Evaluation Method Of Solid Insulation State Based On Pretest Data Of Transformer

With the rapid development of the electric power industry,the scale of the power grid is also expanding rapidly.The safety of power equipment determines whether the power grid can operate safely and reliably.As the hub equipment of power system,the operation condition and health level of oil-immersed transformer are directly related to the safety and stability of power grid.The aging of oil-immersed transformers is an important topic in the research of power equipment.The aging of oil-immersed transformers is mainly reflected in the aging of its solid insulation.In order to understand the operation status of transformers in a timely and effective manner,the regulations stipulate that a number of preventive tests should be carried out on power transformers on a regular basis.In this paper,the correlation between various characteristic quantities reflecting the solid insulation operation status of oil-immersed transformer is explored through the actual pre-test data of large volume itself.so as to realize the condition evaluation of the solid insulation life of oil-immersed transformer.Considering the above contents,this paper firstly analyzes the causes and influencing factors of transformer solid insulation aging.By analyzing the pre-test data of transformer,combining with the relevant regulations and experts' experience,20 characteristic quantities which may be related to the remaining service life of transformer are selected.In combination with a large number of transformer pre-test data,the correlation between characteristic quantity and service life data was judged by Spearman and Kendall correlation coefficient,and several characteristic quantities related to residual life were screened out.By using the collinearity judgment method based on multiple linear regression,it is concluded that there is redundancy among the characteristic quantities.Furthermore,by using the feature selection method based on mRMR.the redundant characteristic quantities are removed,and 8 characteristic quantities are obtained for the evaluation of the solid insulation life condition of the transformer.To avoid single empowerment method has one-sidedness,this article has taken the subjective method of analytic hierarchy process(AHP)and the objective method of random forest algorithm,on the above characteristics in transformer solid insulation life conditions are applied to solve the weights of assessment,and through the empowerment of the subjective and objective on the basis of the principle of minimum identification information fusion method for fusion weight;Finally,by determining the judgment result set V,the judgment factor set U,the weight coefficient set W,and the membership function to determine the judgment matrix L,the remaining life evaluation model of transformer solid insulation based on fuzzy judgment is constructed.In addition,SVM model optimized by ALO was established.Based on the fault information extracted from the transformer pre-test data,the training data set and the test data set were established,and the transformer fault diagnosis was carried out by using ALO-SVM model.Through example,found that described in this article based on the pilot test data of transformer solid insulation residual life evaluation method has certain accuracy and feasibility,can through the actual trial test data itself,exploring reflect the running status of oil-immersed transformer solid insulation between the various characteristics of relevance,assessment of transformer solid insulation life situation,combined with the transformer fault diagnosis pilot test data,play a role in the practical engineering application.