An improved method for MPPT using ANN and GA with maximum power comparison through Perturb & Observe technique

The solar photo voltaic (PV) systems manifest their utility in a number of ways and have emerged as one of the promising renewable sources of electrical power. Solar PV array has a non-linear characteristic. The voltage across the output terminals of the PV array and its internal resistance vary along with changes in the ambient conditions of temperature and insolation. As irradiation and temperature falling onto the panel vary, its voltage across the load as well as internal resistance varies. The aim of this paper is to simulate the maximum power point tracking (MPPT) of solar PV module with Perturb and Observe (P&O) technique and compare the results with those of genetic algorithm (GA) optimized artificial neural network (ANN) based MPPT approach. In this study the Bayesian regulation based ANN algorithm is used in predict the maximum power points at given values of temperature and irradiance. The data obtained from optimization technique using GA is used to train ANN. The simulation is carried out using MATLAB-Simulink. The results obtained show that the GA optimized ANN based approach meets the requirement of the optimum power supply to the load and attenuates the fluctuations around this operating point.