Power transformer fault diagnosis based on immune evolutionary clustering algorithm

In this paper, Fuzzy C-Means (FCM) algorithm is combined with the immune evolutionary algorithm to propose a new method which can be applied to transformer fault diagnosis. First, the paper makes attributes reduction in decision table with the use of rough set, then the continuous attributes in the decision table use immune evolutionary clustering algorithm for discrete processing, and then builds transformer fault diagnosis system. Experiments show that the transformer fault diagnosis system has the feasibility and high accuracy.