A Modified Least Mean Square Technique for Harmonic Elimination

Abstract

A modified least mean square (MLMS) technique-based shunt active filter is proposed for selective harmonic elimination in a grid integrated nonlinear system applications. To provide a better and cost-effective solution, a reduced switch voltage source inverter (RSVSI) is used for a three-phase active filter operation. In addition to that, for better power quality, a MLMS technique is applied to separate the positive and negative sequence components from the non-linear load current. The MLMS technique improves its performance by estimating the appropriate current error through an adaptive step size selection. To validate the MATLAB/Simulink designed the proposed approach, different test conditions are presented. From the comparative analysis, the proposed system shows that the implementation of the novel technique leads to a substantial gain in the signal to noise ratio, by which the distribution system reduces the output error and significantly improved the convergence speed through an adaptive harmonic cancellation.