Performance Evaluation of P&O, ANN and ANFIS Based MPPT Controllers of PV Array Under Partial Shading Effect in South Pole and Leading Solar Sites

Abstract

Fuel oil is a primary source used by the research stations located at Antarctica. However, shipping fuel oil to this remote region is costly and harms the environment. Due to global warming, it has become a necessity to implement green energy technologies. Due to the nonlinear nature of PV cells and their dependence on solar irradiation and temperature, obtaining maximum power is a challenge. The system must be optimized to obtain the maximum power using Power Electronics. One way of achieving this is called Maximum Power Point Tracking (MPPT). Our research evaluates the performance of conventional method, Perturb and Observe (P&O) based MPPT technique, with modern techniques, Artificial Neural Network (ANN) and Adaptive neuro-fuzzy inference system (ANFIS) based MPPT techniques. Simulations of the ANN, ANFIS and P&O algorithms were carried out using MATLAB-Simulink. The comparison between these controllers is made considering the partial shading and their performance at different temperatures at the leading solar sites and South pole. Results indicate that there is an improvement in MPP tracking for both ANN and ANFIS controllers as compared to the P&O algorithm with respect to the settling time, overshoot, oscillations and time to achieve MPP at both environments.