A Neural Network MPPT For Photovoltaic Grid Connected System

Abstract

A photovoltaic (PV) output power depends nonlinearly on the climatic changes such au solar irradiation and temperature, there exists a unique maximum power point (MPP). Thus, a maximum power point tracking (MPPT) controller is needed to continuously extract the highest possible power the PV panel to increase the efficiency of such as systems. Many papers have presented different techniques to track the MPP, Base on the conventional methods, as perturb and observe (P&O) and incremental conductance (INC), or artificial intelligence (AI) techniques which they are becoming useful as alternate approaches to conventional techniques and they can be integrated easily in such as systems. In this paper an artificial intelligence technique is presented based on a feed forward neural network. The inputs of the MPP tracker are the sort circuit current and the open circuit voltage of the photovoltaic panel, however the output is the optimum current. This output is taken as a reference by the control algorithm of the grid connected inverter, to inject an optimal sinusoidal wave form into the utility. It is proved through experimentally results that the algorithm has a good performances and very fast response to climatic changes. Simulation and experimental results are presented to prove the efficiency of the proposed technique