Extremal Nelder–Mead colony predation algorithm for parameter estimation of solar photovoltaic models

Abstract Measurement data based on current and voltage of photovoltaic (PV) systems the establishment more accurate stable solar system models are typical significance for design, control, evaluation optimization PV systems. Accurate parameter needs to be efficient techniques achieve energy conversion from energy. Therefore, this paper proposes a novel technique enhanced colony predation algorithm solve complex identification problem named ECPA. By fusing extremal strategy Nelder–Mead simplex method enables ECPA further develop in neighborhood potential optimal solutions while improving position inferior agent candidates, finally has ability search globally beyond local optimum. To verify efficiency ECPA, first part verifies solving high‐dimensional multimodal problems by conducting competitive comparison experiments at IEEE CEC 2020 benchmark case. In second part, is compared with nine similar published state‐of‐the‐art algorithms, tests under single diode model, double triple model module conducted. Finally, we focused three different commercial (thin film ST40, monocrystalline SM55, multicrystalline KC200GT) test accuracy evaluating parameters. The results show that able maintain high level stability when dealing environments. experimental demonstrate outperforms other algorithms terms fitting, stability, convergence speed accuracy. All can best performance identifying parameters determined