Intelligent transformer is important electrical power equipment in smart grid andsmart substation, intelligent level of intelligent transformer relates with intelligentsubstation operation reliability and economy. And through the research of intelligenttransformer digital on-line monitoring and information fusion diagnosis technology maytimely detect fault and shorten the fault diagnosis time, which is important for the safeoperation of the system and transformer status maintenance.On-line monitoring and fault diagnosis technology is one of the key technologies ofintelligent transformer. the power transformer commonly used on-line monitoringdevice’s principle, diagnostic methods and technical characteristics including dissolvedgas in oil, partial discharge, winding hot spot on-line monitoring are analyzed in thisarticle, providing the theoretical basis for the application of intelligent transformer. Andaccording to problems of the the transformer on-line monitoring and diagnosistechnology, raises a diagnosis method of information fusion based on D-Sevidential theory and BP neural network, which achieved comprehensive analysis andprocessing of information, the reliability and efficiency of diagnosis improved greatly. Inthis paper, multi-level information fusion technology based on D-S evidence theoryand BP neural network is successfully applied to fault diagnosis of intelligent powertransformer, also on-line monitoring and fault diagnosis system of110kV intelligenttransformer is designed.During operating,110kV intelligent transformer on-line monitoring and faultdiagnosis system find out the potential faults of transformer timely, also judge the faulttype and position accurately, proved the advance and reliability of the on-line monitoringand diagnosis method. |