In this paper, a simulation study of the maximum power point tracking (MPPT) for a photovoltaic system using an artificial neural network is presented. Maximum power point tracking (MPPT) plays an important role in photovoltaic systems because it maximizes the power output from a PV solar system for all temperature and irradiation conditions, and therefore maximizes the power efficiency. Since the maximum power point (MPP) varies, based on the PV irradiation and temperature, appropriate algorithms must be utilized to track it in order maintain the optimal operation of the system. The software Matlab/Simulink is used to develop the model of PV solar system MPPT controller. The system simulation is elaborated by combining the models established of solar PV module and a DC/DC Boost converter. The system is studied using various irradiance shading conditions. Simulation results show that the photovoltaic simulation system tracks optimally the maximum power point even under severe disturbances conditions.

This paper presents an intelligent control strategy that uses a feedforward artificial neural network in order to improve the performance of the MPPT (Maximum Power Point Tracker) MPPT photovoltaic (PV) power system based on a modified Cuk converter. The proposed neural network control (NNC) strategy is designed to produce regulated variable DC output voltage. The mathematical model of Cuk converter and artificial neural network algorithm is derived. Cuk converter has some advantages compared to other type of converters. However the nonlinearity characteristic of the Cuk converter due to the switching technique is difficult to be handled by conventional controller. To overcome this problem, a neural network controller with online learning back propagation algorithm is developed. The NNC designed tracked the converter voltage output and improve the dynamic performance regardless load disturbances and supply variations. The proposed controller effectiveness during dynamic transient response is then analyze and verified using MATLAB-Simulink. Simulation results confirm the excellent performance of the proposed NNC technique for the studied PV system.