Underground power cable fault detection using complex wavelet analysis

Abstract

Fault detection in underground power cables can be accomplished by examining voltage and current signals using signal processing techniques. In this paper, we explore the feasibility of applying complex wavelet analysis to fault detection. We combine complex wavelets with continuous wavelet transform (CWT), and calculate the impedance from the voltage and current data in the wavelet domain. We then examine the magnitude and phase distributions of the impedance under various conditions. We test our analysis approach with measurement data from different types of cables. The results show that the complex wavelet analysis based approach is able to provide unique signatures for distinguishing between the cables, thus very promising for fault detection.