Gabor transform based fault locator for transmission lines

Accurate estimation of current and voltage transient parameter is critical for efficient and accurate fault location computation. In this paper, a new fault location scheme is proposed for transmission systems using Gabor transform for signal processing purposes instead of the conventional Fourier methods. The transform is distinctive with more accurate performance specially when dealing with some certain circumstances such as sudden signal changes, dc decaying, non integer harmonics and non stationary quantities for fault signals. For a better extraction of Gabor coefficients, a dedicated artificial neural network is employed. The contribution of this new transform to power system fault location is evaluated through various simulation tests using the Electromagnetic Transient Program EMTP. Simulation results show the potential of the proposed transform in accurate estimation of phasors for more accurate fault location estimation.