Neural Network based Maximum Power Point Tracking Technique for PV Arrays in Mobile Applications

Abstract

Maximum power point tracking (MPPT) of photovoltaic (PV) arrays is an essential concern to enhance the efficiency of the whole PV system. Under partially shaded conditions (PSC) that all modules do not receive uniform illumination, the tracking turns out to be challenging, due to the output power-voltage characteristic of the PV array exhibits multiple peaks. In mobile applications, PSC becomes more troublesome since the partial shading patterns change very fast. Therefore the tracking of the MPP should be quick and precise. In this paper a two-stage MPPT Method that combines Artificial Neural Network (ANN) and Hill Climbing (HC) is presented. In the first stage an ANN estimates the vicinity of the MPP and in the second stage, HC is performed to obtain the exact MPP. The approach is very fast which makes it suitable for mobile applications and is able to extract maximum power under uniform irradiation and PSC. The validity of the proposed method is investigated by simulations in MATLAB/Simulink environment. The simulation results show that the proposed method provides a quick and accurate tracking.