| Partial discharge(PD)detection technology can be used to not only accurately recognize cable latent fault,but also locate the cable defects,thereby improving the reliability of highvoltage(HV)cable power supply.The PD signals selected in the industrial site is mixed with various interference signals,and even the PD signals are submerged in the interference signal.These makes the detection and judgment of the PD signals difficult.Moreover,due to the unique single-core multi-layer structure of the HV cables and many kinds of shielded grounding models in HV cables,the PD signals can be attenuated and dispersed when it is transmitted along the cable.These also makes the PD recognition and location more difficult.Therefore,the research based on the PD detection technology in HV cables is studied in the paper.Firstly,the propagation characteristics of PD signals are analysised in the paper.Then the marginal effects and recognition of PD signals are studied in the paper.Finally,a method is proposed for high-precision location of PD signals.The main results are as follows:(1)The propagation law of PD signals in the HV cables based on the frequency-dependent characteristics of semi-conductive layers is studied in the paper.In the frequency-dependent parameter model,the attenuation characteristics of PD signals propagating along the cable conductor layer and the metal shield layer are analysed,and the coupling effect of PD signals in cable joint is taken into account.Then,the dynamic range of characteristic parameters of PD signals under different cable length is studied.Moreover,the characteristic parameters of the PD signals are statistical analysised based on a least square algorithm.Subsequently,the attenuation mathematical functions of characteristic parameters are established.(2)The marginal effect of the PD signals characteristic parameters is studied in the paper.A series data set of power frequency period time containing PD signals is obtained by the platform of PD detection in HV cables,which composed of three artificial defects.The wavelet decomposition and reconstruction method is used to remove the interference of periodic narrow-band pulses in the data set.The peak value,rise time of the PD signals and the time interval between adjacent PD signals are selected as the characteristic parameters.The(35)t atlas is plotted and the distribution law of the peak value and rise time of the PD signals is obtained.The upper and lower bounds of the original characteristic parameters based on the marginal effect is established.(3)A partial discharge bandpass blocking method based on marginal effect is proposed.The density clustering algorithm is used to analysis on the characteristic parameters of the original PD signals.The upper and lower bounds of the characteristic parameters at different distances can be obtained by the attenuation mathematical function of the characteristic parameters.An example analysis verifies that the bandpass blocking method can effectively identify the same type of PD signals with propagating information and remove other types of PD signals.(4)The high-precision PD location technology based on multi-point synchronous detection system is studied in the paper.Through the multi-point synchronous distributed PD detection technology,a high-precision PD signals location technology based on the phase difference algorithm of the energy method is proposed.The problems of the uncertainty of the PD signal wave velocity and low location accuracy can be solve by the proposed method.The simulations and experiments have been conducted and the results show that the position of the cable defect can be effectively located by this method with high accuracy.Especially,in a low signal-to-noise ratio environment,this method can still locate the PD signal. |