A novel hybrid model-based MPPT algorithm based on artificial neural networks for photovoltaic applications

In this paper, a novel hybrid Maximum Power Point Tracking (MPPT) strategy has been proposed. This algorithm employs an idea similar to the Fractional Short Circuit Current (FSCC) method in order to improve the performance of the conventional Perturbation and Observation (P&O) algorithm. It should be considered that in the conventional FSCC method, the short circuit current needs to be measured at different time instances and hence, the power generated by PV module is not permanently delivered to the load. In the proposed strategy, a model of the PV cell is obtained based on a Multi-Layer Perceptron (MLP) structure. Then, the PV short circuit current is easily approximated using this model and is employed as an auxiliary information for the conventional P&O algorithm in order to simultaneously achieve fast and accurate tracking performance. The output of the proposed algorithm is the reference current value which is then fed to a predictive controller used to control the DC-DC boost converter placed between the PV cell and the load. Finally, the system is simulated in the MATLAB/Simulink environment. The simulation results verify the appropriate performance of the proposed method.