(PDF) A Data-Driven Approach for Generator Load Prediction in Shipboard Microgrid: The Chemical Tanker Case Study

energies Article A Data-Driven Approach for Generator Load Prediction in Shipboard Microgrid: The Chemical Tanker Case Study Tayfun Uyanık 1,2 , Nur Najihah Abu Bakar 2 , Özcan Kalenderli 3 , Yasin Arslanoğlu 1 , Josep M. Guerrero 2 and Abderezak Lashab 2, * 1 Maritime Faculty, Istanbul Technical University, 4469 Istanbul, Türkiye 2 Center for Research on Microgrids, AAU Energy, 9220 Aalborg, Denmark 3 Faculty of Electrical and Electronics Engineering, Istanbul Technical University, 4469 Istanbul, Türkiye * Correspondence:

Abstract: Energy efficiency and operational safety practices on ships have gained more importance due to the rules set by the International Maritime Organization in recent years. While approximately 70% of the fuel consumed on a commercial ship is utilized for the propulsion load, a significant portion of the remaining fuel is consumed by the auxiliary generators responsible for the ship’s onboard load. It is crucial to comprehend the impact of the electrical load on the ship’s generators, as it significantly assists maritime operators in strategic energy planning to minimize the chance of unexpected electrical breakdowns during operation. However, an appropriate handling mechanism is required when there are massive datasets and varied input data involved. Thus, this study implements data-driven approaches to estimate the load of a chemical tanker ship’s generator using a 1000-day real dataset. Two case studies were performed, namely, single load prediction for each generator and total load prediction for all generators. The prediction results show that for the single generator load prediction of DG1, DG2, and DG3, the decision tree model encountered the least errors for MAE (0.2364, 0.1306, and 0.1532), RMSE (0.2455, 0.2069, and 0.2182), and MAPE (17.493, 5.1139, and 7.7481). In contrast, the deep neural network outperforms all other prediction models in the case of total generation prediction, with values of 1.0866, 2.6049, and 14.728 for MAE, RMSE, and MAPE, respectively. Citation: Uyanık, T.; Bakar, N.N.A.; Keywords: data-driven; generator load prediction; maritime; shipboard microgrid Kalenderli, Ö.; Arslanoğlu, Y.; Guerrero, J.M.; Lashab, A. A Data-Driven Approach for Generator Load Prediction in Shipboard Microgrid: The Chemical Tanker 1. Introduction Case Study. Energies 2023, 16, 5092. Energy efficiency is becoming more crucial for commercial ship operations due to https://doi.org/10.3390/en16135092 rising concerns about fuel price fluctuations and strict emission regulations by the IMO Academic Editor: Ying-Yi Hong (International Maritime Organization) [1]. Accordingly, the process of incorporating tech- nological advancements into the maritime industry has accelerated [2]. Ship transportation, Received: 15 February 2023 in particular, is rapidly evolving toward full electrification, with the creation of what are Revised: 5 May 2023 known as all-electric ships (AES). Innovations in electric propulsion offer flexibility during Accepted: 8 June 2023 voyages where the emissions from diesel combustion can be kept under control [3]. Due Published: 30 June 2023 to the incorporation of the microgrid concept into shipboard power systems, AES are regarded as part of the maritime microgrid. They feature components that are similar to a typical microgrid, where the generators transmit electricity via an energy line to power the Copyright: © 2023 by the authors. propulsion and the onboard loads. AES act as mobile microgrids during voyage operations Licensee MDPI, Basel, Switzerland. at sea, where they are considered to be in the stand-alone mode [4]. During this mode, the This article is an open access article entire onboard load and ship propulsion are powered by their onboard power system. distributed under the terms and The main engine, which is the fundamental part of the propulsion system in a ship, conditions of the Creative Commons consumes a large amount of fuel during voyage operations. Additionally, the other essential Attribution (CC BY) license (https:// devices in the ship that must be considered when it comes to fuel consumption include creativecommons.org/licenses/by/ pumps and auxiliary machinery. Generators are the most vital component in the ship’s oper- 4.0/). ational system because they generate electricity and supply energy to the whole ship. Thus, Energies 2023, 16, 5092. https://doi.org/10.3390/en16135092 https://www.mdpi.com/journal/energies Energies 2023, 16, 5092 2 of 20 generators are regarded as the highest fuel consumers during ship voyage operations [5]. Knowing how much power will be used during the voyage and in various operational activities is extremely important for the long-term sustainability and energy efficiency of ship systems. Power interruptions during onboard operations due to generator failures may raise the risk of ship malfunctions and result in more serious mishaps. Furthermore, if sudden loads cannot be foreseen, collapses may occur on the generator side. Strategies to prevent breakdowns can be developed by understanding the individual behaviour of the generator and loads, and enable estimation of optimal fuel consumption [6–8]. For this reason, it is necessary to perform highly accurate load generator predictions [9–11]. From this perspective, this paper aims to predict the load of a chemical tanker in voyage operations in two scenarios, namely, single load prediction and total load prediction. However, to obtain a highly accurate prediction, input selection must be treated as the critical part of the forecasting model. The best way to imitate the actual power value, regardless of the conditions, is by considering the actual measurements of the ship’s main components. Accordingly, this study preselects the input variables from the main engine, diesel generator, and boiler, which are considered the main components in shipboard power. Additionally, to ensure that predictions can be practically used and reflect the actual power consumed, the usage of real data from chemical tanker ships is necessary. Thus, the data utilized in the study should be collected from real ship voyage operations and not generated by simulations. Existing research forecasting ship power, such as the study in [12], considers environmental disturbances such as the wind (speed, direction), waves (height, direction), and position displacement, which react with weather impacts as input data. Despite the fact that this benefits the dynamic positioning of the ship, it does not represent the whole usage of power during its journey at sea. In voyage operations, the sea condition (calm and rough sea) alone does not represent the actual usage of the power. The actual power reading depends on various uncertain factors, such as dynamic consumer behaviour, ship length, carriage capacity, voyage duration, and speed of steaming. This understanding is supported by another recent publication [13] that considers another set of inputs for power prediction, such as latitude, roll angle, net propulsion power, and speed. This scenario justifies considering a diversity of inputs that may be involved in the power usage of a ship. None of them focus on the ship exclusively, but rather focus on the specific operations, as in the case study in [12], which specifically leveraged forecasting outputs to tune the ship’s position by using the outputs for system control. This research gap highlighted the significance of the input selection conducted in this study. A great challenge lies in how to interpret the relationship between the input data and the prediction output. A good understanding of how inputs and outputs relate to each other assists in the development of a reliable forecasting algorithm with the fewest errors and deviations. Among the preliminary assumptions for the input for load generator prediction used in this study are the speed and running hours of the main engines, the consumption and running hours of the diesel generator, temperature, boiler usage, and freshwater generator readings. To ensure a clear rationale behind all the initial assumptions, these pre-assumptions for the input-output relationship in the prediction need to be verified. However, the bulk set of data and its varying values increase the complexity and necessitate a careful approach to data handling. Despite the barrier of complexity, a large amount of data is necessary for the prediction model of the algorithm to understand the hidden relationships between data for the output prediction. This is because the prediction training draws on all the insight patterns of past data. The more data used, the more accurate the prediction model. In order to find the right approach to handling large volumes of data, the research conducted in [3] highlighted that the data-driven approach is a powerful technique for uncovering meaningful patterns, complicated relationships, and correlation variables among massive amounts of data with the help of artificial intelligence (AI). Recently, used models in data-driven research, include support vector machine, multiple-linear regression, artificial neural network, deep neural network, k-nearest neighbour, random forest, extreme gradient boosting, and decision tree algorithms, which were broadly applied Energies 2023, 16, 5092 3 of 20 in the maritime sector, showing promising results. This trend can be seen in recent research publications estimating and optimizing ship speeds with machine learning algorithms to determine energy efficiency [14], collision avoidance, effective manoeuvring using AIS (automatic identification system) data for autonomous vessels [15], vessel traffic modelling and implementation via weather and historical data [16], fuel oil consumption prediction using voyage data [17–20], object detection for vessel safety [21], and shaft power prediction for determining ship energy efficiency [22]. Another study where a container ship was examined with data-driven techniques proves that the models can provide significant advantages in estimating fuel consumption and determining propeller performance [23]. The proven effective data-driven approaches to maritime applications discussed in the literature could also provide a good prediction model for chemical tanker load generators. This justifies the selection of a data-driven strategy for this study. Accordingly, this paper presents a data-driven approach to the load generator predic- tion of a chemical tanker ship with the following contributions: • First, this paper attempts to perform load generator prediction for a chemical tanker in two case studies, namely, single load generator prediction for each generator in the ship and total load generator prediction for the total load. The prediction for the single load generator will assist the ship operator with pre-planning for an unexpected event due to the real issues raised by one malfunctioning generator during a voyage. Meanwhile, total load generator prediction can prevent large operational failures that might cause severe disruptions of the voyage and big losses to shipping lines. • Second, to address the uncertain number of influence factors that affect the actual reading of the power generator, this paper provides a correlation analysis that demon- strates the strong relationship between the chosen inputs and the varying patterns of output prediction. To make this approach realistic and applicable to new data pre- dictions, it is integrated using 1000 days of a real voyage dataset instead of a dataset generated through software simulation. • Third, eight different data-driven algorithms were tested in this study, including support vector machine, multiple-linear regression, artificial neural network, deep neural network, k-nearest neighbour, random forest, extreme gradient boosting, and decision tree algorithms. Comparative performance analysis of these approaches will provide the best forecasting model for the ship’s load generator of a chemical tanker ship with minimum error. This model can be used in energy scheduling and power planning to optimize system operations. The other parts of the study are organized as follows: in Section 2, data driven approaches in the maritime sector are discussed; in Section 3, the prediction methodology is explained; in Section 4, the important findings in the simulation study are discussed; and conclusions of the study are explained in Section 5. 2. Data-Driven Approaches in the Maritime Industry In recent years, data-driven methods have been applied successfully in various dis- ciplines, and their scheme has been widely applied in solving various problems. In the assessment of process safety, approximately 500 studies conducted between 1990 and 2020 were sampled [24]. Recent data-driven approaches in the maritime field generally focus on the assessment of ship performance and management from data collected during voyages [25]. There are several studies in the literature that conduct research on the estimation technique for the ship’s main generator. For instance, considerable work shows that machine learning is effective in fuel consumption estimation, and its output precision can be increased by hyperparameter adjustment [26]. According to a study that highlighted the role of the maritime sector in global emissions, data-driven methodologies could be used to predict the fuel consumption of dry cargo ships [27]. Artificial neural networks were used in a study dealing with estimating the fuel consumption of a bulk carrier [18]. A study mentioned that fuel consumption estimation with data-driven methods is an important Energies 2023, 16, 5092 4 of 20 measure to achieve better energy efficiency and the sustainability of maritime transport. It also could help calculate voyage costs [28]. Meanwhile, in a case study that involved photovoltaic energy for a hybrid ship microgrid, various machine learning algorithms were used for energy estimation, but the scores obtained at the end of the study were not satisfactory enough [4]. ANN, MLR, decision tree, random for- est, and XG Boost algorithms were used to solve the ship berthing for cold ironing problem [3]. Shaft speed predictions have been made with data-driven approaches [12]. Nine different algorithms were used in a study to estimate a container ship’s main engine shaft power and fuel consumption. However, this study contains some innovative aspects. Classical processes such as fuel consumption estimation were also carried out in the mentioned study [13]. These methods are useful for the maritime field problems. In contrast to the literature, marine diesel generator power prediction and total load estimation of the chemical tanker vessel electrical microgrid are performed in this paper. 3. Prediction Methodology This study estimated marine diesel generator power using support vector machine, multiple-linear regression, artificial neural network, deep neural network, k-nearest neigh- bour, random forest, extreme gradient boosting, and decision tree algorithms. Approxi- mately 1000 days of the voyage dataset were obtained from a commercial chemical tanker ship built in 2017. This study’s dataset is divided into two parts: training and test data. The first of these parts is the training data, consisting of 750 samples, and the other part is the test data, consisting of 250 samples. Then, the dataset was processed by purifying the missing samples using data preprocessing techniques, and it was then converted into a format that algorithms could operate with. This dataset includes information about the main engine, auxiliary machines, and diesel generators from various sensors on the ship. Pearson correlation was used to examine the correlations in the dataset, and a pair plot was used to investigate the data frames with high correlation. A separate pair plot illustrated the correlation between the data frames with high correlation between the dataset and the total load on the diesel generators. The simulation results were digitized with the error metrics mean absolute error (MAE), root mean square error (RMSE), and mean absolute percentage error (MAPE). As a result of the simulations, the loads on the marine diesel generators during the voyage were successfully estimated with data-driven algorithms. In addition, a separate simulation has been successfully carried out to estimate the total load on the system. The k-fold cross-validation method was used to verify the process of the results. The methodology of the study outlined above is visualized in Figure 1. In this ship microgrid, diesel generators are the energy generators, while, hotel loads, alarms, pumps, sensors, and boilers, are the energy consumers. Figure 2 shows the general structure of the chemical tanker ship microgrid used in this case study. Three diesel generators power the entire shipload, while a fourth emergency generator is available in case of a main generator failure. 3.1. Data Processing Data collection is the fundamental process in data-driven studies [29]. A solid and clean dataset can also facilitate the achievement of the targeted results in the study [30]. Initially, in this study, a 1461-day voyage dataset was gathered for investigation. Then, the sections of this dataset that had null values were eliminated, resulting in 1000 days of refined data that could be used in this study. The ship’s characteristics and the brief statistical information of the dataset are given in Tables 1 and 2, respectively. The data features used in this study are as follows: • Main engine speed (revolutions per minute (rpm)). • Diesel generator fuel consumption (t/day), power (kW). • Main engine running hours (h/day). • Diesel generator running hours (h/day). • Main engine output maximum continuous rating (%). Energies 2023, 16, 5092 5 of 20 • Main engine power (kW), fuel consumption (t/day). • Fresh water generator running hours (h/day). • Energies 2023, 16, x FOR PEER REVIEW Scavenge air temperature (◦ C), scavenge pressure (bar). 5 of 22 • Main engine exhaust temperature (◦ C). • Boiler running hours (h/day). Figure 1. Methodology of the study. Figure 1. Methodology of the study. In this ship microgrid, diesel generators are the energy generators, while, hotel loads, alarms, pumps, sensors, and boilers, are the energy consumers. Figure 2 shows the general Energies 2023, 16, x FOR PEER REVIEW 6 of 22 Energies 2023, 16, 5092 structure of the chemical tanker ship microgrid used in this case study. Three diesel gen- 6 of 20 erators power the entire shipload, while a fourth emergency generator is available in case of a main generator failure. Figure Figure 2. The 2. The generalstructure general structure of of the theinvestigated investigatedchemical tanker chemical shipship tanker microgrid. microgrid. 3.1.1.Data Table Processing Vessel specifications. Data collection is the fundamental process in data-driven studies [29]. A solid and Item Specification clean dataset can also facilitate the achievement of the targeted results in the study [30]. Initially, in this Vessel study,Type a 1461-day voyage dataset was gathered Oil/Chemical Tanker Then, for investigation. the sections ofGross Tonnage this dataset 29,590 tin 1000 days of that had null values were eliminated, resulting refined data Deadweight that could Tonnage 49,990 be used in this study. The ship’s characteristics t the brief sta- and Length/Breadth 183/32.3 tistical information of the dataset are given in Table 1 and Table 2, respectively. m The data Year Built features used in this study are as follows: 2017 Average/Maximum Speed 13.4/17.3 knots • Main engine speed (revolutions per minute (rpm)). Main Engine Power 7000 kW • Diesel generator Generator fuel consumption (t/day), power (kW). Power/Count 580 kW/3 • Main engine runningDraft hours (h/day). 8m • Diesel generator running hours (h/day). • Main engine output maximum continuous rating (%). Table• 2. Summary Main engine of the dataset. power (kW), fuel consumption (t/day). • Fresh water generator running hours (h/day). Main Engine Generator•FuelScavenge Mainair temperature Engine Fuel (°C), scavenge pressureGenerator Main Engine (bar). 1 Generator Generator 3 ... Power 2 Power Speed (rpm) • (t/d) Consumption Main Consumption engine exhaust (t/d)temperature Power (°C). (kW) (kW) (kW) Power (kW) • Boiler running hours (h/day). Mean 79.46 1.45 9.95 ... 4169.8 175.33 267.24 281.27 Std 16.24 1.03 8.44 ... 1559.2 244.77 248.57 253.12 Min 0 0.01 0 ... 0 0 0 0.04 Max 98.11 4.5 26.18 ... 7000.0 580.00 580.00 580.00 In data-driven studies, correlation analysis explains the relationship between the data and determines the variables. The Pearson correlation coefficient (r) was used to determine the relationship between two variables in the dataset. The correlation coefficient takes values between −1 and +1. Figure 3 shows the Pearson correlation matrix of the dataset. Here, the “+” sign in front of the correlation coefficient indicates a positive correlation Energies 2023, 16, 5092 7 of 20 between the two variables and the “−” sign indicates a negative correlation between the two variables. For example, in this dataset, diesel generators’ fuel consumption and Energies 2023, 16, x FOR PEER REVIEW 8 ofthe 22 main engine fuel consumption values are strongly correlated (0.9), while diesel generator 1 running hours and the scavenge air temperature value correlation value is 0. Figure 3. Pearson Figure 3. Pearson correlation correlation matrix matrix of of the the dataset. dataset. The pair plot The pair plot method methodcan canenable enablea amore more detailed detailed examination examination of the of the correlation correlation be- between tween the variables in the dataset. The pair plot of the variables with a high correlation the variables in the dataset. The pair plot of the variables with a high correlation coefficient in this study is shown in Figure 4. The relationship between these variables and coefficient in this study is shown in Figure 4. The relationship between these variables and the total load of the generators is also shown in Figure 5. While the main engine and power the total load of the generators is also shown in Figure 5. While the main engine and power variables show a strong correlation for all values, other variables cannot form such a linear variables show a strong correlation for all values, other variables cannot form such a linear graph. When Figure 5 is examined, it can be seen that the total load of the diesel generator graph. When Figure 5 is examined, it can be seen that the total load of the diesel generator shows a correlation within a specific range with the scavenge pressure variable. shows a correlation within a specific range with the scavenge pressure variable. Energies 2023, 16, Energies 2023, 16, 5092 x FOR PEER REVIEW 98 of of 20 22 Figure 4. Pair plot of the highly correlated variables. Figure 4. Pair plot of the highly correlated variables. SHAP values (SHapley Additive exPlanations 3.7.7) follow a method that is based on game theory and enables the determination of the influence of inputs on the output variables in the dataset [31,32]. The SHAP values that were discovered quantified the importance of the inputs in determining the system’s output in this prediction model, and they are illustrated in Figure 6. Figure 6A shows the effect of input variables on the output while estimating the power of diesel generator 1. Figure 6B,C show how input variables affect output for generators 2 and 3, respectively. Meanwhile, Figure 6D shows the effect of the input variables on the output in the model used to estimate the total load of the generators. For the DG1 power prediction process (Figure 6A), it can be said that DG2 power, DG1 running hours, and DG3 power are more important than the other features in the dataset. Figure 6B shows that DG1 power, DG2 running hours, and DG3 power values are more important features for the DG2 power prediction phase. For the DG3 power prediction, DG3 running hours, DG2 power, and DG1 power values are more important than the other features (Figure 6C), and important values are presented in the Figure 6D for the total load prediction case. Energies 2023, 16, Energies 2023, 16, 5092 x FOR PEER REVIEW 109 of of 20 22 Energies 2023, 16, x FOR PEER REVIEW 11 of 22 Figure 5. Figure 5. Pair Pair plot plot of of the the total total load load of of generators generators and and highly highly correlated correlated variables. variables. SHAP values (SHapley Additive exPlanations 3.7.7) follow a method that is based on game theory and enables the determination of the influence of inputs on the output vari- ables in the dataset [31,32]. The SHAP values that were discovered quantified the im- portance of the inputs in determining the system’s output in this prediction model, and they are illustrated in Figure 6. Figure 6A shows the effect of input variables on the output while estimating the power of diesel generator 1. Figure 6B,C show how input variables affect output for generators 2 and 3, respectively. Meanwhile, Figure 6D shows the effect of the input variables on the output in the model used to estimate the total load of the generators. For the DG1 power prediction process (Figure 6A), it can be said that DG2 power, DG1 running hours, and DG3 power are more important than the other features in the dataset. Figure 6B shows that DG1 power, DG2 running hours, and DG3 power values are more important features for the DG2 power prediction phase. For the DG3 power prediction, DG3 running hours, DG2 power, and DG1 power values are more im- portant than the other features (Figure 6C), and important values are presented in the Figure 6D for the total load prediction case. Figure6.6.Feature Figure Featureimportance importance values values for formodel modeloutput: output:(A) DG1 (A) DG1power prediction, power (B) (B) prediction, DG2DG2 power power prediction, (C) DG3 power prediction, (D) total load prediction. prediction, (C) DG3 power prediction, (D) total load prediction. 3.2. Estimation Models 3.2.1. Support Vector Machines (SVM) The support vector machine (SVM), developed by AT&T Bell Labs, is a supervised learning method frequently used in classification and regression problems. This algorithm Energies 2023, 16, 5092 10 of 20 3.2. Estimation Models 3.2.1. Support Vector Machines (SVM) The support vector machine (SVM), developed by AT&T Bell Labs, is a supervised learning method frequently used in classification and regression problems. This algorithm has an effective usage area for accuracy, proper generalization, and precision theories [33]. In this method, the predicted results that fall within a given range are considered successful, whereas those that fall outside of that range are considered failed. Vectors that limit this range value are called support vectors [34]. The model of the SVM is given below. f ( x ) = w T x + w0 (1) N λ H (w, w0 ) = ∑i=1 (y i − f (x i )) + 2 kwk2 (2) 0 i f |r | < ε Vε (r ) = (3) |r | − ε otherwise Here, w is the normal vector, x is an independent variable, λ is the regularisation parameter, w0 is coefficient, V is the error function,ε is the error margin, and r is the error [35]. Figure 7 describes the components of the support vector machine. The blue dots represent successful predictions within the permissible margin borders (support vectors), and it can be seen in the figure that the red dots show unsuccessful results or another Energies 2023, 16, x FOR PEER REVIEW 12 of 22 cluster [19]. Figure7.7.Support Figure Supportvector vectormachine. machine. 3.2.2. 3.2.2.Multiple MultipleLinear LinearRegression Regression(MLR) (MLR) The The multiple linear regression(MLR) multiple linear regression (MLR)method methodisisaastatistical statisticaltechnique techniquethat thatconsists consists ofofdependent dependent and independent variables. In this method, dependent variables cancan and independent variables. In this method, dependent variables be be cal- calculated with culated with thethe help help of more of more than than oneone independent independent variable variable and and coefficient coefficient [36,37]. [36,37]. Mul- Multiple linear regression can be expressed by Equation tiple linear regression can be expressed by Equation (4), (4), y =𝑦 a=0 + 𝑎 a+ 1 x𝑎1 + 𝑥 .+ . .⋯ ++an𝑎xn𝑥+ + ε Ɛ (4) (4) where 𝑎 , 𝑎 , … , 𝑎 are coefficients, y is the dependent variable, Ɛ is the model error, and 𝑥 , 𝑥 , … , 𝑥 are independent variables. In this method, 𝑎 (coefficients) are calculated us- ing Equation (5). ∑ ( ∑ ) 𝑎 = 𝑎𝑟𝑔𝑚𝑖𝑛( ) (5) 3.2.3. Artificial Neural Network (ANN) Energies 2023, 16, 5092 11 of 20 where a0 , a1 , . . . , an are coefficients, y is the dependent variable, ε is the model error, and x1 , x2 , . . . , xn are independent variables. In this method, an (coefficients) are calculated using Equation (5). (∑in=1 (yi − a0 −∑nj=1 a j xij )2 ) an = argmin(a) (5) 3.2.3. Artificial Neural Network (ANN) The artificial neural network method, which imitates the working process of the brain, is a structure consisting of artificial neurons and nodes [38]. In this structure, neurons receive and process the signal and transmit it to other neurons with which it is connected. The signal formed in this structure is the sum of the signals coming from the connected neurons [34]. Other considerations in artificial neural networks are the weight variables and the activation function [39]. The weight variable affects the value of the output function by increasing or decreasing the strength of a signal in the neural network. The activation function, on the other hand, determines how the output signal, which is determined by the weights, will be transferred from a certain layer of the network to another layer or the output of the network. A typical artificial neural network is represented in Figure 8. When the figure is examined, I values represent the input values of the network, while X values form the input layer. Here, h values create the hidden layer, while A values create Energies 2023, 16, x FOR PEER REVIEW 13 of 22 the output layer. O values represent the output values of the network. In addition, the w values in the figure represent the weights [39–42]. Figure 8. A typical ANN structure. Figure 8. A typical ANN structure. 3.2.4.Deep 3.2.4. DeepNeural NeuralNetwork Network(DNN) (DNN) AAdeep deepneural neuralnetwork networkcan canbebedefined definedasasananadvanced advancedANN ANNstructure structure[43,44]. [43,44].This This method has the same weight and activation functions as an ANN structure, method has the same weight and activation functions as an ANN structure, but it differs in but it differs in number the the number of hidden of hidden layers. layers. Inway, In this this much way, much more parameter more parameter optimization optimization can be can be made inmade in the the deep deepnetwork neural neural network structure,structure, and more and more successful successful results results can can be obtained be obtained [45]. A [45]. ADNN typical typical DNN structure structure is shown is inshown Figure in FigureFigure 9. When 9. When9 isFigure 9 is examined, examined, I values I values form the form the inputs, inputs,form x values x values formlayer, the input the input layer,form w values w values form theand the weights, weights, andform h values h values the form the hidden hidden layer. layer. Values formValues form the the output output layer, layer, while while Oare O values values are thevalues the output outputof values the system [46,47]. [46,47]. of the system method has the same weight and activation functions as an ANN structure, but it differs in the number of hidden layers. In this way, much more parameter optimization can be made in the deep neural network structure, and more successful results can be obtained [45]. A typical DNN structure is shown in Figure 9. When Figure 9 is examined, I values form the inputs, x values form the input layer, w values form the weights, and h values Energies 2023, 16, 5092 12 of 20 form the hidden layer. Values form the output layer, while O values are the output values of the system [46,47]. Figure9.9.AAtypical Figure typicalDNN DNNstructure. structure. 3.2.5.K-Nearest 3.2.5. K-NearestNeighbours Neighbours(KNN) (KNN) The Thek-nearest k-nearest neighbours algorithm algorithmisisbased basedononthe the principle principle of determining of determining the the dis- distance of k-nearest neighbours to a certain point [48]. Here, the k parameter may tance of k-nearest neighbours to a certain point [48]. Here, the k parameter may vary ac- vary according cording totothe themodel model [49,50]. [49,50]. While While making making predictions with thethe KNN KNN algorithm, algorithm,the the model modelcontinues continuestraining by by training making makinguse of usepast of data. past Minkowski distance data. Minkowski (Lp ), an (L distance impor- p), an tant parameter used in this algorithm, is used for the specification of the margin between a start point (xq ) and any other point (xj ) (Equation (6)). 1 p p L p x j , xq = ∑i | x j,i − xq,i | (6) Here, if p = 1 is the Manhattan distance, p = 2 is the Euclidean distance [51]. 3.2.6. Random Forest (RF) In the random forest (RF) algorithm, an advanced form of the decision tree algorithm, the results are obtained through multiple decision trees [52]. When used for classification purposes, the output produced by the most trees is considered the final result. In contrast, the average results obtained from all of the trees in regression problems give the final result [53]. A typical RF structure is presented in Figure 10. 3.2.7. Extreme Gradient Boosting (XGBoost) The extreme gradient boosting method is an approach that uses a decision tree and gradient boosting structure [54]. In this approach, the value obtained by summing the results from the decision tree algorithm is f (x), and the model equation is as follows (Equation (7)) [55]. N ŷi = ∑ j=1 f j ( xi ) (7) The XGBoost algorithm advances by making improvements to the objective function, whose main task is to optimize the complexity penalty and loss function. Here, the loss function (LF) is expressed in Equation (8) and the complexity penalty (CP) in Equation (9) [56]. n LF = ∑i=1 l (yi , ŷi ) (8) k CP = ∑k Ω( f k ) (9) 3.2.6. Random Forest (RF) In the random forest (RF) algorithm, an advanced form of the decision tree algorithm, the results are obtained through multiple decision trees [52]. When used for classification purposes, the output produced by the most trees is considered the final result. In contrast, Energies 2023, 16, 5092 13 of 20 the average results obtained from all of the trees in regression problems give the final result [53]. A typical RF structure is presented in Figure 10. Energies 2023, 16, x FOR PEER REVIEW 15 of 22 Figure10. Figure 10.Random Randomforest forestalgorithm algorithmstructure structure[16]. [16]. 3.2.8. 3.2.7.Decision 3.2.8. DecisionTree Extreme TreeRegressor Gradient Boosting Regressor (DTree) (XGBoost) (DTree) The Thedecision decisiontree extreme algorithm gradient tree is aissupervised boosting algorithm method a supervisedismachine an approach learning machine thatmethod that is that uses method learning a decision frequently tree is and fre- used in quently regression gradientusedboosting and classification structureand in regression problems [54].classification In this approach, in the problemsliterature the invalue because they obtained by the literature are generally summing because they the are understandable results from generally theand practical decision understandable and[57]. tree In regression algorithm practical is problems, f(x),Inand [57]. the model regression if the targetifis equation problems, variable as target the takes follows con- (Equa- variable tinuous tion (7)) takes values, [55]. thevalues, continuous decision thetree regressor decision treecan be usedcan regressor at this stage.atDuring be used the estimation this stage. During the process, estimationthe process, dataset isthe iteratively dataset ispartitioned iterativelyand the algorithm partitioned and the runs until theruns algorithm partitioning until the process stops [58]. The decision tree ŷ = ∑can be method 𝑓 (𝑥 ) divided into decision nodes, (7) branches, partitioning process stops [58]. The decision tree method can be divided into decision and leaves. nodes,The A simple XGBoost branches, and decision algorithm Atree leaves. advances model simple by ismaking shown decision inmodel Figureis11 improvements tree [20]. to the shown in objective Figure 11function, [20]. whose main task is to optimize the complexity penalty and loss function. Here, the loss function (LF) is expressed in Equation (8) and the complexity penalty (CP) in Equation (9) [56]. 𝐿𝐹 = ∑ 𝑙(𝑦 , ŷ ) (8) 𝐶𝑃 = ∑ 𝛺(𝑓 ) (9) Figure11. Figure 11.Decision Decisiontree treealgorithm algorithmstructure. structure. 3.3. Validation and Evaluation 3.3.1. Mean Absolute Error (MAE) Mean absolute error (MAE) is an error metric found by calculating the average abso- lute values of the distances between the actual and predicted values [59,60]. Calculation of the mean absolute error is shown in Equations (10) and (11). Energies 2023, 16, 5092 14 of 20 3.3. Validation and Evaluation 3.3.1. Mean Absolute Error (MAE) Mean absolute error (MAE) is an error metric found by calculating the average absolute values of the distances between the actual and predicted values [59,60]. Calculation of the mean absolute error is shown in Equations (10) and (11). 1 N N ∑ i =1 i MAE = e (10) ei = | a i − p i | (11) Here, ei is an error value, ai is the actual value, and pi is the predicted value. 3.3.2. Root Mean Square Error (RMSE) Root mean square error (RMSE) is a measure of the distance between values predicted by a model and actual values. It is calculated by taking the square root of the mean of the square of the difference between the actual values and the predicted values [25,59]. The calculation of RMSE is given in Equation (12), s ∑iN=1 ei2 RMSE = (12) N where ei = ai − pi represents the error value, ai is the actual value, pi is the predicted value, and N is amount of data. 3.3.3. Mean Absolute Percentage Error (MAPE) The mean absolute percentage error (MAPE) is the percentage of difference between actual and predicted values. The calculation of MAPE is expressed in Equation (13) [61]. 100% n yi − pi n ∑ i =1 y i MAPE = (13) In this equation, y is the actual value and p is the predicted value. 3.3.4. K-Fold Cross-Validation K-fold cross-validation is a verification method in which algorithms are checked for overfitting problems [62]. In this method, the dataset is divided into equal parts; one test is divided for validation and the other parts are separated as training data. The validation process continues until all parts are processed, and the average of the results obtained is calculated as the validation score [60]. The k-fold cross-validation process is described in Figure 12. 3.3.4. K-Fold Cross-Validation K-fold cross-validation is a verification method in which algorithms are checked for overfitting problems [62]. In this method, the dataset is divided into equal parts; one test is divided for validation and the other parts are separated as training data. The validation process continues until all parts are processed, and the average of the results obtained is Energies 2023, 16, 5092 15 of 20 calculated as the validation score [60]. The k-fold cross-validation process is described in Figure 12. Figure 12. K-fold cross-validation. Figure 12. K-fold cross-validation. 4.4.Simulation SimulationResults Resultsand andDiscussion Discussion Achievinggenerator Achieving generatorload loadprediction predictionoutput outputvalues valuesthat thatare areclose closeto tothe thereal realvalues valuesisis the main objective of the proposed training model. A forecasting model the main objective of the proposed training model. A forecasting model that produces the that produces the lowestrate lowest rateof oferror errordemonstrates demonstratesaahigh highprediction predictionaccuracy accuracyand andisiscapable capableof ofgenerating generating reliableresults reliable resultsfor fornewnewsets setsof ofdata. data.In Inorder ordertoto figure figureout out the the prediction prediction model model for for aa data- data- driven drivenapproach approachthat thatcan cansatisfy satisfythe theprediction predictionobjective, objective,eight eightdifferent differentalgorithms algorithmswerewere utilized, utilized,which whichare areSVM, SVM,MLR,MLR,ANN,ANN,DNN, DNN,KNN,KNN,RF, RF,XGB, XGB,and andDtree. Dtree.The Thecomparative comparative results resultsfrom fromall allofofthe thealgorithms algorithms provide provide the best the bestprediction predictionmodel model forfor thethe generator load generator of load the chemical tanker. The findings from the analysis of each prediction of the chemical tanker. The findings from the analysis of each prediction model’s perfor-model’s performance are addressed mance in this section. are addressed in this section. This Thisstudy studyused usedthe theTensorFlow TensorFlowenvironment environmentof ofPython Python programming programming languagelanguage ver-ver- sion sion 3.7.7 3.7.7ininsimulations simulationsfor formarine marinediesel dieselgenerator generatorpower powerprediction. prediction. The Thecomputer computer hardware hardware utilized utilized an an Intel Intel Core Core i7-9750H i7-9750H 2.60 2.60 GHz GHz processor, processor, 32 32 GBGB of ofRAM, RAM,and andan an NVIDIA GeForce RTX 2070 graphics card. In the first stage of the simulation, NVIDIA GeForce RTX 2070 graphics card. In the first stage of the simulation, the hyperpa- the hyperpa- rameters rametersof ofthe theestablished establishedmodels modelswere weretuned tunedsince sincesome someof ofthe thealgorithms algorithmswere wereunable unable to to produce the desired prediction scores. Hyperparameter tuning has the advantageof produce the desired prediction scores. Hyperparameter tuning has the advantage of providing providingoptimized optimizedvalues valuesforforthe theprediction predictionmodel, model,maximizing maximizingthe thepredictive predictiveaccuracy accuracy as much as possible. This is an essential step in controlling the behaviour of the predictive model. Without the hyperparameter tuning step, the prediction model will make more errors during the data simulation. Table 3 shows the model hyperparameters obtained as a result of this process. The performance of each prediction algorithm was evaluated by several error metrics such as MAE, RMSE, and MAPE. In the first scenario, the prediction model was conducted for each generator of DG1, DG2, and DG3 separately. Table 4 compares the results of the eight prediction algorithms for the power prediction in each generator. It shows that the decision tree (Dtree) algorithm is more efficient than the other algorithms used for this specific case study. It reached values of 0.2364 for MAE, 0.2455 for RMSE, and 17.493 for MAPE for the DG1 power prediction case. When DG2 power prediction scores were evaluated, the DTree algorithm reached values of 0.1306 for MAE, 0.2069 for RMSE, and 5.1139 for MAPE. For DG3 power prediction, the DTree algorithm reached values of 0.1532 for MAE, 0.2182 for RMSE, and 7.7481 for MAPE. The DTree algorithm is a typical model-based learning method. The difference of this method from other algorithms is the modelling of the relationship between the input Energies 2023, 16, 5092 16 of 20 variables and the output variable. For this reason, it can achieve successful results in both linear and nonlinear systems [63]. On the other hand, XGB, RF, and DNN also show a good performance, with their error measurement performing slightly better than Dtree’s. Meanwhile, the SVR technique has the worst performance with the highest error rate. This could be due to it performing operations with a margin between hyperplanes that allows a certain degree of error. Meanwhile, the other prediction algorithm is intended to minimize the error rate as much as possible. That justifies the highest error performance for the case study. Table 3. Model hyperparameters. Model Hyperparameter SVM None MLR None solver = ‘lbfgs’, alpha = 0.00001, max_iter = 10,000, activation = ‘tanh’, hidden_layer_sizes = (5000), power_t = 0.7, ANN validation_fraction = 0.3, batch_size = 250 input_dim = 16, hidden_layer_count = 17, input_layer_activation_function = ‘relu’, DNN hidden_layer_activation_function = ‘linear’, output_layer_activation_function = ‘linear’, optimizer = ‘Adam’, epochs = 1500 KNN n_neighbors = 3, weights = “distance”,algorithm = “kd_tree”, p = 20 RF n_estimators = 350, max_depth = 150 XGB None Dtree None Table 4. The prediction results for the single power diesel generator. MAE RMSE MAPE DG1 DG2 DG3 DG1 DG2 DG3 DG1 DG2 DG3 SVR 1.6736 1.9411 1.6893 6.3752 4.8703 7.5162 30.865 33.814 36.241 MLR 1.1652 1.3124 1.1804 2.6701 3.1117 2.4007 29.597 28.877 31.195 ANN 0.9589 1.0027 0.9324 1.9204 2.1717 1.8913 34.265 24.977 29.988 DNN 0.5783 0.4025 0.5308 1.8067 1.4948 1.8733 31.714 36.591 33.773 KNN 1.4294 1.5561 1.2901 4.4118 4.4908 4.0399 34.652 32.794 35.763 RF 0.3413 0.2252 0.2892 0.4733 0.2662 0.4714 31.879 8.1517 9.9462 XGB 0.2817 0.1424 0.2926 0.4212 0.2431 0.5093 29.547 6.1015 11.184 DTree 0.2364 0.1306 0.1532 0.2455 0.2069 0.2182 17.493 5.1139 7.7481 Knowing the effects of loads on generators can assist maritime companies in pre- dicting malfunctions, improving energy efficiency, and conducting maintenance-attitude studies [64]. When the generator load is known, the fuel consumption can be calculated. In this way, fuel consumption can be optimized, and steps can be taken toward greater energy efficiency [65]. Knowing the fuel consumption can also enable the detection of a malfunction or an error in the generator system when the consumption differs from the normal state, i.e., when the energy efficiency changes. This can help shipping companies in terms of predictive maintenance activities. In the second scenario, the total load power generator prediction model was developed by taking into account all of the chemical tanker ship’s generators. Table 5 shows the prediction results for the total load estimation from eight prediction models. To determine whether the models were overfitting, a 5-fold cross-validation procedure was used. The training dataset was divided into five parts with 200 samples each in this process; one of these parts was designated as the validation dataset, while the other four were used as training data in the model training. The cross-validation process was continued for five iterations, and the validation scores are given in Table 6 in terms of the mean absolute error metric. Energies 2023, 16, 5092 17 of 20 Table 5. The prediction results for the total power diesel generator. Total Load Prediction MAE RMSE MAPE SVR 1.3520 8.3232 15.966 MLR 1.5283 5.6608 23.302 ANN 1.6575 6.1073 25.048 DNN 1.0866 2.6049 14.728 KNN 1.8220 9.6310 27.184 RF 1.3146 4.9486 14.513 XGB 1.4152 5.7916 16.805 DTree 1.4452 9.0413 19.748 Table 6. 5-fold cross-validation scores (MAE) for the total load prediction. Model SVR MLR ANN DNN KNN RF XGB DTree Fold 1 1.4235 1.5813 1.7434 1.1517 1.9589 1.4257 1.4789 1.4925 Fold 2 1.4387 1.5975 1.6928 1.2146 1.9827 1.3952 1.4912 1.5014 Fold 3 1.4291 1.6124 1.7126 1.1738 1.9186 1.3573 1.4671 1.4874 Fold 4 1.4413 1.5731 1.6891 1.1513 1.8917 1.3841 1.4397 1.4731 Fold 5 1.4275 1.5629 1.6973 1.1725 1.9226 1.3619 1.4515 1.4667 Mean Score 1.4320 1.5854 1.7070 1.1727 1.9349 1.3848 1.4656 1.4842 In this case, the deep neural network (DNN) algorithm outperforms all other algo- rithms, with minimum error performance values of 1.0866 (MAE), 2.6049 (RMSE), and 14.728 (MAPE). In light of the findings obtained in the first simulation study, it was re- vealed that model-based algorithms for the generator power estimation case phase are more advantageous due to the format of the dataset. In addition, it can be said that the XGBoost algorithm, which uses a progressive learning methodology, stands out among other algorithms used in the simulation process. In the second part of the simulation study, concerning the prediction of the total load of the generators, it was revealed that the DNN algorithm, which works more efficiently in difficult and complex systems, achieves more successful prediction scores than other algorithms. Unlike ANN, there are multiple hidden layers in DNN. This allows the algorithm to behave more effectively in more complex prob- lems. In addition, the DNN algorithm is also a successful method for nonlinear systems. At this stage, reducing the number of inputs of the system significantly reduced the success of the models established with MLR, ANN, K-NN, RF, XGBoost, and DTree algorithms. However, the DNN and SVR algorithms improved their scores in this process. When the DNN and SVR algorithms were compared, it was determined that the success of the DNN algorithm increased significantly in the second simulation. It can be observed that the success of the SVR algorithm also increased, but this increase was not at a satisfactory level. 5. Conclusions Generators are the second-largest fuel consumer on a ship after the primary engine and, thus, are an important component when evaluating fuel utilization and enhancing energy efficiency. If the electrical load on the generators cannot be correctly predicted and planned for, electrical failures may also occur under sudden loads. This issue may force the ship’s operations to cease and might result in the ship’s severe damage and workplace accidents. This study employed data-driven algorithms to estimate the load on a commercial ship’s generators. The results from the simulation indicated that the DTree model was the most successful method for the generator load prediction, with MAE scores of 0.2364, 0.1306, and 0.1532; RMSE scores of 0.2455, 0.2069, and 0.2182; and MAPE scores of 17.493, 5.1139, and 7.748 for diesel generators DG1, DG2, and DG3, respectively. On the other hand, the DNN model was the most successful method for the total load Energies 2023, 16, 5092 18 of 20 prediction case, with 1.0866 MAE, 2.6049 RMSE, and 14.728 MAPE scores. The 5-iteration k-fold cross-validation procedure used to determine whether the models had an overfitting condition revealed no overfitting condition. It can be concluded from the results that data-driven algorithms can be useful for estimating the electrical load on ship generators. The problems arising from the sudden loads experienced in the ship’s electrical grid may be prevented by injecting these developed models in this study into the ship microgrid control in the future. This will make it possible to take preventative action against electrical failures, prevent potential malfunctions, reduce potential additional maintenance costs, and increase energy efficiency. Author Contributions: T.U.: Conceptualization, Methodology, Software, Validation, Formal analysis, Writing—original draft, N.N.A.B.: Conceptualization, Methodology, Validation, Writing—review & editing, Writing—original draft, Ö.K. and Y.A.: Supervision, Investigation, Writing—review & editing, Writing—original draft, J.M.G. and A.L.: Supervision, Writing—review & Editing. All authors have read and agreed to the published version of the manuscript. Funding: This work was supported by The Scientific and Technological Research Council of Turkey BIDEB 2214-A International Doctoral Research Fellowship Programme reference number:1059B142100334. Josep M. Guerrero and Abderezak Lashab are funded by VILLUM FONDEN under the VILLUM Inves- tigator Grant (25920): Center for Research on Microgrids (CROM). Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Not applicable. Acknowledgments: This work was supported by the Scientific and Technological Research Council of Turkey BIDEB- 2214 International Doctoral Research Fellowship Programme. This article is produced from the PhD dissertation entitled “Deep Learning Applications in Ship Electric Grids” which has been executed in the Maritime Transportation Engineering Program of ITU Graduate School. Conflicts of Interest: The authors declare no conflict of interest. References 1. Sáez Álvarez, P. From Maritime Salvage to IMO 2020 Strategy: Two Actions to Protect the Environment. Mar. Pollut. Bull. 2021, 170, 112590. [CrossRef] [PubMed] 2. Ammar, N.R.; Seddiek, I.S. Enhancing Energy Efficiency for New Generations of Containerized Shipping. Ocean Eng. 2020, 215, 107887. [CrossRef] 3. Bakar, N.N.A.; Bazmohammadi, N.; Çimen, H.; Uyanik, T.; Vasquez, J.C.; Guerrero, J.M. Data-Driven Ship Berthing Forecasting for Cold Ironing in Maritime Transportation. Appl. Energy 2022, 326, 119947. [CrossRef] 4. Gan, M.; Hou, H.; Wu, X.; Liu, B.; Yang, Y.; Xie, C. Machine Learning Algorithm Selection for Real-Time Energy Management of Hybrid Energy Ship. Energy Rep. 2022, 8, 1096–1102. [CrossRef] 5. Pagoropoulos, A.; Møller, A.H.; McAloone, T.C. Applying Multi-Class Support Vector Machines for Performance Assessment of Shipping Operations: The Case of Tanker Vessels. Ocean Eng. 2017, 140, 1–6. [CrossRef] 6. Planakis, N.; Papalambrou, G.; Kyrtatos, N. Ship Energy Management System Development and Experimental Evaluation Utilizing Marine Loading Cycles Based on Machine Learning Techniques. Appl. Energy 2022, 307, 118085. [CrossRef] 7. Hou, J.; Sun, J.; Hofmann, H. Adaptive Model Predictive Control with Propulsion Load Estimation and Prediction for All-Electric Ship Energy Management. Energy 2018, 150, 877–889. [CrossRef] 8. Hardan, F.; Norman, R. Balancing Loads of Rotating Generators Utilizing VSC Direct Power Controllers in a Ship AC/DC Smartgrid. Electr. Power Syst. Res. 2020, 182, 106200. [CrossRef] 9. Omer, H.A.; Mahjoub, K.O.; Karrar, A.A. On the Stability of Generators Load Sharing; IFAC: New York, NY, USA, 2014; Volume 19, ISBN 9783902823625. 10. Roy, N.B.; Das, D. Optimal Allocation of Active and Reactive Power of Dispatchable Distributed Generators in a Droop Controlled Islanded Microgrid Considering Renewable Generation and Load Demand Uncertainties. Sustain. Energy Grids Netw. 2021, 27, 100482. [CrossRef] 11. Kusakana, K. Optimal Peer-to-Peer Energy Sharing between Prosumers Using Hydrokinetic, Diesel Generator and Pumped Hydro Storage. J. Energy Storage 2019, 26, 101048. [CrossRef] 12. Li, X.; Sun, B.; Jin, J.; Ding, J. Speed Optimization of Container Ship Considering Route Segmentation and Weather Data Loading: Turning Point-Time Segmentation Method. J. Mar. Sci. Eng. 2022, 10, 1835. [CrossRef] Energies 2023, 16, 5092 19 of 20 13. Uyanık, T.; Yalman, Y.; Kalenderli, Ö.; Arslanoğlu, Y.; Terriche, Y.; Su, C.-L.; Guerrero, J.M. Data-Driven Approach for Estimating Power and Fuel Consumption of Ship: A Case of Container Vessel. Mathematics 2022, 10, 4167. [CrossRef] 14. Bassam, A.M.; Phillips, A.B.; Turnock, S.R.; Wilson, P.A. Ship Speed Prediction Based on Machine Learning for Efficient Shipping Operation. Ocean Eng. 2022, 245, 110449. [CrossRef] 15. Murray, B.; Perera, L.P. Proactive Collision Avoidance for Autonomous Ships: Leveraging Machine Learning to Emulate Situation Awareness. IFAC Pap. 2021, 54, 16–23. [CrossRef] 16. Rawson, A.; Brito, M.; Sabeur, Z.; Tran-Thanh, L. A Machine Learning Approach for Monitoring Ship Safety in Extreme Weather Events. Saf. Sci. 2021, 141, 105336. [CrossRef] 17. Peng, Y.; Liu, H.; Li, X.; Huang, J.; Wang, W. Machine Learning Method for Energy Consumption Prediction of Ships in Port Considering Green Ports. J. Clean. Prod. 2020, 264, 121564. [CrossRef] 18. Anh Tran, T. Comparative Analysis on the Fuel Consumption Prediction Model for Bulk Carriers from Ship Launching to Current States Based on Sea Trial Data and Machine Learning Technique. J. Ocean Eng. Sci. 2021, 6, 317–339. [CrossRef] 19. Ahlgren, F.; Mondejar, M.E.; Thern, M. Predicting Dynamic Fuel Oil Consumption on Ships with Automated Machine Learning. Energy Procedia 2019, 158, 6126–6131. [CrossRef] 20. Uyanık, T.; Karatuğ, Ç.; Arslanoğlu, Y. Machine Learning Approach to Ship Fuel Consumption: A Case of Container Vessel. Transp. Res. Part D Transp. Environ. 2020, 84, 102389. [CrossRef] 21. Kong, M.C.; Roh, M.-I.; Kim, K.S.; Lee, J.; Kim, J.; Lee, G. Object Detection Method for Ship Safety Plans Using Deep Learning. Ocean Eng. 2022, 246, 110587. [CrossRef] 22. Laurie, A.; Anderlini, E.; Dietz, J.; Thomas, G. Machine Learning for Shaft Power Prediction and Analysis of Fouling Related Performance Deterioration. Ocean Eng. 2021, 234, 108886. [CrossRef] 23. Saettone, S.; Tavakoli, S.; Taskar, B.; Jensen, M.V.; Pedersen, E.; Schramm, J.; Steen, S.; Andersen, P. The Importance of the Engine-Propeller Model Accuracy on the Performance Prediction of a Marine Propulsion System in the Presence of Waves. Appl. Ocean Res. 2020, 103, 102320. [CrossRef] 24. Wen, H.; Khan, F.; Amin, M.T.; Halim, S.Z. Myths and Misconceptions of Data-Driven Methods: Applications to Process Safety Analysis. Comput. Chem. Eng. 2021, 158, 107639. [CrossRef] 25. Lang, X.; Wu, D.; Mao, W. Comparison of Supervised Machine Learning Methods to Predict Ship Propulsion Power at Sea. Ocean Eng. 2022, 245, 110387. [CrossRef] 26. Zhou, T.; Hu, Q.; Hu, Z.; Zhen, R. An Adaptive Hyper Parameter Tuning Model for Ship Fuel Consumption Prediction under Complex Maritime Environments. J. Ocean Eng. Sci. 2021, 7, 255–263. [CrossRef] 27. Yan, R.; Wang, S.; Du, Y. Development of a Two-Stage Ship Fuel Consumption Prediction and Reduction Model for a Dry Bulk Ship. Transp. Res. Part E Logist. Transp. Rev. 2020, 138, 101930. [CrossRef] 28. Yuan, Z.; Liu, J.; Zhang, Q.; Liu, Y.; Yuan, Y.; Li, Z. Prediction and Optimisation of Fuel Consumption for Inland Ships Considering Real-Time Status and Environmental Factors. Ocean Eng. 2021, 221, 108530. [CrossRef] 29. Schäfers, P.; Mütze, A.; Nyhuis, P. Integrated Concept for Acquisition and Utilization of Production Feedback Data to Support Production Planning and Control in the Age of Digitalization. Procedia Manuf. 2019, 31, 225–231. [CrossRef] 30. Ellingsen, O.; Aasland, K.E. Digitalizing the Maritime Industry: A Case Study of Technology Acquisition and Enabling Advanced Manufacturing Technology. J. Eng. Technol. Manag. JETM 2019, 54, 12–27. [CrossRef] 31. Futagami, K.; Fukazawa, Y.; Kapoor, N.; Kito, T. ScienceDirect Pairwise Acquisition Prediction with SHAP Value Interpretation. J. Financ. Data Sci. 2021, 7, 22–44. [CrossRef] 32. Wang, D.; Thun, S.; Lindberg, U.; Jiang, L.; Trygg, J.; Tysklind, M. Towards Better Process Management in Wastewater Treatment Plants: Process Analytics Based on SHAP Values for Tree-Based Machine Learning Methods. J. Environ. Manag. 2022, 301, 113941. [CrossRef] 33. Corinna, C.; Vladimir, V. Support-Vector Networks. IEEE Expert Syst. Appl. 1992, 7, 63–72. [CrossRef] 34. Luíza da Costa, N.; Dias de Lima, M.; Barbosa, R. Evaluation of Feature Selection Methods Based on Artificial Neural Network Weights. Expert Syst. Appl. 2021, 168, 114312. [CrossRef] 35. Liu, M.; Zhou, Q.; Wang, X.; Yu, C.; Kang, M. Voyage Performance Evaluation Based on a Digital Twin Model. IOP Conf. Ser. Mater. Sci. Eng. 2020, 929, 012027. [CrossRef] 36. Xie, X.; Wu, T.; Zhu, M.; Jiang, G.; Xu, Y.; Wang, X.; Pu, L. Comparison of Random Forest and Multiple Linear Regression Models for Estimation of Soil Extracellular Enzyme Activities in Agricultural Reclaimed Coastal Saline Land. Ecol. Indic. 2021, 120, 106925. [CrossRef] 37. Caravaggi, P.; Leardini, A.; Giacomozzi, C. Multiple Linear Regression Approach for the Analysis of the Relationships between Joints Mobility and Regional Pressure-Based Parameters in the Normal-Arched Foot. J. Biomech. 2016, 49, 3485–3491. [CrossRef] [PubMed] 38. Hosseini, S.A.; Shirani, A.S.; Lotfi, M.; Menhaj, M.B. Design and Application of Supervisory Control Based on Neural Network PID Controllers for Pressurizer System. Prog. Nucl. Energy 2020, 130, 103570. [CrossRef] 39. Jeon, M.; Noh, Y.; Shin, Y.; Lim, O.K.; Lee, I.; Cho, D. Prediction of Ship Fuel Consumption by Using an Artificial Neural Network. J. Mech. Sci. Technol. 2018, 32, 5785–5796. [CrossRef] 40. Maepa, F.; Smith, R.S.; Tessema, A. Support Vector Machine and Artificial Neural Network Modelling of Orogenic Gold Prospectivity Mapping in the Swayze Greenstone Belt, Ontario, Canada. Ore Geol. Rev. 2021, 130, 103968. [CrossRef] Energies 2023, 16, 5092 20 of 20 41. Lazakis, I.; Raptodimos, Y.; Varelas, T. Predicting Ship Machinery System Condition through Analytical Reliability Tools and Artificial Neural Networks. Ocean Eng. 2018, 152, 404–415. [CrossRef] 42. Choi, S.; Kim, Y.J. Artificial Neural Network Models for Airport Capacity Prediction. J. Air Transp. Manag. 2021, 97, 102146. [CrossRef] 43. Son, S.; Oh, K.-Y. Integrated Framework for Estimating Remaining Useful Lifetime through a Deep Neural Network. Appl. Soft Comput. 2022, 122, 108879. [CrossRef] 44. Lyu, S.-H.; Wang, L.; Zhou, Z.-H. Improving Generalization of Deep Neural Networks by Leveraging Margin Distribution. Neural Netw. 2022, 151, 48–60. [CrossRef] [PubMed] 45. Tanaka, T.; Inui, T.; Kawai, S.; Kuwabara, S.; Nishizawa, H. Monitoring and Diagnostic Technologies Usingdeep Neural Networks for Predictive Optical Network Maintenance [Invited]. J. Opt. Commun. Netw. 2021, 13, 13–22. [CrossRef] 46. Kong, N.C.L.; Kaneshiro, B.; Yamins, D.L.K.; Norcia, A.M. Time-Resolved Correspondences between Deep Neural Network Layers and EEG Measurements in Object Processing. Vis. Res. 2020, 172, 27–45. [CrossRef] [PubMed] 47. ArunKumar, K.E.; Kalaga, D.V.; Kumar, C.M.S.; Kawaji, M.; Brenza, T.M. Forecasting of COVID-19 Using Deep Layer Recurrent Neural Networks (RNNs) with Gated Recurrent Units (GRUs) and Long Short-Term Memory (LSTM) Cells. Chaos Solitons Fractals 2021, 146, 110861. [CrossRef] 48. Su, T.J.; Pan, T.S.; Chang, Y.L.; Lin, S.S.; Hao, M.J. A Hybrid Fuzzy and K-Nearest Neighbor Approach for Debris Flow Disaster Prevention. IEEE Access 2022, 10, 21787–21797. [CrossRef] 49. Wang, Y.; Cao, X.; Li, Y. Unsupervised Outlier Detection for Mixed-Valued Dataset Based on the Adaptive k-Nearest Neighbor Global Network. IEEE Access 2022, 10, 32093–32103. [CrossRef] 50. Zhao, W.L.; Wang, H.; Ngo, C.W. Approximate K-NN Graph Construction: A Generic Online Approach. IEEE Trans. Multimed. 2021, 24, 1909–1921. [CrossRef] 51. Naimi, A.; Deng, J.; Shimjith, S.R.; Arul, A.J. Fault Detection and Isolation of a Pressurized Water Reactor Based on Neural Network and K-Nearest Neighbor. IEEE Access 2022, 10, 17113–17121. [CrossRef] 52. Gupta, R.; Pierdzioch, C.; Salisu, A.A. Oil-Price Uncertainty and the U.K. Unemployment Rate: A Forecasting Experiment with Random Forests Using 150 Years of Data. Resour. Policy 2022, 77, 102662. [CrossRef] 53. Zheng, J.; Liu, Y.; Ge, Z. Dynamic Ensemble Selection Based Improved Random Forests for Fault Classification in Industrial Processes. IFAC J. Syst. Control 2022, 20, 100189. [CrossRef] 54. Zhang, N.; Qian, H.; He, Y.; Li, L.; Sun, C. A Data-Driven Method for Power System Transient Instability Mode Identification Based on Knowledge Discovery and XGBoost Algorithm. IEEE Access 2021, 9, 154172–154182. [CrossRef] 55. Gu, X.; Han, Y.; Yu, J. A Novel Lane-Changing Decision Model for Autonomous Vehicles Based on Deep Autoencoder Network and XGBoost. IEEE Access 2020, 8, 9846–9863. [CrossRef] 56. Friedman, J.H. Greedy Function Approximation: A Gradient Boosting Machine. Ann. Stat. 2001, 29, 1189–1232. [CrossRef] 57. Saroj, R.K.; Anand, M. Environmental Factors Prediction in Preterm Birth Using Comparison between Logistic Regression and Decision Tree Methods: An Exploratory Analysis. Soc. Sci. Humanit. Open 2021, 4, 100216. [CrossRef] 58. Choi, J.; Gu, B.; Chin, S.; Lee, J. Automation in Construction Machine Learning Predictive Model Based on National Data for Fatal Accidents of Construction Workers. Autom. Constr. 2020, 110, 102974. [CrossRef] 59. Jackson, E.K.; Roberts, W.; Nelsen, B.; Williams, G.P.; Nelson, E.J.; Ames, D.P. Introductory Overview: Error Metrics for Hydrologic Modelling—A Review of Common Practices and an Open Source Library to Facilitate Use and Adoption. Environ. Model. Softw. 2019, 119, 32–48. [CrossRef] 60. Saud, S.; Jamil, B.; Upadhyay, Y.; Irshad, K. Performance Improvement of Empirical Models for Estimation of Global Solar Radiation in India: A k-Fold Cross-Validation Approach. Sustain. Energy Technol. Assess. 2020, 40, 100768. [CrossRef] 61. McKenzie, J. Mean Absolute Percentage Error and Bias in Economic Forecasting. Econ. Lett. 2011, 113, 259–262. [CrossRef] 62. Wong, T.T. Parametric Methods for Comparing the Performance of Two Classification Algorithms Evaluated by K-Fold Cross Validation on Multiple Datasets. Pattern Recognit. 2017, 65, 97–107. [CrossRef] 63. Liu, C.; Lin, B.; Lai, J.; Miao, D. An Improved Decision Tree Algorithm Based on Variable Precision Neighborhood Similarity. Inf. Sci. 2022, 615, 152–166. [CrossRef] 64. Brito, L.C.; Susto, G.A.; Brito, J.N.; Duarte, M.A.V. An Explainable Artificial Intelligence Approach for Unsupervised Fault Detection and Diagnosis in Rotating Machinery. Mech. Syst. Signal Process. 2022, 163, 108105. [CrossRef] 65. Yiğit, K. Evaluation of Energy Efficiency Potentials from Generator Operations on Vessels. Energy 2022, 257, 124687. [CrossRef] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.