Design of an intelligent MPPT based on ANN using a real photovoltaic system data

Abstract

Maximum power point tracking (MPPT) methods are a fundamental part in photovoltaic (PV) system design for increasing the generated power of a PV array. Whilst several methods have been introduced, the artificial neural network (ANN) is an attractive method for MPPT due to its less oscillation and fast response. However, accurate training data is a big challenge to design an optimized ANN-MPPT technique. In this paper, an ANN-MPPT technique based on a large experimental training data is proposed to avoid the system from having a high training error. Those data are collected during one year from experimental tests of a PV system installed at Brunel University, London, United Kingdom. The irradiation and temperature of weather conditions are selected as the input, and the available power at MPP from the PV system as the output of the ANN model. To assess the performance, the Perturb and Observe (P&O) and the proposed ANN-MPPT methods are simulated using a MATLAB/Simulink model for the PV system. The results show that the proposed ANN method accurately tracks the optimal maximum power point and avoids the phenomenon of drift problem, whilst achieving a higher output power when compared with P&O-MPPT method.