Mobile QR Code QR CODE : The Korean Institute of Power Electronics
Title Improving the Output Current of Matrix Converter under Abnormal Input Voltage Conditions using a Neural Network Compensator
Authors 李Eunsil Lee ; Kiwoo Park ; 李Kyo-Beum Lee
DOI 10.6113/TKPE/2010.15.3.199
Page pp.199-206
ISSN 1229-2214
Keywords Matrix converter; Neural network; Current compensator; Abnormal input voltage; Compensation
Abstract Matrix converter is an energy conversion device of controlled power semiconductor switches that directly connects the three-phase source to the three-phase load. With no dc-link components for energy storage in the matrix converter the input current depends directly upon the load currents and the switch state of the converter. Therefore the unbalanced and distorted input voltages can result in unwanted output harmonic currents. This paper presents a current compensator based on neural network to improving output current quality for matrix converter under abnormal input voltage conditions. The effectiveness and feasibility of the proposed technique has been proven through numerical simulations and experimental tests.