A novel TDOA estimation method for UHF sensor based PD localization system in HV power apparatus

Abstract

Detection and localization of Partial Discharge (PD) accurately has become an increasingly important topic for researchers in High Voltage (HV) system. PD measurement helps in condition monitoring of the total power system utilities with minimal expenditure, as it acts as an evidence to fault detection prior to outage. For complex power distribution systems, it is also very essential to locate PD coordinate exactly apart from its measurement, to identify the particular point of disorder. For accurate PD source detection and its localization, UHF signal based non-contact detecting system is proposed in this paper. In this presented work, Micro-strip Patch antenna is used as UHF sensor for detection as well as localization of PDs in HV power apparatus. A pair of sensors is being employed for detection of the same UHF PD signal, placed at different locations. These captured signal pair are then used for estimation of Time Delay of Arrival (TDOA) between two sensors by the proposed method, i.e. calculating Joint of Entropy (JE) between the signals. TDOA is further transformed into distance and used for estimation of the PD location by solving non linear equations. The K-means clustering technique is used for identifying the multiple PD sources. The experimental result proves the authenticity and efficacy of the proposed method for detection and localization of PD sources in HV power apparatus.