On the Accuracy of Correlation Estimator Based Fault Location Methods in Transmission Lines via Polynomial Chaos and Regression Analysis

Abstract

The electromagnetic time reversal (EMTR) method has recently shown a novel advantage in locating faults along transmission lines. Under the ideal condition for EMTR, the key parameters in the transmission line between the direct and reversed phases are assumed to be identical. Thus, the fault position can be pinpointed via the cross-correlation matching theory. However, due to natural and geographical causes, the parameters vary along the line; therefore, accurately obtaining the parameters is difficult. In this article, we first investigate the impact of uncertain parameters on the performance of EMTR-based fault location methods. It has been observed that the performance is insensitive to the variation of the parameters in the lossless case of transmission lines. But in the case of the overhead line above the lossy ground, a significant error in fault location is witnessed. So, under parameter uncertainty, a novel approach to finding the confidence fault interval is proposed, based on utilizing the polynomial chaos and regression analysis. A simulation case with uncertain parameters for different scenarios is presented to validate the performance of the proposed approach.