A population diversity-controlled differential evolution for parameter estimation of solar photovoltaic models

The cleanliness and renewability of photovoltaic (PV) system make it stand out as a promising energy source. However, limited by the developments of PV equipment, the conversion efficiency of solar energy is still very low. In order to further improve the conversion efficiency, an accurate model with well-estimated parameters is much more important for PV systems. Motivated by this demand, we propose a new differential evolution variant (PDcDE) to tackle the parameter estimation of several kinds of solar PV models. Innovations in this paper include an auto-controlled population strategy to adjust the population size during a search process, a diversitycontrolled parameter setting method to decide the scale factor, and a backward search to avoid local optima. The performance of PDcDE is verified on six PV modules against eleven state-of-the-art meta-heuristic algorithms. Extensive comparative results reveal the excellent performance of PDcDE. Moreover, the results analyzed by two statistical methods report its superior efficiency and effectiveness for the parameter estimation of PV systems.

Automated summary

This study proposes a novel differential evolution variant (PDcDE) for the parameter estimation of solar photovoltaic (PV) models. The PDcDE algorithm features an auto-controlled population strategy, a diversity-controlled parameter setting method, and a backward search to enhance its performance. Comparative results demonstrate the superior efficiency and effectiveness of PDcDE in the parameter estimation of PV systems.