil phase, thus forming a stable W/O emulsion. How- ever, under an acidic environment (pH 4), the dynamic imine bond in AT–FA breaks 322 G. Wang et al. and AT is protonated to cationic form AT+ [13]. Because cationic AT+ , like traditional cationic surfactants, can effectively reduce the oil–water IFT and because of its strong hydrophilicity, the emulsifier layer is convex toward the water phase, thereby forming a stable O/W emulsion. Due to the reversibility of the dynamic imine bond, when the alkaline environment is restored, AT–FA is reformed, and stable W/O emulsions are obtained. 4 Summary A pH-responsive emulsifier (AT–FA) based on dynamic covalent bonds was prepared and used to develop reversible OBDFs to address the difficulty in emulsified mud removal. The small-molecule AT–FA exhibited excellent interfacial activity and emulsification stability at alkaline pH (7.6). Due to the presence of dynamic imine bonds in the molec- ular structure, AT–FA can be used to prepare pH-responsive reversible W/O emulsions. At pH 7.6, the AT–FA-stabilized emulsions were a highly stable W/O microemulsion. However, when the pH was adjusted to 4, the W/O microemulsion was converted to an O/W microemulsion. The reversible phase inversion behavior of emulsions depends on the mechanism of dynamic imine bond formation under alkaline environments and decomposition under acidic environments. Finally, reversible OBDFs were formulated using the pH-responsive emulsifier AT–FA and achieved acid-induced high emulsified mud removal efficiency. Acknowledgements. This work was supported by the National Natural Science Foundation of China (grant numbers 51874329, 52004297, 51991361) and the China Postdoctoral Innovative Talent Support Program (BX20200384). References 1. Lichinga, K.N., Maagi, M.T., Wang, Q., Hao, H., Gu, J.: Experimental study on oil based mudcake removal and enhancement of shear bond strength at cement-formation interface. J. Pet. Sci. Eng. 176, 754–761 (2019) 2. Fayad, M.M., Mohamed, D.E., Soliman, E.A., Abd El-Fattah, M., Ibrahim, S., Dardir, M.M.: Optimization of invert emulsion oil-based drilling fluids performance through heterocyclic imidazoline-based emulsifiers. Colloid Surf. A-Physicochem. Eng. Asp. 613, 126092 (2021) 3. Ren, Y., et al.: CO2 /CaO-responsive behavior and application potential of amine emulsion. Pet. Explor. Dev. 48(5), 1014–1022 (2021) 4. Ball, A.S., Stewart, R.J., Schliephake, K.: A review of the current options for the treatment and safe disposal of drill cuttings. Waste Manag. Res. 30(5), 457–473 (2012) 5. Ren, H.Y., Zhou, S.Y., Wang, B., Peng, L., Li, X.G.: Treatment mechanism of sludge contain- ing highly viscous heavy oil using biosurfactant. Colloid Surf. A-Physicochem. Eng. Asp. 585, 24117 (2020) 6. Li, Z., Zhou, Y., Meng, X., Wang, S.: Harmless and efficient treatment of oily drilling cuttings. J. Pet. Sci. Eng. 202, 108542 (2021) 7. Meng, R., Wang, C., Shen, Z.: Optimization and characterization of highly stable nanoemul- sion for effective oil-based drilling fluid removal. SPE J. 25, 1259–1271 (2020) Synthesis of a Novel pH-Responsive Emulsifier 323 8. Yan, J.N., Menezes, J.L., Sharma, M.M.: Wettability alteration caused by oil-based muds and mud components. SPE Drill. Complet. 8, 35–44 (1993) 9. Ren, Y., Jiang, G., Li, F., Zhou, H., An, Y.: Cleanup characteristics and mechanisms of reversible invert emulsion drilling fluid. Colloid Surf. A-Physicochem. Eng. Asp. 133, 296– 303 (2015) 10. Zhang, R., Zhu, X., Cai, Y.: The phase transformation mechanism of bentonite-stabilized and cetyltrimethylammonium bromide-stabilized emulsions and application in reversible emulsification oil-based drilling fluids. J. Surf. Deterg. 22, 525–534 (2019) 11. Ren, G., et al.: pH-responsive nanoemulsions based on a dynamic covalent surfactant. Nanomaterials 11, 1390 (2021) 12. Chu, Z., Feng, Y.: Vegetable-derived long-chain surfactants synthesized via a “green” route. ACS Sustain. Chem. Eng. 1, 75–79 (2013) 13. Zhu, T., et al.: Fabrication of a pH-responsive emulsifier for heavy oil recovery based on dynamic imine bond. J. Mol. Liq. 332, 115916 (2021) 14. Zentner, C.A., Anson, F., Thayumanavan, S., Swager, T.M.: Dynamic imine chemistry at complex double emulsion interfaces. J. Am. Chem. Soc. 141, 18048–18055 (2019) 15. Zhang, Y., Lu, H., Wang, B., Wang, N., Liu, D.: pH-responsive non-pickering emulsion stabilized by dynamic covalent bond surfactants and nano-SiO2 particles. Langmuir 36, 15230–15239 (2020) 16. Li, M., et al.: Fluid state transition mechanism of a ternary component aqueous solution based on dynamic covalent bond. J. Mol. Liq. 332, 115849 (2021) 17. Akanni, M.S., Burrows, H.D.: Surface tensions of some molten divalent metal carboxylates. Thermochim. Acta 115, 351–357 (1987) Removal of Aqueous Cu2+ by NaCl Modified 5A Zeolite Jiantao Yang1(B) and Jianwen Wei2 1 College of Forestry, Henan Agricultural University, Zhengzhou 450002, China

2 Guilin University of Technology, Guilin 541004, China Abstract. In this work zeolite is activated by NaCl aqueous solution, and then it was used to adsorb Cu2+ . When the initial Cu2+ concentration is 63.5 mg/L, 5A modified by 0.5 M NaCl may be the good adsorbent whose removal rate is 82.44% and adsorption quantity is 52.3 mg/g. It is expected that NaCl can improve the adsorption capacity of copper ion which is usually not effectively removed. Then effective and low-cost adsorbents are expected to be developed and used to remove the heavy metals in wastewater. Keywords: Adsorption · Cu2+ · 5A zeolite first section 1 Introduction A number of heavy metals have been found in wastewater from industrial applications, such as electroplating, pesticides, medicine, mining and painting. These pollutants can enter human body through food chain and accumulate in the human body, in the end cause serious harm to the health of human beings [1, 2]. Therefore, efficient removal of heavy metals from aqueous solutions has been drawing an increasing number of attentions. Cu2+ is a kind of common hazardous pollutant and one of the most widespread heavy metal contaminants in the environment. Cu2+ is released from several sources of industries such as printed circuit board production, metallurgical, smelting and refining, paints and pigments, and municipal [3]. However, high concentrations Cu2+ in drinking water can cause vomiting, abdominal pain, nausea, diarrhea, liver and kidney damage, chronic disorders, cramps in the calves and gastrointestinal catarrh [4]. Therefore, it is crucial to remove Cu2+ from wastewaters effectively before discharge into the water bodies. Zeolites have been well developed and have generated considerable interest because they exhibit many superior physical properties such as high specific surface areas and porosity, narrow size distributions and easy to modify the inner surface [5]. But the virgin zeolite shows poor adsorption capacity to remove the aqueous heavy metals. So, it needs modification to improve its removal capacity of heavy metals [6]. There are many methods to modify it, including high temperature activation, alkali activation [7], acid activation [8], inorganic salt activation [9] and organic activation [10]. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 324–329, 2023. https://doi.org/10.1007/978-981-19-3983-9_28 Removal of Aqueous Cu2+ by NaCl Modified 5A Zeolite 325 In this work zeolite is activated by NaCl aqueous solution, and then it will be used to adsorb Cu2+ . It will be expected that NaCl will improve the adsorption capacity of copper ion which is usually not effectively removed. Then effective and low-cost adsorbents are expected to be developed and used to remove the heavy metals in wastewater. 2 Experimental Work 2.1 Preparation of the Adsorbents First, the adsorbent needs to be prepared, and then the adsorption experiments will be carried out. To study the effects of the factors, such as adsorption time, and initial concentrations of heavy metal ions on the adsorbents, chemicals and instruments, such as Atomic Absorption Spectroscopy needed are listed in Table 1 and Table 2. Table 1. Containers or instruments used in the experiments Materials Specification Quantity 1 AAS 1 2 Conical flask(stopcock) 150 ml 5 3 Beaker 500 ml 1 4 Measuring cylinder flask 100 ml 1 6 Plastic wash bottle 250 ml 1 8 Reagent bottles 500 ml 1 9 Reagent bottles 1000 ml 1 10 Glass rod 2 11 Volumetric flask 10 ml 10 12 Volumetric flask 100 ml 5 13 Volumetric flask 1000 ml 1 14 Pipette 1 ml 1 15 pipette 5 ml 1 16 Analytical balance 0.1 mg 1 17 Centrifugal tube 10 ml 20 18 Filter F50 mm 1 package 19 Cotton mask 1 20 Refrigerator 1 21 Pan paper 1 22 Label 1 23 Centrifugal tube rack 50 holes 1 326 J. Yang and J. Wei Table 2. Reagents used in the experiments Reagents Specification Quantity CAS 1 Powder Zeolite Aldrich 250 g 12173-98-7 2 Deionized water 3 Cu(NO3 )2 .3H2 O AR 250 g 10031-43-3 4 NaCl AR 250 g 7647-14-5 5 HNO3 AR 500 ml 7697-37-2 10 g of the 5A zeolites were added to 200 mL of 0.1, 0.3, or 0.5 mol/L NaCl solution, respectively, and was stirred for 22 h at 80–90 °C. After filtration, the zeolites were washed with deionized water for three times, and then put in an oven at 80 °C. The zeolites were placed in a muffle furnace after drying. Subsequently, the zeolites were heated up to 400 °C with a heating rate of 3–5 °C/min and roasted for 4 h. As cooling to room temperature, the NaCl reactivated zeolites were kept in reagent bottles for use. 2.2 Adsorption Experiments Please note that the first paragraph of a section or subsection is not indented. The first paragraphs that follows a table, figure, equation etc. does not have an indent, either. Subsequent paragraphs, however, are indented. Batch adsorption experiment was performed in the 150 mL conical flasks containing 100 mL of single metal nitrate solution and 100 mg of adsorbent stirring with magnetic stirrer at room temperature of 22 °C. A stock solution of Cu2+ was prepared by dis- solving an exact amount of Cu (NO3 )2 .3H2 O in deionized water. Adsorbent–solution mixtures were stirred for 2 h or 4 h and then were filtered to remove the solids. Metal concentrations, both in the initial and in the final solutions, were analyzed by an atomic absorption spectrometer. The removal efficiency of the metal ions (R%) was calculated by the following equation: R = (C0 − Ce )/C0 × 100%. (1) where C 0 and C e are the initial and equilibrium concentrations of Cu2+ in mg/L, respectively. Results for the metal ions uptake are calculated by the following equation: qe = (C0 − Ce ) × V /m. (2) where qe is the equilibrium adsorption capacity of the adsorbent in mmol · g−1 , C 0 is the initial concentration in mg/L, C e is the concentration at equilibrium of metal ions in mg/L, V is the volume in L of metal ions solution and m is the weight in g of the adsorbent. 3 Results and Discussions As seen in Fig. 1 and Fig. 2, when the C0 is 10 mg/L, all the removal rates of the modified 5A zeolites are superior to 90% and the original 5A is 89.01% which showed Removal of Aqueous Cu2+ by NaCl Modified 5A Zeolite 327 that the adsorbents can removal the low aqueous copper effectively. But their adsorbent quantities are all below 10 mg/g. When the C0 is 63.5 mg/L, all the removal rates of the adsorbents decreased, but 5A modified by 0.5 M NaCl still showed relatively high removal rate of 82.44% and their adsorption quantities increased to 35.6, 36.5, 39.6 and 52.3 mg/g. When the C0 was increased to even higher, 127 mg/L, all the removal rates of the adsorbents decreased sharply to smaller than 40%, but 5A modified by 0.5 M NaCl still showed relatively high removal rate of 38.38%, in the four adsorbents. At this initial Cu2+ concentration, their adsorption quantities increased to higher than 40 mg/g except the 0.1 M modified 5A, and 5A modified by 0.5 M NaCl showed highest adsorption quantity of 48.7 mg/g. Fig. 1. Removal rates of 5A and 5A modified by a series of NaCl concentrations Because there is Na residual after the adsorbents were rinsed three times with 100 ml of deionized water, showing that the adsorbents need more rinse. When the initial Cu2+ concentration is lower than 10 mg/L, copper ion can be effectively removed by all the adsorbents. However, due to the Na pollution, 5A is the most effective one in the four adsorbents. When the initial Cu2+ concentration is 63.5 mg/L, 5A modified by 0.5 M NaCl may be the good adsorbent whose removal rate is 82.44% and adsorption quantity is 52.3 mg/g. 328 J. Yang and J. Wei Fig. 2. Adsorption quantities of 5A and 5A modified by a series of NaCl concentrations Therefore, compared to the adsorption quantity of organically modified mesoporous silicas, which is about 35 mg/g at 25 °C, our adsorbent has higher adsorption quantity and can be produced easily and economically. 4 Conclusion 5A zeolite modified by different concentrations of NaCl (0.1 M, 0.3 M and 0.5 M) was used to adsorb Cu2+ in wastewater. When the initial concentration of Cu2+ was 63.5 mg/L, the 5A zeolite modified with 0.5 M NaCl exhibited the highest Cu2+ adsorption capacity, with a removal rate of 82.44% and an adsorption capacity of 52.3 mg/g. The 5A zeolite modified by NaCl showed excellent Cu2+ adsorption performance, which is expected to provide possibility for industrial application. Acknowledgements. The authors sincerely acknowledge the support of National Natural Science Foundation of China (No. 51966002) and Natural Science Foundation of Guangxi Province (No. 2020GXNSFAA159144). References 1. Lofrano, G., et al.: Polymer functionalized nanocomposites for metals removal from water and wastewater: an overview. Water Res. 92, 22–37 (2016) Removal of Aqueous Cu2+ by NaCl Modified 5A Zeolite 329 2. Jin, Y., Wu, S., Zeng, Z., Fu, Z.: Effects of environmental pollutants on gut microbiota. Environ. Pollut. 222, 1–9 (2017) 3. Bilal, M., et al.: Waste biomass adsorbents for copper removal from industrial wastewater-a review. J. Hazard. Mater. 263, 322–333 (2013) 4. Yu, S., et al.: Rational design of carbonaceous nanofiber/Ni-Al layered double hydroxide nanocomposites for high-efficiency removal of heavy metals from aqueous solutions. Environ. Pollut. 242, 1–11 (2018) 5. Sun, J., Liu, L., Yang, F.: A WO3/PPy/ACF modified electrode in electrochemical system for simultaneous removal of heavy metal ion Cu2+ and organic acid. J. Hazard. Mater. 394, 122534 (2020) 6. Delkash, M., Bakhshayesh, B.E., Kazemian, H.: Using zeolitic adsorbents to cleanup special wastewater streams: a review. Microporous Mesoporous Mater. 214, 224–241 (2015) 7. Sannino, F., Ruocco, S., Marocco, A., Esposito, S., Pansin, M.: Cyclic process of simazine removal from waters by adsorption on zeolite H-Y and its regeneration by thermal treatment. J. Hazard. Mater. 229, 354–360 (2012) 8. Liu, J., Cheng, X., Zhang, Y., Wang, X., Zou, Q., Fu, L.: Zeolite modification for adsorptive removal of nitrite from aqueous solutions. Microporous Mesoporous Mater. 252, 179–187 (2017) 9. Soetardji, J.P., et al.: Ammonia removal from water using sodium hydroxide modified zeolite mordenite. RSC Adv. 5(102), 83689–83699 (2015) 10. Peng, Z., et al.: Heavy metal leachability in soil amended with zeolite- or biochar-modified contaminated sediment. Environ. Monit. Assess. 190(12), 1–13 (2018). https://doi.org/10. 1007/s10661-018-7124-2 11. Jimenez-Castaneda, M.E., Medina, D.I.: Use of surfactant-modified zeolites and clays for the removal of heavy metals from water. Water 9(4), 235–246 (2017) Author Index A G Abbood, Mohammed, 138 Gao, Ling, 212 Al Slik, Ghada, 138 Awadhya, Gokaran Prasad, 190 H He, Yinbo, 313 Hermawan, Andreas Aditya, 268 B Huang, Linsheng, 258 Baowen, Yang, 172 Bhat, Sajad Ahmed, 15 J Bhatt, Anita, 15 Jayatharan, Mayuran, 268 bin Hassim, Mohd Mawardi, 49 Jha, Aayush, 190 Buckley, AyseOzcan, 289 Jiang, Guancheng, 313 C K Cao, Hui, 87 Ke, Yin, 163 Cao, Zheng Bo, 229 Chen, Yit-Jin, 98 L Chen, Yiwen, 128 Laveti, Suneelkumar, 98 Chrysanidis, Theodoros, 149 Li, Hao, 87 Cruz, Lovely Jasmin Dela, 234 Li, Peitong, 128 Cui, Chenxing, 3, 245 Li, Wei, 229 Li, Weiwen, 58 Li, Xiaohua, 73 D Liu, Bifeng, 108 Dangzalan, Ondrea Nhika, 234 Liu, Jie, 58 Dong, Yunpeng, 163 Liu, Jinliang, 3, 245 Liu, Ran, 3, 245 Liu, Yuchi, 128 E Lv, Gao, 221 Esguerra, Ezekiel, 234 M F Mendonça, Paulo, 281 Feng, Ming Yue, 212 Monjardin, Cris Edward, 234 Fu, Yue, 313 Mosca, C., 120 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 E. Strauss (Ed.): ICOCE 2022, LNCE 276, pp. 331–332, 2023. https://doi.org/10.1007/978-981-19-3983-9 332 Author Index P V Paglia, C., 120 Vasudevan, Gunalaan, 49 Papadimitriou, Georgios, 149 Vieira, Clara, 281 Q W Qian, Changzhao, 108 Wang, Guoshuai, 313 Wei, Jianwen, 324 R Wenpo, Wang, 172 Rossi, Pier Paolo, 30 Wu, Meizhong, 58 S X Saingam, Panumas, 38 Xingcan, Wei, 172 Shaoli, Qi, 172 Sharma, Sanidhya, 190 Y Song, Li, 3, 245 Yang, Jiantao, 324 Spinella, Nino, 30 Yang, Jun, 313 Strauss, Eric J., 289 Yang, Xu, 58 Ye, Luo, 221 T Yu, Xiyang, 73 Talei, Amin, 268 Tang, Shiying, 58 Z Thakur, Mrinal, 190 Zahidi, Izni, 268 Topacio, Anjerick, 98 Zhang, Hong Qiang, 212, 229 Zhang, Jiahao, 58 U Zhang, Zhenjun, 221 Uthayasirppi, Kribadharan, 49 Zhibin, Zhong, 172