Application of Evidence Theory in Transformer Fault Diagnosis Based on Distance Measure

When traditional evidence theory is used to analyze the fault of transformer, the reliability of the selected bodies of evidence is unscientific and subjective. In this paper, the relation between each bodies of evidence is fully considered and the reliability is ascertained through the distance measure of evidence, through which the data of Dissolved Gas Analysis and electrical tests data are combined effectively and the operation state of transformer is reflected exactly and all sidedly. Firstly, a similarity matrix is established by the proposed algorithm and the credibility of evidence is achieved through the distance of evidence. Secondly, evidence is preprocessed through introducing credibility and the basic probability assignment is constructed based on distance measure. Finally, preprocessed evidence is integrated effectively using Dempster's rule. Through the proposed approach, the multi-characteristic signal is utilized adequately and the diagnostic accuracy is improved, and the practical engineering problems are disposed effectively with enhancing the ability of distinguish the uncertainty data.