Differential Evolution Strategy Research Articles

Neuroevolution reinforcement learning for multi-echelon inventory optimization with delivery options and uncertain discount

The advanced information technology has enabled supply chain to make centralized optimal decision, allowing to make a global optimal solution. However, dealing with uncertainty is important in inventory management. Besides demand and supply uncertainties, supplier discounts also often arise unexpectedly. Further, suppliers or third-parties typically offer various delivery options in which trade-off occurs between cost and lead time. Thus, this study introduces new problem namely Multi-Echelon Inventory Optimization with Delivery Options and Uncertain Discount (MEIO-DO-UD). As a solution, Neuroevolution Reinforcement Learning (NERL) framework is developed for minimizing total system cost. The environment is modeled via System Dynamics (SD) and actor is presented by integration of Artificial Neural Network and Evolutionary Algorithm (EA), creating an effective decision-making model under dynamic uncertainty. The experimental study has been conducted where two different supply chain networks are given namely serial and divergence. Three EA algorithms are compared namely Differential Evolution (DE), Memetic Algorithm (MA), and Evolution Strategy (ES). Furthermore, NERL is also compared with the EA-optimized classical continuous review model namely (s,Q). The result shows that regardless what EA type is used, the proposed NERL always outperforms EA-optimized (s,Q) model. The more complex the problem, the further improvement can be made i.e. cost reduction up to 58%, followed by the fill rate improvement. The result also shows that NERL can avoid overfitting. Managerial implications are highlighted where NERL provides the more stable inventory level among all supply chain partners and bull-whip effect can be damped.

Reentry trajectory optimization of hypersonic glide vehicle based on improved particle swarm algorithm

The reentry process of hypersonic glide vehicle presents a host of intricate constraint issues. A method for optimizing the reentry trajectory of hypersonic glide vehicle is presented to achieve the shortest possible reentry time. The approach utilizes the improved particle swarm optimization (IPSO) algorithm. Firstly, population initialization is performed using the quasi-oppositional differential evolution (QODE) strategy to promote diversity within the population. Then an adaptive adjustment algorithm is designed for the inertia weight coefficient and learning factor to achieve a balance between global and local search capabilities. Finally, utilizing the IPSO algorithm for optimizing the flight parameters of the reentry trajectory, the reentry time is reduced from 338 seconds to 335 seconds. The simulations results indicate that the IPSO algorithm outperforms both the standard PSO algorithm and genetic algorithm (GA) in terms of optimization performance index.

Distribution Network Reconfiguration Based on an Improved Arithmetic Optimization Algorithm

Aiming to address the defects of the arithmetic optimization algorithm (AOA), such as easy fall into local optimums and slow convergence speed during the search process, an improved arithmetic optimization algorithm (IAOA) is proposed and applied to the study of distribution network reconfiguration. Firstly, a reconfiguration model is established to reduce network loss, and a cosine control factor is introduced to reconfigure the math optimization accelerated (MOA) function to coordinate the algorithm’s global exploration and local exploitation capabilities. Subsequently, a reverse differential evolution strategy is introduced to improve the overall diversity of the population and Weibull mutation is performed on the better-adapted individuals generated in each iteration to ensure the quality of the optimal individuals generated in each iteration and strengthen the algorithm’s ability to approach the optimal solution. The performance of the improved algorithm is also tested using eight basis functions. Finally, simulation analysis is carried out by taking the IEEE33 and IEEE69 node systems and a real power distribution system as examples; the results show that the proposed algorithm can help to reconfigure the system quickly, and the system node voltages and network losses were significantly improved after the reconfiguration.

Evolutionary multitasking for multiobjective optimization based on hybrid differential evolution and multiple search strategy

Multitask optimization uses the knowledge transfer between tasks to deal with multiple related tasks simultaneously, which obtains better optimization performance. Currently, most multitask optimization algorithms have shortings such as slow convergence speed, difficulty in generating high-quality solutions and insufficient population diversity. To solve these problems, an new multiobjective multitask evolutionary algorithm is proposed in this paper. The algorithm introduces hybrid differential evolution strategy and multiple search strategy to generate offspring and high-quality solutions. In order to accelerate the population convergence and improve the lack of population diversity, we mixed two differential mutation operators with different functions to generate offspring. While generating high-quality solutions to improve the population convergence speed, some random solutions can also be generated to maintain the population diversity, which enhances the ability of the algorithm to jump out of the local optimal. Secondly, the multiple search strategy uses the information collected from multiple dimensions to optimize decision variables, and generates high-quality solutions by migrating useful knowledge. Finally, the performance of the algorithm is verified on two classical multitask test set. Compared with other related algorithms, the proposed algorithm has a faster convergence speed and better distribution performance, and shows a good ability to jump out of local optimal.

MSAO: A multi-strategy boosted snow ablation optimizer for global optimization and real-world engineering applications

Snow Ablation Optimizer (SAO) is a cutting-edge nature-inspired meta-heuristic technique that mimics the sublimation and melting processes of snow in its quest for optimal solution to complex problems. While SAO has demonstrated competitive performance in comparison to classical algorithms in early research, it still exhibits certain limitations including low convergence accuracy, a lack of population diversity, and premature convergence, particularly when addressing high-dimensional intricate challenges. To mitigate the above-mentioned adverse factors, this paper introduces a novel variant of SAO with featuring four enhancement strategies collectively referred as MSAO. Firstly, the good point set initialization strategy is employed to generate a uniformly distributed high-quality population, which facilitates the algorithm to enter the appropriate search domain rapidly. Secondly, the greedy selection method is adopted to reserve better candidate solutions for the next iteration, thus striking a robust exploration–exploitation balance. Then, the Differential Evolution (DE) scheme is introduced to expand the search range and enhance the exploitation capability of the algorithm for higher convergence accuracy. Finally, to reduce the risk of falling into local optima, a Dynamic Lens Opposition-Based Learning (DLOBL) strategy is developed to operate on the current optimal solution dimension by dimension. With the blessing of these strategies, the optimization performance of MSAO is comprehensively improved. To comprehensively evaluate the optimization performance of MSAO, a series of numerical optimization experiments are conducted using the IEEE CEC2017 & CEC2022 test sets. In the IEEE CEC2017 experiments, the optimal crossover probability CR=0.8 is determined and the effectiveness of each improvement strategy is ablatively verified. MSAO is compared with the basic SAO, various state-of-the-art optimizers, and CEC2017 champion algorithms in terms of solution accuracy, convergence speed, robustness, and scalability. In the IEEE CEC2022 experiments, MSAO is compared with some recently developed improved algorithms to further validate its superiority. The results demonstrate that MSAO has excellent overall optimization performance, with the smallest Friedman mean rankings of 1.66 and 1.25 on both test suites, respectively. In the majority of test cases, MSAO can provide more accurate and reliable solutions than other competitors. Furthermore, six realistic constrained engineering design challenges and one photovoltaic model parameter estimation issue are employed to demonstrate the practicality of MSAO. Our findings suggest that MSAO has excellent optimization capacity and broad application potential.

Time–jerk optimal trajectory planning for industrial robots with coupled interpolation function selection

AbstractIn the contemporary field of optimal trajectory planning for industrial robots, it is customary to construct trajectories through the manual predefinition of interpolation functions. Unfortunately, this method frequently overlooks the influence of the interpolation function itself on the optimization objectives, resulting in suboptimal outcomes. To remedy this limitation, an optimal trajectory planning method with coupled interpolation function selection is proposed, in which the total task time and the integral squared jerk are defined as optimization objectives. This method minimizes the optimization objectives while also factoring in the optimal interpolation function, and avoiding subjective interference. To address the aforementioned biobjective optimization problem better, an Improved MultiObjective Golden Eagle Optimizer is introduced. Population diversity and the ability to escape local optima are enhanced through the incorporation of Chaotic Mapping, Opposition‐Based Learning, Differential Evolution, and adaptive inertia weight strategy into the algorithm. The superiority of the algorithm is validated through a series of simulations on 17 benchmark functions. In the context of the robotic stirring operation within the automated block cast charging process, the proposed method is utilized to derive the time–jerk optimal trajectory. The results demonstrate the effectiveness of the proposed method.

DP-EPSO: Differential privacy protection algorithm based on differential evolution and particle swarm optimization

In deep learning differential privacy protection, adding noise based on gradient has become a mainstream algorithm, but excessive gradient perturbation noise causes accuracy degradation. To solve this problem, a differential privacy protection algorithm based on differential evolution and particle swarm optimization is proposed to realize hyperparameter optimization in differential privacy, reduce the impact of noise on the model, and effectively improve the accuracy. On the one hand, the differential evolution scheme performs selection, crossover and mutation on learning rate η, make it approach the global optimal solution, and improve the computational efficiency of the algorithm. On the other hand, the particle swarm optimization scheme is adopted. Without changing the parameters and gradient structure, the parameters are optimized by using the network propagation attributes, which reduces the influence of noise on the accuracy. Experiments are performed on three datasets: Cifar10, Mnist and FashionMnist. Compared with the existing differential privacy algorithms, under the same privacy budget, the proposed algorithm has better accuracy and higher efficiency.

Using Optimized Long Short-Term Memory For Time-Series Forecasting of Electric Vehicles Battery Charging

The last decade has seen a significant rise in the adoption and development of Electric Vehicles (EVs), driven by environmental concerns, technological advancements, and governmental support. Batteries, central to EVs, have witnessed groundbreaking innovations in terms of energy density, charging speeds, and longevity. Expanding charging infrastructure and the automotive industry's investment in EV research have made them more mainstream. Effective Battery Management (BM), which includes monitoring essential parameters and thermal management, is critical for the longevity and reliability of EVs. Accurate charge prediction, in particular, aids in trip planning, reduces range anxiety and facilitates cost-effective charging coordinated with dynamic electricity pricing. Traditional models like linear regression and Autor-Rgressive Integrated Moving Average (ARIMA) have been standard for EV battery charge prediction. However, these often struggle with the dynamic nature of EV charging data. Even models like the vanilla Long Short-Term Memory (LSTM), which are adept at recognizing long-term patterns, require meticulous hyperparameter tuning. This work introduces the DWT-DE-LSTM model, which utilizes the Discrete Wavelet Transform (DWT) to dissect battery charging data at different resolutions and a Differential Evolution (DE) strategy for model optimization. Tests using the Panasonic 18650PF Li-ion Battery Dataset revealed the superior efficacy of the DWT-DE-LSTM model, emphasizing its suitability for real-world battery charge prediction.

Blockchain event ordering algorithm based on high-dimensional multi-objective differential evolution

Abstract Since the birth of Bitcoin, blockchain technology has been widely used in many fields, such as finance, supply chain management, smart contracts, etc.and has been highly valued for its decentralized and tamper-proof characteristics. With the expansion of application scenarios, blockchain faces inefficient data processing challenges, particularly in event sequencing and processing. Developing efficient event sequencing algorithms is a research priority to enhance the performance of blockchain systems. This study addresses the efficiency problem in blockchain event sequencing by introducing a high-dimensional multi-objective differential evolution strategy to improve blockchain technology’s performance in various fields. This paper uses the NSGA-III-WA algorithm to efficiently sort blockchain events by combining differential evolution strategies. Compared to the traditional sorting algorithm, this algorithm reduces latency by 30% and improves memory usage by about 20% when dealing with events of the same size, ensuring sorting accuracy. By optimizing the weight vector adjustment strategy, the convergence speed and distribution uniformity of the algorithm and the solution can be optimized. In this paper, the algorithm can effectively improve the efficiency and performance of blockchain event processing, which is critical in optimizing blockchain systems.

An improved composite particle swarm optimization algorithm for solving constrained optimization problems and its engineering applications

<abstract><p>In the last few decades, the particle swarm optimization (PSO) algorithm has been demonstrated to be an effective approach for solving real-world optimization problems. To improve the effectiveness of the PSO algorithm in finding the global best solution for constrained optimization problems, we proposed an improved composite particle swarm optimization algorithm (ICPSO). Based on the optimization principles of the PSO algorithm, in the ICPSO algorithm, we constructed an evolutionary update mechanism for the personal best position population. This mechanism incorporated composite concepts, specifically the integration of the $ \varepsilon $-constraint, differential evolution (DE) strategy, and feasibility rule. This approach could effectively balance the objective function and constraints, and could improve the ability of local exploitation and global exploration. Experiments on the CEC2006 and CEC2017 benchmark functions and real-world constraint optimization problems from the CEC2020 dataset showed that the ICPSO algorithm could effectively solve complex constrained optimization problems.</p></abstract>

Optimal wind power generation system by honey badger algorithm with differential evolution strategies

The purpose of this study is to build an optimal hybrid wind power system consisting of a permanent magnet direct-drive wind power generation unit, a hybrid energy storage system (HESS), a power electronic converter, and loads. Moreover, a reasonable control method is designed for each part, and a honey badger algorithm (HBA) with differential evolution strategies is employed to realize the coordinated control of each unit and improve the system stability. The real anti-interference capability of wind power system can be improved by the proposed HBA with differential evolution strategies (IHBA). Firstly, the wind hybrid system model is constructed, in which the wind power generation unit adopts the variable step climbing method to achieve the maximum power point tracking control; the HESS designs the power distribution method based on the bus voltage; the converter supplying energy to the AC load introduces the virtual synchronous generator (VSG) control strategy. Secondly, due to the existence of energy exchange in each unit of the system, VSG has more parameters and its control performance is influenced by the system itself, this study proposes HBA with differential evolution strategies for system parameter optimization and constructs an IHBA-VSG model with the objective of minimizing the bus voltage fluctuation value. The IHBA improves original honey badger algorithm by three mechanisms which are time control function, Gaussian variation factor and differential evolution strategy. Finally, the parameters obtained from the HBA with differential evolution strategies optimization search are substituted into the wind hybrid system, and the simulation system is built by MATLAB/Simulink. The simulation results show that when the environment of the system changes or the load on the AC side changes abruptly, the proposed control method can reduce the bus voltage fluctuation by 2.9%− 18.22% compared with the honey badger algorithm Optimized VSG (HBA-VSG), and can reduce the bus voltage fluctuation by 5.09%− 17.98% compared with the traditional droop control without considering the system interaction. This study can effectively improve the stability of wind power generation system and promote the development of new energy industry.

An equilibrium honey badger algorithm with differential evolution strategy for cluster analysis

Data clustering is a machine learning method for unsupervised learning that is popular in the two areas of data analysis and data mining. The objective is to partition a given dataset into distinct clusters, aiming to maximize the similarity among data objects within the same cluster. In this paper, an improved honey badger algorithm called DELHBA is proposed to solve the clustering problem. In DELHBA, to boost the population’s diversity and the performance of global search, the differential evolution method is incorporated into algorithm’s initial step. Secondly, the equilibrium pooling technique is included to assist the standard honey badger algorithm (HBA) break free of the local optimum. Finally, the updated honey badger population individuals are updated with Levy flight strategy to produce more potential solutions. Ten famous benchmark test datasets are utilized to evaluate the efficiency of the DELHBA algorithm and to contrast it with twelve of the current most used swarm intelligence algorithms and k-means. Additionally, DELHBA algorithm’s performance is assessed using the Wilcoxon rank sum test and Friedman’s test. The experimental results show that DELHBA has better clustering accuracy, convergence speed and stability compared with other algorithms, demonstrating its superiority in solving clustering problems.

A Novel Parameter Estimation Scheme for Vehicle Suspension Systems Based on Response and Test Track Prioritization

In this paper, a system identification methodology based on vehicle response and test track prioritization is presented. The proposed method can be used to perform time-domain parameter estimation by simply driving the vehicle on different road profiles, thus eliminating the necessity of having a special test setup or excitation input feed. The method is based on exploiting the dependence of the identifiability of vehicle parameters on vehicle responses and test tracks. To prioritize the test tracks, the responses and test tracks are weighed based on their capabilities to estimate the vehicle parameters, and these weights are obtained in a preliminary analysis of a seven-degree-of-freedom ride model developed in MATLAB® (R2021b). An optimization framework based on a differential evolution scheme with an optional archive scheme (JADE) is used to estimate the twelve vehicle parameters. The proposed system identification scheme is verified on tests using a high-fidelity vehicle model developed in ADAMS® (29.2)using several tracks with different severities. Experimental validation is performed on a test vehicle, and it is demonstrated that the vehicle suspension system parameters can be identified accurately with fast convergence performance.

Flood susceptibility mapping using AutoML and a deep learning framework with evolutionary algorithms for hyperparameter optimization

Flooding is one of the most common natural hazards that have extremely detrimental consequences. Understanding which areas are vulnerable to flooding is crucial to addressing these effects. In this work, we use machine learning models and Automated machine learning (AutoML) systems for flood susceptibility mapping in Kerala, India. In particular, we used a three-dimensional convolutional neural network (CNN) architecture for this purpose. The CNN model was assisted with hyperparameter optimization techniques that combine Bayesian optimization with evolutionary algorithms like differential evolution and covariance matrix adaptation evolutionary strategies. The performances of all models are compared in terms of cross-entropy loss, accuracy, precision, recall, area under the curve (AUC) and kappa score. The CNN model shows better performance than the AutoML models. Evolutionary algorithm-assisted hyperparameter optimization methods improved the efficiency of the CNN model by 4 and 9 percent in terms of accuracy and by 0.0265 and 0.0497 with reference to the AUC score.

Research on variable time-scale SOC and SOH asynchronous collaborative estimation strategy for electric vehicle power lithium iron phosphate batteries

Battery state parameters such as state-of-health (SOH) have slow time-varying characteristics, while state-of-charge (SOC) parameters have fast time-varying features. Using the same time scale algorithm framework to estimate the above state parameters simultaneously will waste BMS system calculation and cause severe parameter fluctuations due to frequent parameter updates, affecting the estimation accuracy. To address the above issues, this work proposes a variable time-scale SOC and SOH asynchronous collaborative estimation strategy for automotive power lithium iron phosphate batteries. Among them, the real-time changing battery state SOC is estimated through micro time scale filtering algorithms, the slowly changing battery parameter SOH is estimated through macro time scale filtering algorithms, and the time scale parameters that can minimize the cost function when maximizing noise are obtained through the robust optimization strategy of minimum maximization differential evolution. The results indicate that the asynchronous collaborative estimation strategy of variable time scale SOC and SOH proposed in this paper can effectively balance the accuracy of SOC estimation and SOH estimation.

Anti-sine-cosine atom search optimization (ASCASO): a novel approach for parameter estimation of PV models.

Nowadays, solar power generation has gradually become a part of electric energy sharing. How to effectively enhance the energy conversion efficiency of solar cells and components has gradually emerged as a focal point of research. This paper presents a boosted atomic search optimization (ASO) with a new anti-sine-cosine mechanism (ASCASO) to realize the parameter estimation of photovoltaic (PV) models. The anti-sine-cosine mechanism is inspired by the update principle of sine cosine algorithm (SCA) and the mutation strategy of linear population size reduction adaptive differential evolution (LSHADE). The working principle of anti-sine-cosine mechanism is to utilize two mutation formulas containing arcsine and arccosine functions to further update the position of atoms. The introduction of anti-sine-cosine mechanism achieves the populations' random handover and promotes the neighbors' information communication. For better evaluation, the proposed ASCASO is devoted to estimate parameters of three PV models of R.T.C France, one Photowat-PWP201 PV module model, and two commercial polycrystalline PV panels including STM6-40/36 and STM6-120/36 with monocrystalline cells. The proposed ASCASO is compared with nine reported comparative algorithms to assess the performance. The results of parameter estimation for different PV models of various methods demonstrate that ASCASO performs more accurately and reliably than other reported comparative methods. Thus, ASCASO can be considered a highly effective approach for accurately estimating the parameters of PV models.

Feature selection based on a multi-strategy African vulture optimization algorithm and its application in essay scoring

Reducing the dimensions of the original data set while preserving the information as much as possible is conducive to improving the accuracy and efficiency of the model. To achieve this, this paper presents a multi-strategy African vulture optimization algorithm that is the chaotic and elite opposition-based African vulture optimization with the simplex method and differential evolution strategy(CESDAVO). Three main improvements are introduced into African vultures optimization(AVO) to improve its capabilities in this study. Firstly, the chaotic elite opposition-based learning strategy is used to initialize and diversify individual positions of vultures. Secondly, the simplex method is used to optimize those poor individuals so as to further improve the local exploitation ability of the algorithm. Thirdly, the differential evolution strategy is used to make the algorithm escape from the local optimum and improve the global optimization capability of the algorithm. The results of the ablation experiments show that mixing the three strategies greatly improves the optimization performance of the algorithm. In addition, Nine algorithms are compared with CESDAVO on 15 benchmark functions, and this experimental result shows that its optimization capability is superior to the others. Then, the proposed CESDAVO is employed for feature selection, and 12 standard datasets are used for experiments. According to the experimental results, CESDAVO obtained the highest average classification accuracy on 11 datasets and the highest feature selection rate on 8 datasets, which is significantly better than other algorithms. Finally, CESDAVO is also applied to feature reduction for essays, removing 24 features and significantly improving the classification accuracy on multiple classifiers.

Evolutionary Multitasking for Optimization Based on Generative Strategies

Evolutionary multitasking (EMT) is one of the emerging topics in evolutionary computation. EMT can solve multiple related optimization tasks simultaneously and enhance the optimization of each task via knowledge sharing among tasks. Many EMT algorithms have been proposed and achieved success in various problems, yet EMT for multi-objective optimization remains a big challenge. The existing multi-objective EMT algorithms tend to suffer from slow convergence and difficulty in generating high-quality knowledge. To alleviate these issues, this paper proposes a new EMT algorithm namely EMT-GS for multi-objective optimization based on two generative strategies. Particularly, generative adversarial networks and inertial differential evolution are introduced to generate transferable knowledge and offspring, respectively. A generative adversarial network is trained periodically for each source-target task pair, based on which helpful knowledge is generated from the source task and transferred to boost the solving of the target task. To accelerate the population convergence, the inertial differential evolution strategy is put forward to generate offspring in a promising direction according to the individuals from the previous generation and the transferred knowledge. The performance of EMT-GS is validated on three multitasking multi-objective benchmark problems. The experimental results highlight the excellent competitiveness of EMT-GS compared to other state-of-the-art multi-objective EMT algorithms.

Reliable task allocation for soil moisture wireless sensor networks using differential evolution adaptive elite butterfly optimization algorithm.

Wireless sensor technology advancements have made soil moisture wireless sensor networks (SMWSNs) a vital component of precision agriculture. However, the humidity nodes in SMWSNs have a weak ability in information collection, storage, calculation, etc. Hence, it is essential to reasonably pursue task allocation for SMWSNs to improve the network benefits of SMWSNs. However, the task allocation of SMWSNs is an NP (Non-deterministic Polynomial)-hard issue, and its complexity becomes even higher when constraints such as limited computing capabilities and power are taken into consideration. In this paper, a novel differential evolution adaptive elite butterfly optimization algorithm (DEAEBOA) is proposed. DEAEBOA has significantly improved the task allocation efficiency of SMWSNs, effectively avoided plan stagnation, and greatly accelerated the convergence speed. In the meantime, a new adaptive operator was designed, which signally ameliorates the accuracy and performance of the algorithm. In addition, a new elite operator and differential evolution strategy are put forward to markedly enhance the global search ability, which can availably avoid local optimization. Simulation experiments were carried out by comparing DEAEBOA with the butterfly optimization algorithm (BOA), particle swarm optimization (PSO), genetic algorithm (GA), and beluga whale optimization (BWO). The simulation results show that DEAEBOA significantly improved the task allocation efficiency, and compared with BOA, PSO, GA, and BWO the network benefit rate increased by 11.86%, 5.46%, 8.98%, and 12.18% respectively.

Hybrid intelligent deep learning model for solar radiation forecasting using optimal variational mode decomposition and evolutionary deep belief network - Online sequential extreme learning machine

Accurate prediction of solar radiation is of great significance to improve the utilization of solar energy for photovoltaic power generation on the roofs of urban buildings. Therefore, a hybrid model, namely the OVMD-PACF-ISSA-DBN-OSELM model, is being proposed in this study for solar radiation prediction. Firstly, the data is decomposed by optimal variational mode decomposition (OVMD) technology. Secondly, the partial autocorrelation function (PACF) technology selected features for components decomposed by OVMD technology. Thirdly, a deep belief network-online sequential extreme learning machine (DBN-OSELM) prediction model is constructed, and sparrow search algorithm (SSA) is utilized to optimize the model hyperparameters. Aiming at the shortcomings of SSA algorithm, Tent map and differential evolution strategy are introduced to improve the SSA algorithm. Finally, the model is used to predict each component, and the prediction results of each component are summarized to obtain the prediction results of solar radiation. To verify that the proposed method can obtain accurate prediction results, experiments are conducted using solar radiation data for January, May and October 2022, and compared with other models. Taking January as an example, compared with shallow model BP, the RMSE, MAE and R of the proposed model are increased by 66.83%, 58.65% and 5.33%, respectively. Moreover, the experiments of the proposed model in other months have excellent results. The current findings demonstrate that the suggested model is a strong contender in the arena of solar radiation prediction, and can consequently provide valuable support in enhancing the efficiency of photovoltaic equipment installed on urban building roofs.