Research On Condition Evaluation And Fault Diagnosis Of Power Transformer

As one of the basic equipment of power system,the operation state of power transformer is directly related to the long-term stable and healthy operation of power grid.In recent years,with the rapid development of the intelligent construction of China's power grid,the number of power equipment has increased rapidly,and ensuring the stable and healthy operation of power grid is facing greater challenges.Therefore,the in-depth study of power transformer status evaluation and fault diagnosis technology is of great theoretical significance and practical value to ensure the normal and reliable power supply of power system and promote the development and application of condition maintenance.In this paper,the 220 k V oil-immersed power transformer is taken as the research object,and the condition evaluation and fault analysis of the power transformer are studied as follows:First,referring to the standards and relevant regulations of the State Grid Corporation of China,considering the structure and performance characteristics of power transformers,based on the current status of power enterprise technology development,optimizing the factors that affect the operation status of power transformers,proposing the classification criteria and maintenance response strategies of power transformer status,establish effective Multi-level state evaluation model of power transformer based on part performance.Secondly,on the basis of understanding the commonly used index weight determination methods,the method of determining the index weight of power transformer combining fuzzy set value statistics method and entropy value method is adopted by comprehensively considering the advantages and disadvantages of subjective and objective weight methods.This paper analyzes the relationship between power transformer status and extension theory,explores the construction form of the correlation function of the state evaluation index of power transformer,constructs power transformer status evaluation steps based on layered extension method,and achieves accurate evaluation of power transformer status.Thirdly,in consideration of individual differences in actual operation of power transformers,based on the research results of gas to gas ratio in oil at home and abroad,based on common characteristic gases and IEC TC 10 fault database,discrete binary particle swarm optimization and support vector machine algorithm are adopted to optimize the input features of fault model.The test results show that the preferred feature has higher fault diagnosis accuracy than the common characteristic gas and three ratio feature.Finally,combined with the power transformer state evaluation model and fault diagnosis algorithm proposed in this paper,based on Python,My SQL and other development environments,a prototype system for power transformer state evaluation and fault diagnosis is designed and implemented,which can realize the functions of power transformer state evaluation and fault diagnosis analysis,provide visualization of the maintenance decision support for equipment users.