Research on Transformer Fault Diagnosis Expert System Based on DGA Database

This paper analyzes and designs the transformer fault diagnosis system based on dissolved gas analysis (DGA) database in which DGA data is managed by the Oracle database. The fault diagnosis module includes the single analyzing item and the Integrated analyzing item, such as, improvement three-ratio method, grey relational entropy, fuzzy clustering, artificial neural networks, and so on. They reduce the insufficiency in diagnosis method which is used now. The system realizes each function of the modules by using the lamination method, it is able to diagnose problems existing in oil chromatogram analysis data of transformer, and the accuracy of the system is also testified by practical example.