HARMONIC CURRENT COMPENSATION IN ACTIVE POWER FILTER BASED ON ADAPTIVE NEURAL NETWORK

Abstract

an essential role is played by Shunt active power filters (SAPF) in harmonic elimination as well as reactive power compensation in power systems with a large concentration of non-linear loads. Commonly, the (SAPF) efficiency depends on three design criteria: (1) types of current controllers used; (2) methods used to obtain the reference current; (3) DC link voltage control methods. In this paper a comparison between two strategies for extracting the three-phase reference currents for shunt active power filters is shown ;artificial neural network (ANN) and p-q theory. The two strategies performance will be evaluated at the supply current quality rank and system stability at DC link capacitor which ensures suitable transit of powers to supply the inverter. The study has been accomplished using simulation with the MATLAB Simulink Power System Toolbox. The simulation study results of the new identification technique (ANN) are found quite satisfactory by assuring good filtering characteristics and high system stability.