Detection and location of partial discharge and arc faults

Abstract

A system and method of automatically detecting electrical discharges in a gas insulated switchgear equipment and locating the electrical discharges in real-time is disclosed. The method includes sensing an ultra high frequency (UHF) signal generated from the electrical discharges at multiple channels simultaneously. The UHF signal sensed at multiple channels are reviewed simultaneously to determine whether they are of the same electrical discharge or not. The method also provides the sensed UHF signal at multiple channels to an artificial intelligence training model trained to determine the type of electrical discharge such as a partial discharge or an arc. The method provides a location of the electrical discharge based on an attenuation profile associated with the multiple channels and other information including amplitude information and distance information between adjacent channels.