Study Of Fault Diagnosis Method For Power Transformer Based On Cloud Theory

Power transformers,as power equipment for transmitting AC power in the power grid,play an important role in the power system and play an important role in the safe,stable and reliable operation of the power grid.Therefore,the fault diagnosis research of power transformers has important significance.In practical applications,the three-ratio method is mainly used to diagnose the fault of the transformer.The traditional three-ratio method has the disadvantages of misjudgment of boundary position,lack of coding,and failure to diagnose multiple faults at the same time,resulting in low accuracy of the judgment result.However,most of the transformer state predictions have the disadvantages of large error when the abnormal fluctuation is large,it can not effectively solve the non-equal interval time series,and it have large fluctuations in particle size at different time.In view of the above problems,this thesis conducts in-depth research on power transformer fault diagnosis and prediction of dissolved gas concentration in oil based on cloud theory,mainly doing the following work:(1)The paper applys cloud theory to improve the fault diagnosis of transformer three-ratio method.Firstly,the initial cloud model is established,and the data rule is used to establish the cloud rule.Then,the sample data to be tested is input to generate the membership cloud model,and the transformer failure type is obtained by optimal matching between the member cloud model and the initial cloud model.According to the randomness and ambiguity of cloud theory,the model makes the three-ratio method no longer limited to defining the fault type with fixed coding,thus it solves the shortcomings of boundary misjudgment and coding loss in the traditional three-ratio method,and it also reserved the three superior ratios are used as the sample data.The accuracy is improved.(2)The paper studys the prediction of dissolved gas concentration in power transformer oil based on cloud theory,it based on the concentration of dissolved gases in five kinds of oils.it uses cloud theory method to analyze and predict the concentration of five gases to solve the problem of large error in abnormal fluctuation of data and non-equal interval time series data prediction performance fluctuation.Simultaneously it introduced a interval parameter estimation theory,and the interval prediction model of dissolved gas concentration in oil is established.The cloud point prediction based on cloud theory is combined with interval prediction.At the same time,it is compared with the generalized regression neural network model(GRNN),LSSVM model and gray model(GM)to verify the validity of the model.(3)The paper designes a power transformer fault diagnosis system based on MATLAB GUI(Graphical User Interface).It includes a transformer fault diagnosis by traditional three-ratio method,a transformer three-ratio fault diagnosis using cloud theory improvement,and a dissolved gas in transformer oil based on cloud theory Interval prediction.