IDENTIFICATION OF HARMONIC CURRENTS BY NEURAL MVF METHOD USED IN SHUNT ACTIVE POWER FILTER CONTROLLED BY SVM

Abstract

This paper presents a new approach for the identification of harmonic currents that MVF method called the neural, architecture chosen and the type of network occurred, We exposed the Feedforward neural by the backpropagation algorithm and system based on the neural adaptation technique with non-linear loads of variable electrical networks ,neural MVF Used for generate and extract the reference currents which should be injected by three level inverter is presented after the identification using the three-level inverter controlled by space vector modulation (SVM) for minimizing the harmonics of source currents, The regulation of the current is accomplished by PI regulators all results achieved by digital simulation using MATLAB environment.