The Study Of Transformer Fault Diagnosis System Based On Case Based Reasoning

The power transformer is a important component of power system and its operatingstate have important meaning for the security, stability and economic operating of powersystem. It is high necessary to develop transformer fault diagnosis to improve reliabilityduring operating.This paper has achieved the first build of expert system that have the ability ofqualitative diagnosis based on reference of < The Preventative Test Rules Of PowerEquipment> (shorted as from now on), IEC three-ratio method, and CBR. Usedfuzzy mathematics theory to realize indeterminacy reasoning against the shortage that and IEC three-ratio method misjudge fault that near threshold usually. Extractedthe trend characteristic of DGA and then evaluated the dangerous grade of thischaracteristic against the shortage that definition of noticeable value of characteristic gascontent in oil transformer and applied to expert system.The proof of a great deal instance indicated the validity and reliability of this expertsystem towards transformer diagnosis and analysis. All the job completed in this paperhave active and indicative meaning to the fault diagnosis and condition-basedmaintenance of transformer.