Adaptive neuro-fuzzy inference system based maximum power point tracking for stand-alone photovoltaic system

Abstract

The maximum power point tracking (MPPT) plays a very important role to extract the maximum power of the photovoltaic (PV) system by ensuring its optimal production under sunshine and temperature variations. This study presents an algorithm based MPPT named an adaptive neuro-fuzzy inference system (ANFIS) which is built with the combination of the artificial neural network (ANN) and the fuzzy logic controller (FLC). The efficiency of the ANFIS algorithm is tested under Matlab/Simulink and compared with the fixed step conventional perturb and observe (P&O) and the gradient descent techniques under temperature and irradiance change. The obtained results showed a significant improvement in performances of the PV system using the ANFIS-MPPT technique which provides also faster convergence, stability in steady state, less oscillations around the MPP and higher efficiency to track the maximum power from the PV system compared to other techniques under different operating conditions.