Transformer Fault Diagnosis Based on Naive Bayesian Classifier and SVR

Because traditional transformer diagnosing approaches are over-rigidity and need almost complete and accurate testing data, a NB (Naive Bayesian) classifier based model to diagnose transformer faults is presented and constructed in the paper. As the diagnosing performance is depressed by incomplete testing data, SVM regression approach is used to estimate the missing data. Thus a new diagnosis model, which integrates SVM regression and NB classifier, is constructed. The diagnosing experiments of different transformer testing scenarios show that the constructed NB diagnosis model has a good performance given complete testing data, and the proposed SVM regression approach can raise the accuracy of transformer diagnosing even if a certainty quantity of data or important data are missed