The Multi-class SVM Is Applied in Transformer Fault Diagnosis

Abstract

Transformer fault forecast plays an important role in the safe and stable operation of power system. So it is important to detect the incipient faults of transformer as early as possible. In this study, the support vector machine (SVM) is introduced to analyze and diagnosis the transformer fault. According to the accumulation fault data, the SVM forecast model take the RBF as the kernel function and utilize the best pattern to cope with data for reducing imbalance. In order to prove the SVM method efficacious and accuracy, we also make the diagnosis with traditional three ratio method experimental. The results of the final experimental indicate that SVM can make higher diagnosis accuracy and have excellently generalization ability.