Development of detection and diagnostic techniques for partial discharges in GIS

This paper describes detection and diagnostic techniques for partial discharges in gas insulated substations (GIS). Partial-discharge current waveforms of SF/sub 6/ gas and external noise characteristics of actual substations were investigated, and a measuring system having sensors in GIS and employing frequency range of VHF band was developed. By this system, detecting very small partial discharges of several 10 pC level has become possible. Also, a neural network system, focusing attention as a method for recognizing patterns, was applied for determining causes for partial discharge in GIS, and its applicability has been proved.<>