Improved adaline based algorithm for power system frequency estimation

Abstract

This paper intends to present an adaptive algorithm for estimating the frequency, amplitude, and phase of a sinusoid under non stationary condition present in time-varying power signals. The proposed algorithm estimates precisely the frequency variation, phase variation, and the amplitude and shows accuracy in estimation even in the presence of harmonic and inter harmonic as noise. This method uses Taylor series expansion of the signal to cope with the sudden changes. Then a modified ADALINE is used because of its low computational complexity, for which it can be implemented in real time. The performance of the proposed algorithm has been extensively tested and demonstrated.