Robust estimation of photovoltaic model parameters under varying conditions using novel nature-inspired optimization algorithms

The growing need for reliable PV cell and module mathematical modeling has recently become an important issue. Although the common mathematical models of a PV cell/module, Single Diode Model (SDM), Double Diode Model (DDM), and Triple Diode Model (TDM), require a high degree of parameter identification accuracy, this article highlights the use of meta-heuristic optimization techniques, namely the Horse Herd Optimizer (HOA) and the Elk Herd Optimizer (EHO), for PV parameter identification. The proposed framework is tested on SDM, DDM, and TDM and is validated using real-world I-V measurements of actual devices such as the RTC France solar cells and Photowatt-PWP201 modules. A rigorous test has been carried out among existing traditional metaheuristics to compare efficiency, accuracy, and convergence capabilities. The simulation results indicate that HOA and EHO converge better and give more accurate results concerning the parameter values. Further, to evaluate the robustness of this method in practical scenarios, various simulations are performed at different temperatures and irradiance levels. From the above-discussed results and simulation studies presented in this paper, it is concluded that the proposed optimization approach provides an effective solution for PV parameter estimation.