Research On Partial Discharge And Insulation Monitoring Of Mine Cable Based On Step Pulse

The reliability of the mine power grid is a basic condition for ensuring underground safety production.The insulation performance of mining cables is an important basis for ensuring the safety of mine power grids.Mining cables can cause a variety of hidden dangers during the insulation degradation process,such as cable insulation damage,single-phase ground faults,and even cable firing caused by partial discharge in the early stage of cable insulation degradation.Therefore,the monitoring and diagnosis of the insulation status of mining high-voltage cables is very important.During the operation of mining cables,it is necessary to regularly perform preventive failure detection and real-time online monitoring to determine the insulation status of the cables.In order to accurately judge the insulation status of mining cables and propose effective detection methods,this paper has conducted in-depth research from two aspects.On the one hand,the partial discharge of the cable at the initial stage of insulation degradation were detected.On the other hand,we conduct on-line monitoring and line selection of ground faults after further damage to the cable insulation.In the field of partial discharge detection of high-voltage cables for mining,due to the special environment of the coal mine,the ordinary cable partial discharge detection method on the ground cannot be directly used.Inspired by methods for detecting partial discharges in mining motors,high-voltage frequency converters for mining,and power electronic device modules.Our study proposes a new detection method for partial discharge of mining cables.High-voltage step pulse voltage is injected into the core wires of each phase of the mining high-voltage cable to cause partial discharge.By collecting and analyzing the partial discharge signal generated by the high-voltage step pulse excitation,the partial discharge of the cable is detected.In order to fully explore the feasibility of this method,we first conducted a systematic study on the partial discharge characteristics of mining cables under step pulses.Firstly,the electric field simulation model of mining cable is established,the causes of partial discharge of cable due to defects are analyzed,and the existing simulation model of partial discharge of cable is improved.According to the characteristics of cable insulation defects,a cable partial discharge transient process model based on the bubble expansion model was innovatively proposed,and the partial discharge mechanism of the cable under step pulse was clarified.Secondly,a cable partial discharge experimental platform based on high-voltage step pulse injection is established,and a transient process model of the cable partial discharge is verified;At the same time,the partial discharge signals of mining cables under step pulses and AC power frequency sinusoidal voltage were compared through experiments,and the advantages of using step pulses to detect partial discharges were analyzed.Through the partial discharge experiment under a single pulse,the spectral characteristics of the partial discharge signal of the cable are analyzed.Based on the research on the characteristics of partial discharge under step pulse,a new type of partial discharge detection technology and device for mining cable based on step pulse injection method was further proposed.First of all,in terms of hardware,this research solves three key issues of pulse injection,signal acquisition,and detection during the design of system devices.A prototype was further made to conduct a cable partial discharge detection experiment,and the data sample was successfully collected.Secondly,in terms of partial discharge signal processing,in order to solve the problems of denoising,filtering to remove interference and identifying and detecting.Denoising methods of partial discharge signals based on Hankel matrix fast singular value decomposition,partial discharge signal optimization methods based on improved least squares fitting filtering,and cable partial discharge signal discrimination methods based on multi-index combined neural network algorithms are proposed.The intelligent identification and detection of the partial discharge of the cable is realized.In addition,using prototype experiments to further study the transmission characteristics of partial discharge signals in cables,on the basis of this,two types of cable partial discharge positioning methods based on the attenuation characteristic positioning method and time difference method are proposed.Finally,the neural network algorithm was used to optimize the fusion of the two positioning methods to achieve high-precision positioning of the cable partial discharge.The partial discharge detection and location technology method based on step pulse injection for mining cables proposed in this paper can detect partial discharges in the early stages of the degradation of mining cable insulation,and achieves the purpose of early warning of faults at the beginning of cable insulation decline.In terms of cable insulation fault diagnosis and line selection,the existing single-phase ground fault line selection technology for high-voltage power supply systems in mines is affected by distributed capacitance and non-effective grounding methods,resulting in inaccuracy of insulation monitoring.Aiming at this problem,a new type of mining cable insulation on-line diagnosis and rapid line selection technology was proposed.In this study,a 6k V power supply model of the mine was first established based on the actual 6k V power supply structure topology of the coal mine,which was used to monitor and diagnose the insulation status of the cable and analyze the line selection.Then,the various insulation state characteristics of the 6k V cable in the mine are modeled,and the decision logic for the cable insulation state diagnosis and fault judgment is established according to the insulation state characteristics.Based on the machine learning logic algorithm,a method for online insulation diagnosis and fault determination of mine cables is proposed to realize the functions of real-time online monitoring diagnosis and fault determination of cable insulation.Subsequently,a fast fault line selection technique for cable insulation based on the low-voltage step pulse injection method was proposed.The technical principle is analyzed,and the flow selection process and implementation method are given.Finally,simulation and prototype experiments verify the practical reliability and accuracy of the proposed technology in coal mine 6k V power grids.The proposed online fault diagnosis and line selection technology for mining cable insulation based on step pulse can effectively avoid the influence of distributed capacitance on line selection accuracy.The faulty line can be selected when the cable insulation drops to100k? to prevent serious ground faults.Through the combination of the two technologies,the purpose of real-time insulation online monitoring and diagnosis and line selection of the underground power grid cables is achieved.Finally,the article summarizes the entire research,obtained results,conclusions and innovations,and looks forward to the focus and direction of the next research.