Multiple peaks tracking for photovoltaic system using particle swarm optimization with artificial neural network algorithm

Abstract

Photovoltaic (PV) array may receive different level of solar irradiance, such as partially shaded by clouds or nearby building. Multiple peak power points occur when PV module is under partially shaded conditions, which would significantly reduce the energy produced by PV without proper control. Therefore, Maximum Power Point Tracking (MPPT) algorithm is used to extract maximum available PV power from the PV array. However, most of the conventional MPPT algorithms are incapable to detect global peak power point with the presence of several local peaks. A hybrid Particle Swarm Optimization and Artificial Neural Network (PSO-ANN) algorithm is proposed in this thesis to detect the global peak power. The PV system which consists of PV array, dc-dc boost converter and a resistive load, were simulated using MATLAB/Simulink. The performance of the proposed algorithm is compared with that of the standard PSO algorithm. The proposed algorithm is tested and verified by hardware experiment. The simulation results and the experimental results are compared and discussed. It shows that the proposed algorithm performs well to detect the global peak of the PV array under partially shaded conditions. In this work, the tracking efficiency of the proposed algorithm is in the range of 96.8 % to 99.7 %.