Identification of Solar Cell Model Parameters based on PSO with Adaptive Elite Mutation

Abstract

Solar cell model parameters are of great significance in the study of solar cell fault model, power prediction and the maximum power point tracking. Therefore, it is very necessary to study an fast and accurate algorithm for solar cell model parameters identification. Based on characteristics of solar cell model, a solar cell parameters identification method based on particle swarm optimization (PSO) with adaptive elite mutation (AEM) is studied. Traditional evolutionary computation methods have the disadvantage of premature convergence, which will affect the accuracy of identification. In order to accelerate the searching speed, we select elite particles to replace inferior particles in the process of evolution. Meanwhile, adaptive mutation strategy can avoid the risk that particle swarm will lose its diversity and fall into local optimal solution. Combining elite particle and adaptive mutation strategy, we overcome contradiction between exploration and development of the optimization algorithm. The experimental results show that the particle swarm optimization with adaptive elite mutation (PSO-AEM) has a good effect and processing speed for identification of solar cell model parameters.