Shunt Active Power Filter based on Artificial Neural Networks for Grid-connected Photovoltaic system

The aim of this paper is to improve the quality of the energy generated by photovoltaic grid-connected system. In order to compensate harmonic currents caused by nonlinear loads and to reduce total harmonic distortion (THD), shunt active power filter, based on multilayer feed forward neural networks and fuzzy logic techniques, is used. The performance of the proposed control model is verified by MATLAB based simulation results. Results are compared with the instantaneous reactive power technique results.