Comparative analysis of LMS, block LMS and RLS adaptive control methods for shunt voltage source converter

Abstract

This paper gives the comparative analysis of the adaptive control mechanism which are meant for shunt connected voltage source converter (SVSC). The adaptive control mechanism driven in such a way to improve the quality of electric power under non-linear loads. The authors have implemented the three adaptive control algorithms, the adaptive control algorithm has a filtering and adaption section, the adaption section consists of least mean square (LMS), second algorithm is block LMS and then an algorithm recursive least square (RLS) finds the filter weights recursively manner. The reference current is estimate using these adaptive filter algorithms and compare with the SVSCcurrents to generate the required switching signals. The algorithms are compared based on several points, mainly convergence time of the weighted vectors, steady state error of the incorporated adaptive algorithm and finally the total harmonic distortion (TlID) in the source current. The investigation on the algorithm are performed on the Matlab/Simulink platform under nonlinear load and dynamic load behaviour.