Development of an expert system for analyzing faults in power system using artificial neural network based waveform recognition

The authors describe a hybrid fault analysis expert system using faulted section judgment and waveform recognition. Faulted section judgment is executed by an expert system (ES) using information about the operated relays and circuit breakers (CBs). Waveform recognition is performed by an artificial neural network (ANN) using the fault voltage/current waveform. The result of waveform recognition is verified by using qualitative knowledge. With ES and ANN used together, the system can obtain the convenient features of both methods