Research And Application Of The Method For Fault Detection Of High Voltage Power Cables

With the rapid economic development,industry and agriculture business of electric power demand is increasing year by year,lose distribution line length increased by land resources tense GIS equipment increasing,the influence of such factors as proportion of cable line more and more cable line adopts directly buried pipe coating methods such as in the ground,laying cable ontology cannot be observed directly,part of the cable line drawings missing path is not clear,path has such problems as other building facilities,fixed point to be more difficult than traditional overhead lines,the fault diagnosis need for longer periods of time.Efficient high voltage cable fault location method can reduce maintenance time reduce the detection cost,help to improve the grid power supply stability,with good economic and social benefits,therefore,how to improve the cable line especially high voltage cable line fault location technology to raise the level of fault detection,the current is an urgent need to solve the problem.In this paper,a BP neural network fault location method based on particle swarm optimization(PSO)is proposed to solve the problem that it is difficult to accurately locate a cable fault.Firstly,the law of electromagnetic wave propagation in the cable is analyzed,a cable distribution parameter model is constructed,and based on this model,the relationship between the voltage at the fault point and the voltage and current at the beginning and end of the cable is analyzed.BP neural network is introduced to identify cable parameters,which solves the problems of cable parameters change and fault location inaccuracy caused by distributed inductance and capacitance in cable operation.In view of the problem that neural network algorithm is highly dependent on initial value selection and slow convergence,particle swarm optimization is proposed to improve the algorithm.Secondly,the detailed analysis of cable fault location system based on functional requirements,based on the.NET Framework design of cable fault location system architecture,design the system hardware,system software and based on the C# language developed functions the system realize the function of neural network training and cable fault location,and capable of storing data cable fault.Finally,some common cable fault examples in Harbin area are selected for fault location diagnosis,and the application of high voltage cable fault location method based on PSO-BP neural network is realized.The results of ranging test and actual measurement show that the method of fault location based on neural network for cable parameter identification meets the requirements of calculation error,and is effective,feasible and worth popularizing.