Discrimination of partial discharge signal in Gas insulated switchgear from external noise using internal signal processing method

Abstract

GIS (Gas insulated switching gear) is the power equipment with excellent dielectric strength and with economical value in high confidence and stability. Recently the reliability of GIS equipment has been questioned for its long usage, and the importance of the partial discharge on-line diagnosis system has been recognized. A partial discharge (PD) detection system is an effective means for the monitoring and evaluating the dielectric condition of Gas Insulated switchgear (GIS). The ultra-high-frequency (UHF) PD detection technique can detect and locate PD sources from the inside of GIS by detecting electromagnetic wave emitted from the PD source. Therefore, a real-time diagnostic system using UHF detection method was developed for this application. However, the signal of partial discharge occurring in SF6 gas is very weak and susceptible to external noises which mainly consist of PD in the air. Thus, it is very important to distinguish signal from the external noise signals. Unfortunately, these external noise signals and the partial discharge signals have very similar characteristics. For this reason, the external noise signal is not easy to remove. Therefore to solve this problem, we need the signal processing method to distinguish partial discharge signals from the external noise signals for the improvement of signal to noise ratio (SNR) and sensitivity. In this paper, we proposed the internal signal processing method to remove the external noise signals with built-in preamplifier and frequency conversion circuit.