The Study Of Oil-paper Insulation Condition Based On Improved Grey Target Theory

Transformers are extremely important power equipment in power system,and the insulation aging is a major cause of transformers being out of service.Therefore,to ensure the safety and reliability of power system operation,the transformer oil-paper insulation condition should be kept good.This paper mainly studies the oil-paper insulation system of electric equipment,and the equivalent circuit model is constructed on the basis of the measured data in time-domain dielectric spectroscopy,which can well reflect the actual relaxation process of oil-paper insulation system,this paper analyzes the influence law of time domain dielectric spectroscopy characteristic quantity affected by oil-paper insulation aging state,and proposes a comprehensive diagnosis method of oil-paper insulation condition based on improved grey target algorithm.The main research content of this article is as follows.1.This paper considers the relaxation characteristics of oil-paper insulation comprehensively,based on the theory of time-domain medium,the extended Debye equivalent circuit model is selected because its more fit to the circuit model of actual relaxation response process of oil-paper insulation system,and the mathematical model for parameter identification of the circuit model is deducted.Meanwhile,the artificial bee colony is improved to enhance the search ability of the algorithm,which is applied to identify the equivalent circuit of transformer insulation system to get the result of each circuit parameter identification.2.Based on the return voltage method(RVM),taking the characteristic quantity measured by RVM and the characteristic parameters of extended Debye equivalent circuit model as research object,this paper obtain the influence law of each characteristic parameter affected by oil-paper insulation aging state through theoretical analysis and case verification,and the aging criteria of each characteristic parameter diagnosing transformer oil-paper insulation state is obtained.Among them,the main measured characteristic quantities are peak value of initial slope,peak voltage value of polarization spectrum and dominant time constant.The characteristic quantities of equivalent circuit model are insulation resistance,geometric capacitance,polarized circuit number,average relaxation time constant and half-life residual energy.3.This paper proposes an improved algorithm based on analytic hierarchy process and fuzzy clustering to analyzes several groups of transformers,the weight of each character istic quantities are assigned according to its contribution degree to the diagnosis results of oil-paper insulation deterioration state,and then the transformers of similar deterioration degree are split into the same group,which lays a foundation for the further research on t ransformer insulation state.But,because of the fuzziness in fuzzy clustering algorithm,the diagnosis results are influenced by confidence level.So,the improved algorithm bas ed on gray target theory is applied to analyze the transformer oil-paper insulation state.4.Synthesizing the time domain dielectric spectroscopy characteristic quantities,applying improved grey target algorithm and diagnosing the deterioration state of insulation state comprehensively fill the gaps in the comprehensive application of each time domain dielectric spectroscopy characteristic quantity to evaluate oil-paper insulation deterioration state,and solve the problem that the deterioration state is difficult to diagnose while several characteristic quantities compete.