Modern methods for power system harmonics estimation

Abstract

Harmonics has been present for a long time and its presence shapes the performance of a power system. Therefore, estimation of harmonics is of paramount importance while analyzing a power system network. Following the inception of harmonics, various filters have been devised to achieve an optimal control strategy for harmonic alleviation. This thesis introduces various algorithms to analyze harmonics in the power system. The objective is to estimate the power system voltage magnitude in the presence distortions taking into account the noise by employing different estimation approaches. We have focused our attention towards the study of Least Mean Squares (LMS) based filter, Recursive Least squares (RLS) based filter, Kalman filter (KF) and Extended Kalman (EKF) filter. For a test signal LMS, RLS, KF and EKF based algorithms have been analyzed and results have been compared. The proposed estimation approaches are tested for only static signals.