Fault Diagnosis of Power Transformer Insulation Based on Fuzzy Normal Partition and Logic Reasoning

A new diagnosis method based on fuzzy normal partition and logic reasoning for insulation fault of power transformer is put forward in this paper. First, according to the normal distribution functions, fuzzy processing of the insulation diagnosis parameters and diagnosis conclusions is realized. Secondly, the insulation diagnosis knowledge is acquired and the reasoning rules are built. Finally, the reasoning results are gotten by applying fuzzy reasoning in fault diagnosis. The method may resolve the problem that the relations between actual test data and diagnosis conclusions are difficult to describe and improve the reasoning efficiency. Moreover, the method increases the accuracy of fault diagnosis and maneuverability by actual computation.