A combined ANN and expert system tool for transformer fault diagnosis

Abstract

A combined artificial neural network and expert system tool (ANNEPS) is developed for transformer fault diagnosis using dissolved gas-in-oil analysis (DGA). ANNEPS takes advantage of the inherent positive features of each method and offers a further refinement of present techniques. The knowledge base of its expert system (EPS) is derived from IEEE and IEC DGA standards and expert experiences to include as many known diagnosis rules as possible. The topology and training data set of its artificial neural network (ANN) are carefully selected to extract known as well as unknown diagnosis correlations implicitly. The combination of the ANN and EPS outputs has an optimization mechanism to ensure high diagnosis accuracy for all general fault types. ANNEPS is database enhanced to facilitate archive management of equipment conditions, trend analysis and further revision of the diagnosis rules. Test results show that the system has better performance than ANN or EPS used individually.