Design and Evaluation of an Integrated Ultra-High Frequency and Optical Sensor for Partial Discharge Detection in GIS

Abstract

Partial discharge (PD) detection is an important technique for monitoring and evaluating the insulation condition of gas-insulated switchgear (GIS) equipment. The joint analysis and diagnosis of multiple signals can effectively improve the sensitivity and reliability of PD detection. In this paper, an integrated ultra-high frequency (UHF) and optical sensor is proposed and designed for PD detection. The effectiveness and sensitivity of the designed sensor are experimentally tested. Furthermore, a 500 kV GIS test platform is built, and PD measurements for different types of defects (metal particle on the insulator surface, floating potential, and protrusion) are carried out based on the integrated sensor. The results show that the integrated sensor can detect discharge signals with a minimum apparent charge below 2 pC and has good detection performance for different types of defects. Due to different propagation and attenuation characteristics, there is no strict correspondence between the amplitude of optical and UHF signals. This means that even if the amplitude of the UHF signal is close, the optical signal amplitude may still differ significantly. Compared to UHF signals, the amplitude distribution of optical signals is more dispersed, resulting in differences in the phase-resolved PD pattern characteristics between optical and UHF signals. Moreover, the effectiveness of the optical method is more easily affected by the sensor and defect position compared to the UHF method, and in some cases, the sensitivity of the optical method is lower than that of the UHF method. The results of this study provide a foundation for a reliable and sensitive PD detection technique in the GIS.