Simultaneous classification and localization of partial discharges in gas insulated switchgear using recurrent neural network

Partial discharge (PD) activity is a pre-cursor for insulation degradation which may eventually lead to catastrophic failure of the electrical equipment with severe social and economic consequences. It is therefore imperative that PD is detected at its early stages to permit repair or replacement, prior to total failure. In this work, PD measurements from a test cell inside the L-section of a gas insulated switchgear (GIS) are used to train and evaluate a bidirectional long short-term memory (BiLSTM) recurrent neural network (RNN) for classification and localization of PDs. Evaluating the trained model yielded an accuracy of around 96% with a spatial resolution of 15 cm for simultaneous classification and localization.