A nonlinear insulating dielectric is an insulation material which exhibits variations in conductivity and/or dielectric constant in response to an electric field,thereby enabling adaptive homogenization of the electric field.It finds wide applications in motor end insulation,anti-corona structures,and cable accessory insulation,owing to its exceptional properties.To obtain accurate and reliable dielectric parameters,this dissertation focuses on the response data of the time domain spectrum of the polarization current and the joint current-potential measurement of nonlinear insulating dielectric under step voltage excitation.Nonlinear regression analysis serves as the primary tool to investigate the acquisition method and result evaluation of the nonlinear insulating dielectric’s dielectric parameters,including conductivity,relaxation polarization rate,displacement polarization rate,among others.Firstly,this thesis establishes a nonlinear insulated dielectric three-branch equivalent circuit model on the circuit simulation software Multisim,and appropriate parameter conditions are set.Using this model,the polarization process of the dielectric is simulated,and relevant data are obtained according to the time-domain spectrum of the polarization current at different voltage levels.The extended Debye model is then selected to fit the polarization current time-domain spectra,and the timedomain response information of the absorption and conduction currents are separated from it.The relaxation polarizability and conductivity of the dielectric model are extracted from these two parts of currents by calculation.Subsequently,the proposed calculation method was validated using statistical means such as relative error,coefficient of determination,and confidence interval of the parameters.These statistical means were used to confirm the accuracy and reliability of the method.In order to verify the robustness of the algorithm,the computation process is investigated under the control of polarization time,sampling rate,and noise level.The results show that the algorithm has high robustness and good reliability.Considering the limitations of the test system in acquiring information related to instantaneous displacement polarization in the polarization current time-domain spectrum test,this thesis performs simulations of the combined current and potential test based on the above model,and obtains the corresponding simulation data.Using these simulated data,the displacement polarization rates of nonlinear insulating dielectrics are calculated and analyzed for accuracy,reliability,and robustness of the algorithm.Finally,using the real response data obtained,this thesis performs the above-mentioned dielectric parameter acquisition operations,and analyzes the calculation results using statistical methods such as point estimation and interval estimation.The results demonstrate that the various algorithms used in this thesis provide excellent results,verifying their performance in terms of accuracy,reliability,and robustness,and laying the foundation for more in-depth research. |