Research On Diagnosis Method Of Transformer Oil Paper Insulation Aging

Transformer is one of the key equipments in power system.Transformer fault caused by aging of oil paper insulation occupies a very high proportion.Accurate diagnosis of aging degree of oil-paper insulation and reasonable evaluation of transformer insulation condition are of great significance for ensuring the safe and stable operation of power system.The diagnostic method of aging degree of transformer oil paper insulation is studied in this paper.The main work is as follows:The aging degree of oil-paper insulation of transformer is analyzed and diagnosed by using partial discharge characteristics.Based on the phase distribution map of partial discharge,27 statistical operators are extracted to form the characteristics of partial discharge.In view of the large number of original features and information redundancy,principal component analysis was used to reduce the dimension of the original features,and the original 27-dimensional features were reduced to 8-dimensional.Using Adaboost-SVM classifier to diagnose the aging degree of oil-paper insulation,experiments were carried out based on the original data set and the data set after PCA dimensionality reduction.The experimental results show that the recognition effect of Adaboost-SVM classifier is better than that of single SVM classifier,and the classification effect of data set based on PCA dimensionality reduction is better than that of original data set.In view of the limitations of single dimension feature parameters in diagnosing the aging degree of oil-paper insulation,a diagnostic model of oil-paper insulation aging based on multi-dimension feature parameters was established by using three dimension feature parameters,physical-chemical,dielectric response and partial discharge,and fusing analysis to diagnose the aging degree of oil-paper insulation.In the first part of the model,the initial diagnosis results of physical and chemical dimensions and dielectric response dimensions are determined based on the fuzzy comprehensive evaluation model.The primary diagnosis results of partial discharge dimensions are obtained by using the aging diagnosis model of oil-paper insulation based on partial discharge characteristics constructed in this paper.In the second part of the model,the preliminary diagnosis results of three dimensions are fused by evidence theory,and the final diagnosis results are obtained.An intelligent diagnosis software system for transformer oil paper insulation aging is designed and implemented.Based on the effective analysis of the functional requirements of the software system,three functional modules are designed,which are transformer oil-paper insulation aging information management module,oil-paper insulation aging degree diagnosis module and system management module.The transformer oil paper insulation aging intelligent diagnosis system is develped by using PHP programming language and MySQL database.The Intelligent diagnosis system of oil-paper insulation aging has complete functions,friendly and concise interface,and can meet various functional purposes.