A High-Accuracy Spatial Localization Methodology for Partial Discharge by UHF-Signal Sensor-Arrays

Abstract

Inspired by the idea of mobile targets localization by the antenna array, a novel localization methodology for the partial discharge (PD) source in power systems is proposed by using ultra-high frequency (UHF) sensor-arrays in this paper. Due to the broadband feature of UHF-signals emitted from PD source, a focusing algorithm, i.e., two-sided correlation transformation (TCT) algorithm, is utilized to transform the PD signals to the narrowband ones, which improves the accuracy of the direction-of-arrival (DoA) estimation at the sensor-array under the severe electromagnetic surroundings. Then, with the DoAs obtained from multiple sensor-arrays, a spatial localization paradigm for the PD source is introduced. Finally, the PD source can be localized within a spatial sphere formed by the multiple skew DoA-lines from these sensor-arrays. Based on the simulation with the experimental PD UHF-signal data, it is shown that the proposed methodology accurately localizes the PD source within the a small range of sphere under the severe testing situations, which improves the accuracy of PD source localization compared with the classic time-difference-of-arrival (TDoA) method.