Hybrid salp swarm gaining and sharing knowledge (ssa-gsk) metaheuristic algorithm for extracting photovoltaic cell parameters

Abstract

Maximizing the energy output of photovoltaic cells depends firstly on accurately estimating their parameters using one of the three models. This parameter estimation is a multi-variable, nonlinear, and multi-modal problem. Developing more accurate and advanced solutions for this remains a challenging research problem. This study aims to develop a hybrid algorithm to precisely extract the parameters from photovoltaic cells and improve the results of the Salp Swarm Algorithm (SSA). To achieve this goal, a hybridization approach is implemented by combining two metaheuristic algorithms, SSA and Gaining and Sharing Knowledge (GSK), to extract parameters. The hybrid SSA-GSK algorithm was tested on RTC France photovoltaic cells, amorphous silicon photovoltaic cells (aSi), and STM6-40 using single and double diode models. The performance of the SSA-GSK hybrid algorithm is comparatively analyzed using the root mean square error (RMSE) and statistical analysis. RMSE was calculated using a hybrid SSA-GSK algorithm and obtained as 0.00261 for RTC SDM and 0.00844439 for RTC DDM, 6.3614 10-05 for aSi SDM and 9.4936 10-05 for aSi DDM, 2.0349 10-3 for STM6-40/36 SDM and, 2.0740 10-3 for STM6-40/36 DDM. The results demonstrated that the hybrid SSA-GSK algorithm achieved a lower rate of RMSE in comparison to SAA and other literature-reported algorithms. The algorithm was more proficient in search space, leading to superior optimization results in lesser execution time after hybridization. The proposed algorithm increases the probability of generating high-quality solutions and outperforms some other optimization algorithms while maintaining ease of implementation.