Research On Fault Identification Method And System Design Of High-Voltage Cable Based On On-line Monitoring Of Sheath Current

With the proposal of the goal of "Carbon neutralization and carbon compliance" and the maturity of high-voltage and UHV transmission technology,China has issued relevant policies to vigorously support the development of new energy technologies such as large-scale wind power generation,offshore wind power generation,West to East power transmission and related industries conducive to improving energy distribution.In order to ensure the realization of the double carbon goal,accelerate the construction of new power system and meet the development needs of urban construction,the State Grid Corporation of China has actively changed the development direction of power grid and gradually replaced the traditional overhead transmission line with high-voltage cable.Although compared with overhead lines,cables are less affected by the natural environment and have stable transmission performance,with the increase of operation years,cables are prone to frequent failures in cable insulation,cable joints,grounding boxes and other areas due to external force damage and insulation layer corrosion.At present,when the cable may have faults,the mode of "replacing and repairing" is mostly adopted,which greatly increases the operation cost;Or use the off-line detection method to conduct a series of experiments on the cables that may have hidden dangers,which virtually increases the relevant costs.The traditional manual inspection can reflect the potential faults of the cable to some extent,but this method is time-consuming and labor-consuming;Although the current partial discharge method can effectively detect cable insulation problems,this method is more complex and the fault types identified are relatively limited.Therefore,the research on a simple,accurate and comprehensive fault diagnosis method is of great significance to improve the operation performance and fine management of high-voltage cable.In view of the above problems,this paper has done the following research,Firstly,through in-depth understanding and comparative analysis of the research status and research methods of high-voltage cable faults at home and abroad,it is found that the current fault diagnosis methods have problems such as difficult detection and single identification.Then,aiming at the characteristic signal that the sheath current will change to some extent in case of cable fault,a fault analysis method based on sheath circulation is proposed.By building the simulation models of common cable faults such as body fault and multi-point grounding fault,it is found that when different types of cable faults occur,the sheath current The changes of grounding resistance and other related parameters are different,that is,the sheath current can be used as the characteristic signal of fault diagnosis and identification,which lays the foundation for subsequent chapters.Then,aiming at the problems of difficult fault identification,incomplete fault classification and low classification accuracy of 110kV cross interconnected cable transmission line.In Chapter 3,a fault identification and classification method of cross interconnected cables based on bagging heterogeneous k-nearest neighbor lifting learning is proposed.Firstly,the relevant electrical parameters such as core current,voltage and sheath current obtained during various faults are denoised and normalized,and the characteristic matrix of electrical parameters is constructed;Then,the principal components of the original data are extracted based on principal component analysis to reduce the dimension of the original data and improve the efficiency of data processing and recognition;Then,based on the k-nearest neighbor algorithm,different K values and different distance measures are used as individual learners,and a height difference sub learner is constructed.By introducing bagging algorithm to improve the overall learning efficiency of heterogeneous learners,we can effectively distinguish the faults of cross interconnected transmission cables of different types and regions.Finally,aiming at the problems of low efficiency,high cost and inaccurate fault identification of traditional inspection lines,a set of cable sheath current real-time acquisition and fault analysis system based on STM32 is developed in this paper.In addition to directly processing the data such as filtering,principal component analysis,fault identification and classification,the upper computer system can also collect and upload the data to the server for storage and build an expert diagnosis database.After laboratory test,it basically meets the needs of sheath current sampling and processing.