Power Transformer Fault Diagnosis Based on Integrated of Rough Set Theory and Evidence Theory

When using chromatography data analysis in diagnosis of power transformer fault, fault information cannot be make full use, which can't effectively discover knowledge hidden in data. In this paper a method integreted of rough set theory and evidence theory for transformer fault diagnosis is presented. In this approach, in order to avoid subjectivity of basic probability assignment", "rough set was induced to calculate the importance degree of condition attribute to decision attribute and act as basic probability assignment of recognition framework. Different evidence in the same reconginition framwork was combinated to obtain information on the fault types of decision classification information. A large number of examples analysis show that the rough set theory and evidence combination used in electric power transformer fault diagnosis, not only can effectively improve the single fault diagnosis accuracy, also give the information about compound fault analysis.