Academia.eduAcademia.edu

Figure 5 – uploaded by Prof. Dr. Saad Mekhilef

See full PDFdownloadDownload figure
Figure 5. Details of the recurrent neural network architectures network. Recurrent neural network architecture is from a neural network family, though, is displays differences compared to ANN, in which the looping constraint on the hidden layer turns back to RNN [15]. The feedback constraint is back-propagated to ensure that the subsequent data is looped into the input data from the last step in each neurons first step, as shown in Figure 5. step, as shown in Figure 5.

Figure 5 Details of the recurrent neural network architectures network. Recurrent neural network architecture is from a neural network family, though, is displays differences compared to ANN, in which the looping constraint on the hidden layer turns back to RNN [15]. The feedback constraint is back-propagated to ensure that the subsequent data is looped into the input data from the last step in each neurons first step, as shown in Figure 5. step, as shown in Figure 5.

Related Figures (30)

Figure 1. Methodology of utilizing optimization to find optimal parameters of neural networks were counted as eligible for review of references. In this review, only the meaningful and suitable literature has been considered by evaluating the article’s relevant content and the critical topic of attention of the review. Accordingly, the related papers were designated based on the number of citations and research interest. This review methodology process comprises several stages, and the Prisma guidelines according to [36,37] were followed. Figure 1 shows how the methodology for utilizing optimization to find optimal parame- ters of neural networks. A schematic diagram of the review section process, evaluation, and quality control of the database using the Prisma guidelines is shown in Figure 2.
Figure 1. Methodology of utilizing optimization to find optimal parameters of neural networks were counted as eligible for review of references. In this review, only the meaningful and suitable literature has been considered by evaluating the article’s relevant content and the critical topic of attention of the review. Accordingly, the related papers were designated based on the number of citations and research interest. This review methodology process comprises several stages, and the Prisma guidelines according to [36,37] were followed. Figure 1 shows how the methodology for utilizing optimization to find optimal parame- ters of neural networks. A schematic diagram of the review section process, evaluation, and quality control of the database using the Prisma guidelines is shown in Figure 2.
Figure 2. Schematic diagram of the literature selection, evaluation and quality control process of the database using the Prisma guidelines.
Figure 2. Schematic diagram of the literature selection, evaluation and quality control process of the database using the Prisma guidelines.
Table 1. Advantages and disadvantages of the most popular nature-inspired optimization techniques.
Table 1. Advantages and disadvantages of the most popular nature-inspired optimization techniques.
Figure 3. Artificial neural network architectures with feed-forward and backpropagation algorithms. AVEANLIOERLIRY VI RRL CAN PV ER EER END EEE CHEE VAL EEEAR IT LOUNL Oe See MMiit | El am Ki Another example is feed-forward neural network probabilistic neural network (PNN) in [79] and knowledge-based neural network described in [80,81]. In [82] this ap- AN exa the bed arc proach was used for modeling a solar field in direct steam generation parabolic trough. N is used as an optimizer in many research projects to solve bundling problems; for mple, in [83] it was used to optimize a flight trajectory for rockets. An ANN optimized design and optimization of microwave circuits in [84]. Model-aided wireless AI em- ding expert knowledge in DNN to solve wireless system optimization to find he best hitecture of an ANN [22]. ANN is also used to optimize and control thin film growth processes [85]. A sampling method for the ANN model’s optimal design [86]. A feedfor- ward neural network optimization is applied to synthesize fault-tolerance [87]. ANNs, to- get Some optimized artificial neural her with the Xinanjiang model to employed to explore nonlinear transformations [88]. were done to decrease the cost of aq bloom forecasting accuracy [89]. A using ANN by optimizing heat-integration [90]. ANN solved the optimization problem and extraction of anthocyanins in black rice using orthogonal arrays [91]. ANN solves the opt used as an optimizer and applied to pin. for their convers neural network is shown in Figure 3. Each neuron output includes an activation func l inputs weights, while the neuron input is a sum of all weights included in the bias, as shown in Figure 4. The bias is a constant used to adjust the output and the weighted sum of the inputs to the neuron, while the activation functions are a powerhouse of asum of a process are d imization problems in traffic lig network models for predicting chlorophyll dynamics uatic environmental in-situ monitoring and increase problem of crude oil distillation systems was solved hts timing traffic light controller [92]. Also, ANN is waves energy converters (WEC) to predict overtop- g rates as part of a sustainable op timization of coastal or harbor defense structures and ion for constructing a predictive model [1]. The architecture of the artificial t. neural networks [93-99]. The neural network weights updates in the back-propagation one to get the gradients as a neural network using many hidden layers. The gradient may vanish and explode during the backward propagation [100,101].
Figure 3. Artificial neural network architectures with feed-forward and backpropagation algorithms. AVEANLIOERLIRY VI RRL CAN PV ER EER END EEE CHEE VAL EEEAR IT LOUNL Oe See MMiit | El am Ki Another example is feed-forward neural network probabilistic neural network (PNN) in [79] and knowledge-based neural network described in [80,81]. In [82] this ap- AN exa the bed arc proach was used for modeling a solar field in direct steam generation parabolic trough. N is used as an optimizer in many research projects to solve bundling problems; for mple, in [83] it was used to optimize a flight trajectory for rockets. An ANN optimized design and optimization of microwave circuits in [84]. Model-aided wireless AI em- ding expert knowledge in DNN to solve wireless system optimization to find he best hitecture of an ANN [22]. ANN is also used to optimize and control thin film growth processes [85]. A sampling method for the ANN model’s optimal design [86]. A feedfor- ward neural network optimization is applied to synthesize fault-tolerance [87]. ANNs, to- get Some optimized artificial neural her with the Xinanjiang model to employed to explore nonlinear transformations [88]. were done to decrease the cost of aq bloom forecasting accuracy [89]. A using ANN by optimizing heat-integration [90]. ANN solved the optimization problem and extraction of anthocyanins in black rice using orthogonal arrays [91]. ANN solves the opt used as an optimizer and applied to pin. for their convers neural network is shown in Figure 3. Each neuron output includes an activation func l inputs weights, while the neuron input is a sum of all weights included in the bias, as shown in Figure 4. The bias is a constant used to adjust the output and the weighted sum of the inputs to the neuron, while the activation functions are a powerhouse of asum of a process are d imization problems in traffic lig network models for predicting chlorophyll dynamics uatic environmental in-situ monitoring and increase problem of crude oil distillation systems was solved hts timing traffic light controller [92]. Also, ANN is waves energy converters (WEC) to predict overtop- g rates as part of a sustainable op timization of coastal or harbor defense structures and ion for constructing a predictive model [1]. The architecture of the artificial t. neural networks [93-99]. The neural network weights updates in the back-propagation one to get the gradients as a neural network using many hidden layers. The gradient may vanish and explode during the backward propagation [100,101].
Figure 4. Details of a single perceptron neuron for an artificial neural network.
Figure 4. Details of a single perceptron neuron for an artificial neural network.
Figure 6. Details of the convolution neural network architectures network with convolution layers. To the next convolutional layer, is the convolutional operation, as shown in Figure 6. Each convolutional layer specifies the number of filters used to detect the patterns of shapes in specific object shapes, for example (circle, squire, corner, eyes, feathers, etc.). These filters help extract the right and relevant features from the input data. CNN is the most widely used type of neural network for analyzing images. However, image analysis is but one use of CNN and it can be used for other data analysis problems such as classi- fication problems. Most generally, CNN is a critical neural network specializing in picking out patterns and making sense of them. This pattern detection is what makes CNN so useful for image analysis. These CNN models are used across different applications and domains, especially in image and video processing projects. CNN is related to solving images as multiple-image-based depth estimation or estimates depths in the edges; basi- cally, they are used to classify the edges in backgrounds or reflections [5]. CNN was used to detect wildfire smoke images [104] and forest fire smoke recognition [105]. 7. Overview of Neural Networks Enhanced by Optimization Algorithms
Figure 6. Details of the convolution neural network architectures network with convolution layers. To the next convolutional layer, is the convolutional operation, as shown in Figure 6. Each convolutional layer specifies the number of filters used to detect the patterns of shapes in specific object shapes, for example (circle, squire, corner, eyes, feathers, etc.). These filters help extract the right and relevant features from the input data. CNN is the most widely used type of neural network for analyzing images. However, image analysis is but one use of CNN and it can be used for other data analysis problems such as classi- fication problems. Most generally, CNN is a critical neural network specializing in picking out patterns and making sense of them. This pattern detection is what makes CNN so useful for image analysis. These CNN models are used across different applications and domains, especially in image and video processing projects. CNN is related to solving images as multiple-image-based depth estimation or estimates depths in the edges; basi- cally, they are used to classify the edges in backgrounds or reflections [5]. CNN was used to detect wildfire smoke images [104] and forest fire smoke recognition [105]. 7. Overview of Neural Networks Enhanced by Optimization Algorithms
Table 2. Studies involving PSO for neural network design based on weights and neuron number optimization.
Table 2. Studies involving PSO for neural network design based on weights and neuron number optimization.
Table 3. Studies involving PSO for neural networks design and application enhancement.
Table 3. Studies involving PSO for neural networks design and application enhancement.
Table 4. Studies involving GA for neural network design and application enhancement. 7.3. Artificial Neural Networks-Based Artificial Bee Colony
Table 4. Studies involving GA for neural network design and application enhancement. 7.3. Artificial Neural Networks-Based Artificial Bee Colony
Table 5. Studies involving ABC for neural networks design and application enhancement. 7.4. Artificial Neural Networks Based Evolutionary Algorithm
Table 5. Studies involving ABC for neural networks design and application enhancement. 7.4. Artificial Neural Networks Based Evolutionary Algorithm
Table 6. Studies involving EA for neural networks design and application enhancement. 7.5. Artificial Neural Networks-Based Backtracking Search Algorithm BSA optimization technique is an evolutionary computation technique for producing a trial population that includes two new crossovers and mutation operators proposed by [62]. BSA dominates searching for the best value of the populations and searches in the space boundary to get the exploitation capabilities and very robust exploration [62]. BSA dominates the search’s value for the best populations and the boundary of the space to provide very sturdy exploration and exploitation capabilities. Thus, considerable research has proven it as one of the most powerful optimization techniques [62]. Numerous researchers widely use BSA in modern applications, such as solving the state of charge of lithium-ion batteries by improving a backpropagation neural network (BPNN) by optimizing hidden layer neurons’ optimal value learning rate [167]. The BSA improved neural network with ran- dom weights by combining BSA and a neural network with random weights (NNRWs) to optimize the hidden layer parameters of the single-layer feed-forward network (SLFN), and NNRWs is used to derive the output layer weights [168]. In [169] a modified BSA (MBSA) has been improve by learning and niching together with ANN training and in [170] an ANN prediction method based on adaptive BSA was used for optimizing the connection weights matrix of the echo state network reservoir [170]. These studies involving BSA for neural network design and applications enhancement for the best neural network structure boost the performance level and reduce the time-consuming network setup. Table 7 presents dif- ferent research projects involving BSA for design and application enhancement. boundary to get the exploitation capabilities and very robust exploration [62]. BSA dominates
Table 6. Studies involving EA for neural networks design and application enhancement. 7.5. Artificial Neural Networks-Based Backtracking Search Algorithm BSA optimization technique is an evolutionary computation technique for producing a trial population that includes two new crossovers and mutation operators proposed by [62]. BSA dominates searching for the best value of the populations and searches in the space boundary to get the exploitation capabilities and very robust exploration [62]. BSA dominates the search’s value for the best populations and the boundary of the space to provide very sturdy exploration and exploitation capabilities. Thus, considerable research has proven it as one of the most powerful optimization techniques [62]. Numerous researchers widely use BSA in modern applications, such as solving the state of charge of lithium-ion batteries by improving a backpropagation neural network (BPNN) by optimizing hidden layer neurons’ optimal value learning rate [167]. The BSA improved neural network with ran- dom weights by combining BSA and a neural network with random weights (NNRWs) to optimize the hidden layer parameters of the single-layer feed-forward network (SLFN), and NNRWs is used to derive the output layer weights [168]. In [169] a modified BSA (MBSA) has been improve by learning and niching together with ANN training and in [170] an ANN prediction method based on adaptive BSA was used for optimizing the connection weights matrix of the echo state network reservoir [170]. These studies involving BSA for neural network design and applications enhancement for the best neural network structure boost the performance level and reduce the time-consuming network setup. Table 7 presents dif- ferent research projects involving BSA for design and application enhancement. boundary to get the exploitation capabilities and very robust exploration [62]. BSA dominates
Table 7. Studies involving BSA for neural network design and application enhancement.
Table 7. Studies involving BSA for neural network design and application enhancement.
Table 8. Overview of a variety of optimization techniques based on neural network design and application enhancement
Table 8. Overview of a variety of optimization techniques based on neural network design and application enhancement
Table 9. Overview of a variety of optimization-based neural network weights optimization enhancement. Lite IUBUW ils SLUUIEDS UVELT VIE VY SUILIC CAAGIIPICS UL VPUMEAAUT CLULALICCIILIEILE OL Vdin ious optimization-based neural network parameters (hidden layers, learning rate, neu- rons, and wights). For example, a hybrid lightning search algorithm (LSA)-based ANN can predict the optimal ON/OFF status for home appliances for home energy management by tuning the learning rate value and the number of nodes in the hidden layers in [30]. Also, in [195], FNNs are based on artificial fish swarm optimization (AFSA) to replace the BP process in ANN. A DNN based on multi-objective was used for solving the connecting structure DNNs, particularly the layerwise structure learning method, in [80]. An optimi- zation method for CNNs based on the difference between the present and the immediate past gradient diffGrad optimization technique to solve the problem with basic stochastic gradient descent (SGD) in [103]. A global optimal known as Bayesian optimization (BayesOpt) is a machine learning-based global optimization technique to solve a simple objective function problem in CNN [196]. A systematic quantitative and qualitative anal- ysis and guidelines using CNN-based Ben’s spiker algorithm [197]. In [198], the microca- nonical optimization algorithm (MOA) is used to select the best hyperparameter architec- ture for CNN for a variant of simulated annealing. DNNs with stochastic optimization acceleration update the network parameters to solve PID controller problems in [199]. A hybrid neuro-fuzzy network-based differential biogeography-based optimization (DBBO) for online population classification in an earthquakes optimizer searches for the best pa- rameter for the main network and the subnetwork [200]. An adaptive memetic algorithm with a rank-based mutation (AMARM) is used to design ANN architectures by a simul- taneously fine-tune number of hidden neurons as well as connection weights in [201].
Table 9. Overview of a variety of optimization-based neural network weights optimization enhancement. Lite IUBUW ils SLUUIEDS UVELT VIE VY SUILIC CAAGIIPICS UL VPUMEAAUT CLULALICCIILIEILE OL Vdin ious optimization-based neural network parameters (hidden layers, learning rate, neu- rons, and wights). For example, a hybrid lightning search algorithm (LSA)-based ANN can predict the optimal ON/OFF status for home appliances for home energy management by tuning the learning rate value and the number of nodes in the hidden layers in [30]. Also, in [195], FNNs are based on artificial fish swarm optimization (AFSA) to replace the BP process in ANN. A DNN based on multi-objective was used for solving the connecting structure DNNs, particularly the layerwise structure learning method, in [80]. An optimi- zation method for CNNs based on the difference between the present and the immediate past gradient diffGrad optimization technique to solve the problem with basic stochastic gradient descent (SGD) in [103]. A global optimal known as Bayesian optimization (BayesOpt) is a machine learning-based global optimization technique to solve a simple objective function problem in CNN [196]. A systematic quantitative and qualitative anal- ysis and guidelines using CNN-based Ben’s spiker algorithm [197]. In [198], the microca- nonical optimization algorithm (MOA) is used to select the best hyperparameter architec- ture for CNN for a variant of simulated annealing. DNNs with stochastic optimization acceleration update the network parameters to solve PID controller problems in [199]. A hybrid neuro-fuzzy network-based differential biogeography-based optimization (DBBO) for online population classification in an earthquakes optimizer searches for the best pa- rameter for the main network and the subnetwork [200]. An adaptive memetic algorithm with a rank-based mutation (AMARM) is used to design ANN architectures by a simul- taneously fine-tune number of hidden neurons as well as connection weights in [201].
Table 10. Overview of various optimization-based neural network parameters (hidden layers, learning rate, neurons, weights) optimization enhancement. 7.7. Optimization Search Algorithm-Based Artificial Neural Networks
Table 10. Overview of various optimization-based neural network parameters (hidden layers, learning rate, neurons, weights) optimization enhancement. 7.7. Optimization Search Algorithm-Based Artificial Neural Networks
igement, reliable power generation, = Jo = Ss . © ae Previous research on optimal scheduling controllers was developed for energy man- cost minimization, and carbon emission calculation 64]. A binary BSA algorithm and binary PSO are utilized to search for optimal binary schedules [120,205]. These algorithm techniques have powerful optimization skills, search xploration process, fast convergence for the solution, and other conventional optimiza- ion techniques and overcome local minima traps. Besides, developing an enhanced ANN- »ased BBSA and ANN-based BPSO schedule controller ensures the best performance icross different load conditions. [206,207]. An ANN is on track as a prediction technique o find he best weight values for neural nets designed for efficient system operation. In his paper, these nets operate in optimum ON/OFF status by training on input and output lata patterns obtained from schedul ing controllers [206,208]. This section presents ANN- ased optimization algorithms implementation for ANN-PSO, ANN-GA, ANN-ABC, and \ANN-BSA, respectively, to search for the optimal values of the number of nodes in hidden 1odes layer] and layer2 as well as the best value of the learning rate. The algorithms apply imitations of, e.g., (max and min) number of nodes in each hidden layer and the learning ate. The output data is relayed ona binary schedule (25x24) obtained from the scheduling controller. The input data includes nergy price, battery status, gird sta six inputs, including solar irradiances, wind speed, tus, and diesel fuel status refer to [206,208]. In all the \ANN-based algorithms, the iteration of ANN is set to 100 iterations, and the population size is 20 populations. Table 10 presents a brief list describing the data and the limitations »f the aforementioned algorithm tec hniques. The mean absolute error (MAE) is an objec- ive function that enhances the ANN performance by decreasing the error function ex- oressed in the general flow chart of n Figure 7. All the inputs and outpu optimization of ANN optimal parameters, as shown ts of the ANN-based optimization algorithm training or the virtual power plants system in [208] can be expressed by the following Equations 1) and (2): ANN deep neural networks using feed-forward structures have been adopted in this study. The use of trainlm as a network training function that updates weight and bias values according to Levenberg-Marquardt optimization is considered the fastest back- propagation algorithm in the Matlab toolbox. However, it does require more memory
igement, reliable power generation, = Jo = Ss . © ae Previous research on optimal scheduling controllers was developed for energy man- cost minimization, and carbon emission calculation 64]. A binary BSA algorithm and binary PSO are utilized to search for optimal binary schedules [120,205]. These algorithm techniques have powerful optimization skills, search xploration process, fast convergence for the solution, and other conventional optimiza- ion techniques and overcome local minima traps. Besides, developing an enhanced ANN- »ased BBSA and ANN-based BPSO schedule controller ensures the best performance icross different load conditions. [206,207]. An ANN is on track as a prediction technique o find he best weight values for neural nets designed for efficient system operation. In his paper, these nets operate in optimum ON/OFF status by training on input and output lata patterns obtained from schedul ing controllers [206,208]. This section presents ANN- ased optimization algorithms implementation for ANN-PSO, ANN-GA, ANN-ABC, and \ANN-BSA, respectively, to search for the optimal values of the number of nodes in hidden 1odes layer] and layer2 as well as the best value of the learning rate. The algorithms apply imitations of, e.g., (max and min) number of nodes in each hidden layer and the learning ate. The output data is relayed ona binary schedule (25x24) obtained from the scheduling controller. The input data includes nergy price, battery status, gird sta six inputs, including solar irradiances, wind speed, tus, and diesel fuel status refer to [206,208]. In all the \ANN-based algorithms, the iteration of ANN is set to 100 iterations, and the population size is 20 populations. Table 10 presents a brief list describing the data and the limitations »f the aforementioned algorithm tec hniques. The mean absolute error (MAE) is an objec- ive function that enhances the ANN performance by decreasing the error function ex- oressed in the general flow chart of n Figure 7. All the inputs and outpu optimization of ANN optimal parameters, as shown ts of the ANN-based optimization algorithm training or the virtual power plants system in [208] can be expressed by the following Equations 1) and (2): ANN deep neural networks using feed-forward structures have been adopted in this study. The use of trainlm as a network training function that updates weight and bias values according to Levenberg-Marquardt optimization is considered the fastest back- propagation algorithm in the Matlab toolbox. However, it does require more memory
Table 11. Studies involving neural networks for improving optimization techniques design and application enhancement. 8. Application on Artificial Neural Networks Based Optimization Algorithms
Table 11. Studies involving neural networks for improving optimization techniques design and application enhancement. 8. Application on Artificial Neural Networks Based Optimization Algorithms
Figure 7. General flow chart of optimization algorithms for ANN-PSO, ANN-GA, ANN- ABC, and ANN- BSA for 100 iteration.
Figure 7. General flow chart of optimization algorithms for ANN-PSO, ANN-GA, ANN- ABC, and ANN- BSA for 100 iteration.
Figure 8. ANN Architecture Based on PSO, GA, ABC, and BSA algorithms, using input and output data obtained from [208].
Figure 8. ANN Architecture Based on PSO, GA, ABC, and BSA algorithms, using input and output data obtained from [208].
Table 12. Optimization algorithms data and limitations. In this study, the BSA objective of enhancing the ANN structure toward optimal pa- rameters was the best among the other techniques, which minimize the MA value of 0.0062 [210]. The GA objective was 0.0080, which is not very far objective. Simultaneously, the MAE of the PSO and the ABC was greater 0.0172, respectively, compared to the other two techniques, as shown in BSA’s main principles have been done through its crossover consists of two parts. The first part generates the binary matrix, and the second part compares popula the trial population. Crossover is used to obtain an updated map(i,j). Also, this part works F to reaches a from the BSA at 0.0144 and Figure 9. The tion X(i,j) and on the control mechanism of boundaries for a trial population. As presented, enhancing the neural network could help the system enormously and the enhanced A NN is prover to be overwhelmingly impressive, or at least competitive, by training and testing is as important as the optimal design of the ANN structure. This study also introduces a novel way of solving optimization tasks by the neural network.
Table 12. Optimization algorithms data and limitations. In this study, the BSA objective of enhancing the ANN structure toward optimal pa- rameters was the best among the other techniques, which minimize the MA value of 0.0062 [210]. The GA objective was 0.0080, which is not very far objective. Simultaneously, the MAE of the PSO and the ABC was greater 0.0172, respectively, compared to the other two techniques, as shown in BSA’s main principles have been done through its crossover consists of two parts. The first part generates the binary matrix, and the second part compares popula the trial population. Crossover is used to obtain an updated map(i,j). Also, this part works F to reaches a from the BSA at 0.0144 and Figure 9. The tion X(i,j) and on the control mechanism of boundaries for a trial population. As presented, enhancing the neural network could help the system enormously and the enhanced A NN is prover to be overwhelmingly impressive, or at least competitive, by training and testing is as important as the optimal design of the ANN structure. This study also introduces a novel way of solving optimization tasks by the neural network.
Figure 9. Objectives of optimization algorithms for ANN-PSO, ANN-GA, ANN-ABC, and ANN- BSA for 100 iterations.
Figure 9. Objectives of optimization algorithms for ANN-PSO, ANN-GA, ANN-ABC, and ANN- BSA for 100 iterations.
Figure 10. MATLAB Simulink block of Neural Network Net of ANN-BSA. The o ure 10, inv ptimal enhanced net of ANN-BSA in a Matlab Simulink block is shown in Fig- olving six inputs and twenty-five binary outputs on an hourly basis to manage distributed generators throughout the virtual power plant system. The net block is gener- ated after the training completer by using Equation (3). Table 13 presents the ANN train- ing-based Net modu Net could svstems to PSO, GA, ABC, and BSA using the optimized parameters. The generated ANN e is an AI controller; it is considered a masterpiece and smart controller. This be implemented in cheap microchips and used as a smart device to control hug serve in a very effective smart wav cheaplv.
Figure 10. MATLAB Simulink block of Neural Network Net of ANN-BSA. The o ure 10, inv ptimal enhanced net of ANN-BSA in a Matlab Simulink block is shown in Fig- olving six inputs and twenty-five binary outputs on an hourly basis to manage distributed generators throughout the virtual power plant system. The net block is gener- ated after the training completer by using Equation (3). Table 13 presents the ANN train- ing-based Net modu Net could svstems to PSO, GA, ABC, and BSA using the optimized parameters. The generated ANN e is an AI controller; it is considered a masterpiece and smart controller. This be implemented in cheap microchips and used as a smart device to control hug serve in a very effective smart wav cheaplv.
Table 13. Artificial neural network training-based PSO, GA, ABC, and BSA using the optimized parameters obtained. deep The following figures represent the training performance and regressions for A neural network after using the optimization algorithms’ optimal parame ers. study shows a fair compression based on each optimization technique to find the best frame and error time during training and find the required best save valuable time during the training and testing. Any of ters to serve the system in the best way. These hybrid could give better results than manual parameters tuning. techniques could save huge parameters, using smaller ne Yet, some techniques could the best fitness faster and more efficiently than others, as ANN-BSA in Figure 11, w: shows the best training performance of 6.3695 e” at 2317 epochs and regression (R) reach to best of 1 regression training which the best results it may obtain by training. This the optimization algorithms used NN pa- ria ts to find hich
Table 13. Artificial neural network training-based PSO, GA, ABC, and BSA using the optimized parameters obtained. deep The following figures represent the training performance and regressions for A neural network after using the optimization algorithms’ optimal parame ers. study shows a fair compression based on each optimization technique to find the best frame and error time during training and find the required best save valuable time during the training and testing. Any of ters to serve the system in the best way. These hybrid could give better results than manual parameters tuning. techniques could save huge parameters, using smaller ne Yet, some techniques could the best fitness faster and more efficiently than others, as ANN-BSA in Figure 11, w: shows the best training performance of 6.3695 e” at 2317 epochs and regression (R) reach to best of 1 regression training which the best results it may obtain by training. This the optimization algorithms used NN pa- ria ts to find hich
Figure 11. (a) Performance and (b) regression of ANN training after applying optimal parameters of ANN-BSA.
Figure 11. (a) Performance and (b) regression of ANN training after applying optimal parameters of ANN-BSA.
Figure 12. (a) performance and (b) regression of ANN training after applying optimal parameters of ANN-GA. “ Fair comparison results of Bus1 of 14-bus IEEE test system for virtual power plants utilize the optimized ANN net based on half-hour binary patterns for managing each dis- tributed generation (DG) unit in the system. The binary (ANN-BPSO), binary (ANN- BABC), binary ANN-BGA, and binary (ANN-BBSA) is a controller with binary output 0 or 1 to switch each DG ON or Off based on the inputs. Figure 15 shows that the entire algorithm saved a huge amount of power. Yet, all the saved power was considered with sharing new distributed resources to inject power to the loads instead of supplying power from the utility grid [212]. However, most of the optimized Nets have done an excellent job. However, some Nets are better than the others based on their objectives as can be seen that the total power for the 24 h of the ANN-BBSA Net was 1182.5 MW in comparison to other optimized Nets 1211.3 MW, 1184.3 MW, and 1252.9 MW for, ANN-BGA, ANN- BPSO, ANN-BABC, respectively.
Figure 12. (a) performance and (b) regression of ANN training after applying optimal parameters of ANN-GA. “ Fair comparison results of Bus1 of 14-bus IEEE test system for virtual power plants utilize the optimized ANN net based on half-hour binary patterns for managing each dis- tributed generation (DG) unit in the system. The binary (ANN-BPSO), binary (ANN- BABC), binary ANN-BGA, and binary (ANN-BBSA) is a controller with binary output 0 or 1 to switch each DG ON or Off based on the inputs. Figure 15 shows that the entire algorithm saved a huge amount of power. Yet, all the saved power was considered with sharing new distributed resources to inject power to the loads instead of supplying power from the utility grid [212]. However, most of the optimized Nets have done an excellent job. However, some Nets are better than the others based on their objectives as can be seen that the total power for the 24 h of the ANN-BBSA Net was 1182.5 MW in comparison to other optimized Nets 1211.3 MW, 1184.3 MW, and 1252.9 MW for, ANN-BGA, ANN- BPSO, ANN-BABC, respectively.
Figure 13. (a) Performance and (b) regression of ANN training after applying optimal parameters of ANN-PSO.
Figure 13. (a) Performance and (b) regression of ANN training after applying optimal parameters of ANN-PSO.
Figure 14. (a) Performance and (b) regression of ANN training after applying optimal parameters of ANN-ABC.
Figure 14. (a) Performance and (b) regression of ANN training after applying optimal parameters of ANN-ABC.
Figure 15. Original Bus! of 14-bus IEEE test system compared to ANN-based binary optimization algorithms ANN-BPSC ANN-BABC, ANN-BGA, and ANN-BBSA.
Figure 15. Original Bus! of 14-bus IEEE test system compared to ANN-based binary optimization algorithms ANN-BPSC ANN-BABC, ANN-BGA, and ANN-BBSA.
Table 14. Comparison of the proposed technique with other techniques of enhancing neural networks by finding the optimal parameters of No. of nodes in hidden layers and learning rate. Table 15. Overview of significant studies on NNs based optimization using node numbers in hidden layers and learning rate.
Table 14. Comparison of the proposed technique with other techniques of enhancing neural networks by finding the optimal parameters of No. of nodes in hidden layers and learning rate. Table 15. Overview of significant studies on NNs based optimization using node numbers in hidden layers and learning rate.
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved