Artificial Neural Network Based Duty Cycle Estimation for Maximum Power Point Tracking in Photovoltaic Systems

According to a nonlinear current-voltage characteristic of Photovoltaics (PV) we need to track maximum power output of PV generation units instantly. The aim of this paper is to introduce a non-complicated method for tracking the maximum Power Point without any previous knowledge of the physical parameters linked with a Grid-Connected photovoltaic (PV) system using artificial neural networks (ANN) modelling. The ANN is trained in various conditions of PV Output Voltage and PV Output Current to forecast the Duty Cycle of DC-DC boost converter as the MPPT device. The proposed technique is implemented in Matlab/Simulink and compared with the conventional method of incremental conductance. Simulation results show a good performance of the ANN based MPPT controller. MPPT techniques that properly detect the global MPP has been widely investigated in the literature. They include hill climbing (HC), incremental conductance (IncCond), perturband-observe (P&O), and fuzzy logic controller (FLC). As the best of our knowledge estimation of the duty cycle of the DC-DC boost converter by Artificial Neural Network and using it in place of the whole MPPT controller and using Voltage and current has not been done so far in the literature.