Comparative study between three methods for optimizing the power produced from photovoltaic generator

Abstract

The technics of maximum power point tracking is widely used in solar photovoltaic energy and electric power system applications. Traditionally, these technics are based on conventional methods like perturb and observe and incremental conductance. In this work, three methods based on particle swarms optimization, incremental conductance and adaptive neuro fuzzy inference system are presented. A comparative study is carried out. The study of this paper shows that there is a limitation in the incremental conductance method. To overcome the shortage of this last method, particle swarms and adaptive neuro fuzzy optimization methods are used. The behaviors of the three methods are compared and evaluated in simulation under matlab/simulink. Results demonstrate that the adaptive neuro fuzzy inference system is effective for photovoltaic power optimization even for non-uniform climatic conditions. It has the best performances followed by the particle swarms method.