New Photovoltaic Module Model And A Comparative Study of MPPT Control Techniques Based On Neural Networks

Abstract

Nowadays, renewable energy resources play an important role in replacing conventional fossil fuel energy resources. Solar photovoltaic (PV) energy is a very promising renewable energy resource, which rapidly grew in the past few years. The main problem of the solar photovoltaic is with the variation of the operating conditions of the array, the voltage at which maximum power can be obtained from it likewise changes. In this paper, is first performed the modelling of a solar PV panel using MATLAB/Simulink. After that, a maximum power point tracking (MPPT) technique based on artificial neural network (ANN) is applied in order to control the DC-DC boost converter. This MPPT controller technique is evaluated and compared to the “perturb and observe” technique (P&O). The simulation results show that the proposed MPPT technique based on ANN gives faster response than the conventional P&O technique, under rapid variations of operating conditions. This comparative study is made in terms of temporal variations of the duty cycle (D), the output power ( out P ), the output current ( out I ), the efficiency, and the reference current ( ref I ). The efficiency, D, out P , and out I are the output of the boost DC-DC, and ref I is itsinput. The different temporal variations of the efficiency, D, ref I , out P , and out I (for the two cases: the first case, when T = 25°C and G =1000 W/m2 and the second case, when T and G are variables), show negligible oscillations around the maximum power point. The used MPPT controller based on ANN has a convergence time better than conventional P&O technique.