A case‐based reasoning approach to power transformer fault diagnosis using dissolved gas analysis data

A synthetic diagnosis approach based on case‐based reasoning is presented to diagnose incipient fault of power transformer. First, the base case bank is established according to the characteristics of dissolved gas analysis (DGA) of transformer oil. Based on different artificial intelligence techniques, three kinds of case searching method are adopted to retrieve the set of base cases, which the symptom of every retrieved case is similar to that of suspected transformer to some extent. And then a synthetic estimation method is presented to achieve the optimal base case for all retrieved base cases. Since similar symptoms may be induced by the same kind of faults, the diagnosis conclusion of suspected transformer could be drawn by the similar degree comparison between the optimal case and the suspected transformer. The application results show that the method has higher reliability and is more practical for the transformer incipient fault diagnosis.