Research On Denoising And Location Method Of XLPE Power Cable Partial Discharge Signal

With the vigorous development of the power industry,the use of XLPE power cables in power systems is gradually increasing.Cable operating environment is generally harsh.Most of the cables used in early construction are in the late life.Various kinds of insulation defects gradually appear in the cable,which cause insulation breakdown and bring huge economic losses.Cable insulation defects are the main cause of partial discharges(PD).Through partial discharge positioning,hidden defects of cables can be found in time.In order to ensure the stable operation of the power grid,the cables need to be replaced with faulty sections.Aiming at the PD positioning problem of XLPE cables,this paper introduces the cable preparation process and structural composition,and theoretically studies the aging characteristics and PD characteristics of the cables.Use PSCAD / EMTDC to build simulation circuit,and combine MATLAB to carry out simulation experiment and calculation analysis.Four commonly used mathematical models of PD signals are introduced.The similarities and differences of the four mathematical models of PD signals are analyzed by FFT,S transform and Teager energy operator,and the applicability of the three signal analysis methods to different mathematical models of PD.It has laid a foundation for the research on computer simulation of PD signals.A new adaptive threshold function is proposed,and the empirical wavelet transform algorithm that determines the scale space segmentation Fourier spectrum is suitable for noisy PD signals with complex spectrum,avoiding the choice of wavelet base and decomposition scale by the wavelet analysis algorithm,and avoiding experience The number of decomposition modes is manually selected when the wavelet transform is based on the extreme point to split the Fourier spectrum.The empirical wavelet transform soft threshold denoising algorithm based on scale space segmentation Fourier spectrum is superior to the traditional wavelet denoising algorithm.The empirical wavelet transform adaptive threshold denoising algorithm based on the scale space segmentation Fourier spectrum makes the denoising effect further improved.A partial discharge positioning method for cables based on the improved singleended method is proposed.This method avoids the problem of large positioning errors due to the improper wave velocity of the PD signal selected in the traditional positioning algorithm,and avoids the synchronization problem in the double-ended positioning.Based on the improved single-ended positioning,the situation of PD signal varying with the transmission distance and the uncertainty of the wave speed is considered,which further improves the accuracy of the positioning algorithm and determines the effectiveness of the method through simulation experiments.