Artificial Neural Network Based MPPT Controller for Stand-alone Solar PV System

Abstract

The world s attention on renewable energy supplies has been spurred by the rise in the need for power and the limited availability of fossil fuels. Solar Photovoltaic (PV) energy is thought to have the capacity to meet the rising energy demand due to its abundance. To increase efficiency, the maximum power point of PV array output power is calculated. Traditional Maximum Power Point Tracking (MPPT) techniques have a simpler structure and implementation, but they perform poorer than AI-based systems. Artificial neural network (ANN) approaches include offline training, nonlinear mapping, high-speed response, robust operation, reduced processing effort, and quick results for multiple-variable difficulties. The purpose of this study is to create an intelligent system that uses Artificial Neural Networks to extract a PV array s Maximum Power Point. To improve the control of MPPT for PV systems, this system uses an RBFN architecture. The non-linear output characteristics of a solar array are influenced by insolation, temperature changes, irradiance, and the optimum operating point. As a result, to get the most power out of the system, this must be monitored. For best efficiency, the MPPT controller s output can be used to supervise the DC-DC boost converters.