Research On Moisture Condition Detection Method Of Oil-Paper Insulation Transformer By Fusion Time Domain Dielectric Response Characteristics

Oil-immersed transformer is the key equipment of power system.Once the transformer fails,it will lead to serious blackouts in the power grid.According to the relevant research,the main reason of transformer fault is the insulation fault caused by aging and moisture of the internal insulation material.What's more,the insulation breakdown voltage will drop because of the transformer insulation moisture,which cause transformer fault easily.Therefore,it is of great significance to detect the moisture condition of oil-paper insulation accurately and timely for ensured the reliable power supply of power system.This study is funded by the National Natural Science Foundation of China that the name is "the aging diagnosis method of oil-paper insulation based on circuit analysis and recovery voltage response(No.61174117)".The research object is moisture state of oil-paper insulation transformer.The signal flow graph method is used to identify the polarization equivalent circuit parameters of oil and paper insulation based on the theory of time domain dielectric response.And the extended Debye equivalent circuit model of oil paper insulation is established by the identification result.Then,it analyse the relationship between the characteristics of time domain dielectric response and the moisture state of insulation.A comprehensive moisture condition detection method of oil-paper insulation based on entropy weight and grey correlation analysis is proposed.The main work of this research is as follows:(1)The moisture theory of transformer oil-paper insulation has been analysed.And it puts forward a method for determining the number of polarization branches based on the theory of time domain dielectric response and the change of polarization characteristics caused by oil-paper insulation moisture,that the local peak points of depolarization current quadratic differential time domain spectra is used to determine the polarization branch number of oil-paper insulation extended Debye model.Finally,the topological structure of the oil-paper insulation equivalent circuit model is obtained by the number of polarization branches.(2)The depolarization characteristics of oil paper insulation are analysed by the theory of signal flow graph,which belongs to the depolarization process of oil-paper insulation time domain dielectric response measurement.The Mason formula is taked to calculate the polarization branch transmission gain of oil-paper insulation in depolarization stage.Then,the equivalent circuit model of oil-paper insulation is constructed by the polarization equivalent circuit parameters calculated by the trans:mission gain of the polarization branch.In the end,the accuracy and credibility of parameter identification result of oil-paper insulation equivalent circuit based on signal flow graph method are verified by application examples.(3)This thesis analyses the depolarization relaxation process of oil paper insulation.Furthermore,it draws the following conclusions.The depolarization charge quantity shows the polarization strength and the ability of storage charge for oil-paper insulation.The number of polarization branches reflects the number of relaxation response mechanisms.The peak voltage of the polarization spectrum reports the polarization characteristics of the relaxation response of the insulation dielectric.The main time constant reflects the polarization response rate of the insulation dielectric.The initial slope peak represents the polarization conductivity of oil-paper insulation.The relationship between the time domain dielectric response characteristics and the moisture condition of the insulation is explored,and the criterion for detecting the moisture state is proposed by using the relationship.(4)The comprehensive evaluation index model of oil-paper insulation is established by fusing multiple time domain dielectric response characteristics.The weight of each evaluation index is reasonably distributed by entropy weight method.Then,the grey correlation analysis method is used to calculate the correlation degree between the pending diagnosis sequence and the standard reference sequence.Finally,the moisture state of oil-paper insulation transformer is determined according to the maximum correlation degree,which provides a new method for the moisture condition detection of oil-paper insulation.