An improved DAG-SVM algorithm based on KFCM in Power Transformer Fault Diagnosis

In order to improve the accuracy of the fault diagnosis for power transformer, an improved directed acyclic graph support vector machine (DAG-SVM) method based on kernel-based fuzzy c-means (KFCM) is proposed in this paper. The separating characters among patterns are evaluated by the distances among KFCM cluster centers. It contributes to selecting the proper node and optimizing the topology structure of DAG-SVM. Based on the dissolved gas analysis, experimental results show that the proposed method can distinguish the transformer fault type accurately and effectively.