Study On Partial Discharge Propagation Characteristics And Positioning Of 10kv XLPE Power Cable

Insulated power cables are becoming more and more popular in today's power distribution networks.However,it may occur insulation defects because of aging,deterioration,and various operational and environmental stresses,so it is necessary to monitor the cable in time to avoid unexpected failure.Defects are directly related to partial discharge(PD)phenomena.The positioning of these PDs is a very important aspect of power cable condition monitoring.The research aiming at the partial discharge signal propagation characteristics,filtering noise and partial discharge positioning of 10 kV XLPE cables partial discharge of cables was carried out.Firstly,the simulation is used to study the propagation behavior of PD signals in the branch cable configuration.Secondly,a PD signal denoising hybrid algorithm consisting of dual-tree complex wavelet transform and adaptive singular value decomposition is proposed.Finally,use New reflection measurement technology RBR-based for partial discharge location and evaluation of cables.The simulation circuit was built with PSCAD/EMTDC,and combined with MATLAB for simulation test and calculation analysis.The core contents of this article are:(1)Introduce a method of obtaining a partial discharge signal of a cable.First,a high frequency current transformer that does not damage the cable structure is mounted on the ground line to detect the PD signal.Next,the study of partial discharge of the cable is carried out,first analyze the electric field distribution of the cable,and then analyze the electric field of the defect in theory.(2)Use simulation to study the propagation behavior of PD signals in the branch connection cable.It was found that the characteristic impedance plays an important role and determines the dispersion of PD energy at the cable joint.A detailed analysis of the measured PD signal not only provides a comprehensive explanation of the PD current pulse dispersion at the joint,but also identifies the cable insulation defect portion of the branch cable connected at the joint.(3)A PD signal denoising hybrid algorithm consisting of dual-tree complex wavelet transform and adaptive singular value decomposition is proposed,which is called ADTCWT.In this algorithm,the signal has been decomposed and reconstructed based on a two-tree complex wavelet transform.The threshold based on the dual-tree complex wavelet transform is determined by an adaptive singular value decomposition method.The purpose of the proposed algorithm is to remove noise from the detected signal.(4)A new RBR-based reflection measurement technology is proposed to perform partial discharge localization and evaluation of cables.In the RBR,the wave velocity necessary for the conventional reflection method is not used.Based on the measurement history,a training set is constructed and three feature quantities of the training set are extracted.After performing feature extraction,the GRNN is used to estimate the partial discharge position and reflection coefficient based on the RBR.