Compare the Convergence Behavior of DE and PSO Optimization Algorithms for Parameter Extraction in DDM Equivalent Circuit for the PV Panels

Abstract

Accurate equivalent circuit parameter estimation for solar cells can significantly provide actionable insights for photovoltaic (PV) system designers. In this paper, we present a comparative analysis of two commercial PV modules, Jinko JKM365M (monocrystalline) and Canadian Solar CS3U-365PB-FG (polycrystalline bifacial), tested under distinct real-world environmental conditions, which were conducted under high irradiance with elevated temperature (1000 W/m², 65°C) and low irradiance with moderate temperature (200 W/m², 25°C). A rigorous preprocessing pipeline was applied to enhance data quality and ensure the reliability of the extracted parameters. Both Differential Evolution (DE) and Particle Swarm Optimization (PSO) were implemented in MATLAB and evaluated based on convergence behavior, Root Mean Square Error (RMSE), and alignment with manufacturer specifications. The key findings emphasized the performance of the optimization algorithms and the accuracy of the models. This study makes several noteworthy contributions to the field of photovoltaic modeling, particularly in the context of parameter estimation using field-measured data. One of the primary achievements lies in validating the efficacy of DE as a superior optimization algorithm for PV applications. The findings contribute to more accurate PV modeling, improved system diagnostics, and enhanced design and control of solar energy systems.