Identification of a five-parameter model for a photovoltaic module using hybrid particle swarm optimization and pattern search

Abstract

This paper aims at determining the five parameters of the single-diode model of a photovoltaic (PV) module based on data-sheet parameters. For this purpose, a system of five non-linear analytical equations free of any simplifications, describing the PV module in three main operating points, has been formulated as a constrained nonlinear optimization problem. The latter has been solved by using a hybrid Particle Swarm Optimization and Pattern Search (HPSOPS) algorithm. The proposed procedure with the hybrid evolutionary algorithm has proven its effectiveness and reliability to estimate the five parameters, and its ability to reproduce the PV module characteristics for different solar irradiance and temperature conditions. The impact of solar irradiance and temperature variations on the single-diode parameters values has been also highlighted.