Development of incremental average differential evolution algorithm for photovoltaic system identification

Abstract

The determination of the behavior of photovoltaic (PV) systems is a current subject of research due to the increase in their share in electricity production. This identification problem is generally defined as the estimation of the unknown parameters of the equivalent circuit model. The parameters of the PV model are optimized by minimizing the error between the measured data from the actual PV cell and the results of the model. An efficient optimizer tool is required to obtain the best model’s parameter. This paper presents a novel metaheuristic named incremental average differential evolution algorithm (IncADE) for parameter estimation of PV models. The IncADE is a new variant of average differential evolution (ADE) that enhances the global search ability of ADE algorithm by the incremental population strategies. The performance of the developed IncADE is firstly evaluated on well-known benchmark functions, and the experimental results show that the proposed method improves the accuracy of the concluding solutions and the convergence performance of the basic ADE. Then, the IncADE is employed to estimate the optimal parameters of different PV models, which are single diode, double diode and PV module. Experimental results prove the superiority of IncADE on parameter estimation in terms of accuracy and computational efficiency by comparing with ADE and other metaheuristic algorithms.