A METHOD OF SYNTHESIS BASED ON THE GREY CLUSTER AND FUZZY CLUSTER ABOUT INTERNAL FAULT DIAGNOSIS OF TRANSFORMER

According to the fuzzy and gray features corresponding to the insulation condition of transformer, several typical fault samples are divided into c kinds of gray types and the same number of optimized cluster centers are computed in use of fuzzy cluster method. And then the principle and algorithm to found whitenization weight function of gray type of faults are brought forward on the base of cluster center matrix and gray system theory. By the algorithm, coefficients of gray cluster of all patterns for examination are computed firstly and then a new model of gray cluster integrated with fuzzy cluster is proposed for fault diagnosis of transformer. A large quantities of instant analysis using the model is made and the results show the higher accuracy of diagnosis is achieved by the method proposed in this paper than that achieved by present conventional methods.