Method Of Fault Section Location Based On Distributed Current Collection Terminal

Single-phase ground fault is the most common fault in the distribution network,the probability of which is nearly 90%.The characteristic of the current and voltage is not prominent in the single-phase earth fault of the distribution network,and the measurement equipment in the medium voltage distribution is insufficient,which makes the fault location problem of the distribution network not solved well for a long time.Distributed phasor acquisition terminal is low in cost,which can help to achieve fault location of single-phase earth fault in distribution network.Thus,the fault location technology of distribution network based on distributed current phasor acquisition device is studied in this paper in order to realize the fast and reliable fault location of distribution network.Firstly,the unbalanced current caused by load and its effect on the method of fault location is analyzed in this paper.At the same time,the application of zero-sequence current fault component method in resonant grounding system is researched.Then,the characteristic of three-phase current and phase-to-phase current in single-phase earth fault of distribution network is analyzed,and a fault location algorithm based on distributed phase current fault component and phase-to-phase current fault component is proposed.At the same time,the characteristic of the phase current of the line when the load changes are analyzed,and the line fault and the change of phase current caused by the load change are distinguished to eliminate the interference of the load change to the fault zone.Finally,a new fault location method based on improved binary differential evolution algorithm is proposed to solve the problem of data distortion and missing fault location when the line information is uploaded to the control center.The control parameters of algorithm are improved adaptively.The simulation results show that the algorithm can locate the fault accurately in the case of single point fault and small amount of data distortion and the convergence is improved compared with the algorithm of binary differential evolution.