Research On Fault Diagnose Of Power Transformer Based On Entropy Weight And Grey Relational Analysis

The safe and stable operation of Power transformer is to protect the normal supply of electricity and production and social life.We should find the internal potential fault of the power transformer timely and effectively and improve the level of maintenance of the transformer.The method based on dissolved gas analysis of power transformer fault diagnosis is commonly used nowadays.Hence,how to improve the accuracy and stability of transformer fault diagnosis,which becomes the research hotspot.In this paper,the relevant theory of transformer fault is expounded,and the relationship between transformer fault and characteristic gas is analyzed.To overcome the shortcomings of the IEC three-ratio method,it is proposed to diagnose transformer fault by gray relational analysis.Secondly,in order to solve the problem that standard sequence is difficult to be identified in the previous gray relational degree method,fuzzy clustering is proposed to determine the standard sequence of transformer fault diagnosis.The FCM clustering algorithm is used to cluster the actual collected data of fault transformer,and c optimal clustering centers are obtained.Then we can utilize the gray correlation method to diagnose the fault sequence of the transformer.Thirdly,we expound the basic gray relational model.In order to overcome it's shortcoming,a new improved relevance degree model is constructed by using entropy weight method to improve the relevance degree of the reaction system.Finally,the example of transformer fault diagnosis based on the improved relational model is presented.The fault diagnosis is based on the data of the transformer samples to be detected.The result shows that it has higher correctness and stability of diagnosis than the basic gray correlation model and the IEC three ratio method,and its validity has been proved.Utilizing fuzzy clustering to determine the standard sequence of transformer fault diagnosis,as well as to improve the basic gray relational model,all provide theoretical basis for transformer fault diagnosis technology,and it is a new exploration and attempt.It is of practical significance to improve the stability and operation economically of transformer,and to improve the reliability of power system operation,and it has certain engineering application value.