Algorithm And Software Development For Portable Power Transformer Condition Monitoring Systems

Power transformers are vital components of power network infrastructure.Research on transformer fault diagnosis and on-line monitoring technology has been an important research direction of the transformer industry.Among the existing monitoring methods,the vibration analysis method is sensitive to the internal structure of the transformer and has no electrical connection with the power system,which makes it wildly used for the condition monitoring and fault diagnosis of power transformers.In recent years,some progresses have been made in the power transformer condition monitoring system,however,most of the condition monitoring system is still in experimental research stage and not suitable for field measurements.Also,the lack of effective fault diagnosis technique also greatly reduces the reliability and practicability of condition monitoring system.This paper aims to study the transformer fault diagnosis method and the implementation of portable power transformer condition monitoring system.The main work of this paper is as follows:1.In order to extract the source vibrations generated by the core and winding,a mixing model based on single-channel vibrations is presented and a new-fashioned method based on TIFROM(Time-Frequency Ratio of Mixtures)method is proposed in this paper.The method was applied to separate vibrations of a running power transformer and the core and winding vibrations were successfully retrieved.2.A transformer state classifier based on Naive Bayes is established in this paper.The model divides the transformer state into three classes:health,aging and fault.The classifier is trained on the basis of the four characteristic quantities of the transformer.The trained model can classify the status of the monitored transformer accurately.3.In order to compensate for the shortcomings of cable ICP sensors in terms of power consumption,monitoring security and measurement methods,In this paper,a vibration sensor based on micro-power MEMS sensing element and ZigBee is designed and developed.4.On the basis of the above study,this paper designs and realizes the communication software of the portable power transformer condition monitoring system.The software can meet different data communication needs under different acquisition scenarios of the monitoring system.