Comparative study between four intelligent algorithms to track the Maximum Power Point of Stand-Alone Photovoltaic System

The first step in the optimization of the PV system performance deals with getting the maximum power available from the PV generator. This is done by tracking the Maximum Power Point (MPPT) of the PV generator. Beside the conventional techniques like the Perturb and Observe (P&O), Incremental Conductance (IC) and the Hill Climbing (HC), soft computing techniques namely fuzzy logic (FL), artificial neural networks (ANN), adaptive neuro-fuzzy inference system (ANFIS) and FL-optimized by genetic algorithm (FL-GA) have been nowadays widely used in photovoltaic applications for MPPT due to their better efficiency. In this article, we present four controllers that have been developed with intelligent techniques (FL, ANN, NF, FL-GA) for tracking the MPP in Stand Alone PV system (750Wp silicon monocrystalline modules, 550Ah batteries bench, DC-DC converter and static load) under uniform and variable irradiation and temperature conditions. Simulations under Matlab Simulink environment have been conducted using a real database (irradiation and temperature) gathered with pyranometers and weather station installed at UDES site (Bou-Ismail, Algeria). A comparative study between the developed controllers based on critical parameters like computational requirement, speed tracking, tracking efficiency and implementation complexity will be discussed. Among the developed controllers, the FL-GA controller appears as the best as it is the fastest, with insignificant fluctuation even in steady state or time varying situation.