Kung's component extraction in power system fault location

Abstract

This paper presents a method for the accurate estimation of phasors in fault location applications. One of the essential tasks of power system digital relays is to decrease the consequences that follow faults by fast detection and localization. A fault signal (current and voltage) is typically composed of the fundamental component in addition to a damped DC signal, together with damped harmonics and inter-harmonics. A significant challenge in impedance-based fault analysis is the accurate estimation of the fundamental frequency component. While different variations of the Fourier-based algorithm are the most commonly used in the estimation of the fundamental component, their accuracy tends to be sufficient for protection relaying schemes but not for fault location. This paper proposes a novel fault feature extraction method that is highly accurate; it combines Kung's method for signal decomposition followed by a nonlinear least-squares estimation. The proposed technique has been successfully evaluated in the context of a fault location method; the numerical tests show that the proposed two-stage approach can accurately estimate phasor information in the presence of transients and noise. © 2020 Elsevier Ltd