Research And Design Of On Line Monitoring System For Transformer Operation State In Substation

In the whole power grid,substation is the basis of connecting the whole power grid,and plays an important role in power transformation and distribution.However,due to various historical reasons,the overall supporting facilities of many substations in our country are old,and the influence of human and environment makes the safety and reliability of substations greatly threatened.In order to improve the reliability of power grid operation,it is necessary to carry out fault diagnosis and on-line monitoring for substation,so as to give early warning before the accident and reduce the accident rate.However,the traditional fault diagnosis method directly predicts and diagnoses the original data of the transformer,which is difficult to effectively mine the corresponding characteristic quantity of different fault types of the transformer.It will not only increase the calculation amount of the system,but also have low judgment accuracy,and can not give full play to the early warning function in time.The traditional transformer on-line monitoring and fault diagnosis system has the disadvantages of low monitoring accuracy and poor fault diagnosis ability,so it is necessary to put forward accurate on-line monitoring and fault diagnosis algorithm to achieve accurate on-line monitoring and diagnosis of transformer.In this paper,based on wavelet decomposition and Gru neural network algorithm,the algorithm of transformer on-line monitoring and fault diagnosis is studied.On the basis of algorithm research,the design and system development of transformer on-line monitoring and fault diagnosis method are carried out based on embedded technology,Through the embedded technology,the data of transformer box is collected and transmitted,and the collected information is uploaded to the online monitoring and fault diagnosis software system;Through remote monitoring to achieve online monitoring and fault diagnosis of transformer,the efficiency of online monitoring management and online fault diagnosis of transformer is greatly improved.The main contents of this paper are as follows:(1)The principle of gas produced by transformer box is analyzed,and the relationship between different gases in transformer oil and fault types is analyzed;This paper summarizes the transformer condition analysis method and the process of on-line monitoring and abnormal diagnosis,as well as the demand analysis of transformer remote monitoring and abnormal diagnosis system.At the same time,based on wavelet decomposition and Gru neural network algorithm,the transformer fault diagnosis method is studied.The feature quantity of transformer fault feature data is extracted based on wavelet analysis to provide the extraction accuracy of transformer fault feature quantity;Then,after completing the wavelet analysis,the extracted fault data features are input into Gru neural network for training and testing,and the corresponding fault diagnosis results are obtained.Taking the relevant calculation example as an example,it is verified that the transformer fault diagnosis can be realized based on wavelet decomposition and Gru neural network algorithm.(2)In the third chapter,based on wavelet decomposition and Gru neural network algorithm,the design and development of transformer on-line monitoring and fault diagnosis system is carried out based on embedded technology.Determine the overall design scheme and specific design content of transformer online monitoring and fault diagnosis,and complete the design of software and hardware of the system and the development of the overall interface;Finally,the online monitoring and fault diagnosis system of transformer is tested based on the original data.The test results show that the developed online monitoring and fault diagnosis system can effectively complete a variety of fault diagnosis of transformer.The transformer on-line monitoring and fault diagnosis system developed and designed in this paper can provide more accurate diagnosis means for transformer on-line monitoring and remote diagnosis,improve the management level of transformer on-line monitoring and the efficiency of fault diagnosis,and has important practical significance for improving the safe and reliable operation of transformer.