Optimization of Solar Energy Using ANN Techniques

Abstract

The adaptation of maximum power point tracking algorithm for photovoltaic systems is of immense importance. In this paper, solar photovoltaic system is developed using MATLAB/SIMULINK which is connected to utility, artificial neural network (ANN) is used as maximum power point tracking (MPPT) algorithm for generation of maximum power by the system. Proportional-Integral (PI) controller that supports the information provided by ANN and generates boost converter duty-cycle helps in improving overall system stability. MPPT is evolved from a highly efficient boost converter where artificial neural network (ANN) with Levenberg-Marquardt (LM) algorithm is used for generating reference voltage for MPPT using feed forward back propagation training algorithm; error calculation is done by using the concept of mean square error algorithm. The developed system is verified under different test conditions and its control strategy shows exceptional performance having tracking efficiency exceeding 94.5%. Developed MATLAB/SIMULINK model offers control strategy and tool for the optimization of solar PV system connected to utility.