Research On Small Current Grounding Method Based On Improved MP Algorithm

With the development of the power system,more and more cable lines are used in the distribution network,which directly results in a sudden increase in the capacitive current of the system,which will bring serious security risks to the equipment of relay protection and the insulation equipment,especially to the ungrounded operation of neutral points.According to the DL/T620-1997 electricity industry standard promulgated in 1997,when the system capacitance current exceeds 10 A,the neutral point needs to be grounded by arc suppression coil.The residual current at the fault point of the arc suppression coil grounding system should not be too large,and the arc suppression coil should be overcompensated.The advantage of this system is that it can run for 1-2 hours after the failure,improve the insulation level of the line and the reliability of power supply,reduce the interference to the communication system,and strengthen the protective effect of the electrical equipment.However,after the single-phase grounding fault occurs,how to select the faulty line quickly and accurately brings great challenges to the power workers.Since the steady-state characteristic quantity is not obvious after the fault occurs,the fault location method based on transient characteristic quantity is more suitable for the resonant grounding system.Although there are many theories of fault line selection,the actual application results are still not ideal and the rate of false positives is very high.Therefore,a new fault line selection method based on Improved MP Algorithm is proposed in this paper.Firstly,this paper analyses the transient characteristics of the single-phase grounding fault in the neutral grounding system,and deduces the transient capacitance current,the transient inductance current and the transient grounding current.And through the SIMULINK modeling and simulation observation and analysis of the fault characteristics,and then by setting different fault conditions,summarize the factors that affect the accuracy and reliability of the line selection,and provide the theoretical basis for the later research.Secondly,combining genetic algorithm with matching pursuit algorithm,the improved orthogonal matching pursuit algorithm(GOMP)is used to decompose the zero sequence current and get the each order atomic components of the fault line and the non-fault line.This method not only solves the problem of high computation and low efficiency in the greedy iteration,but also has the smallest residual energy compared with the improved MP and OMP,that is,it can match the original signal more accurately,so as to get the sparse expression of the original signal.On this basis,the atomic energy operator AEO(Atomic Energy Operator)is introduced.Due to the conservation of energy before and after decomposition,the relative energy value of each line from the fault time to the end of the simulation is compared by obtaining the integral of each line AEO in the time domain,and the fault line selection can be realized more accurately.Finally,modeling and simulation the typical of resonant grounding system--underground low-voltage distribution network,verify the feasibility of the method by setting different fault conditions.Then,Chaoyang primary change is selected as simulation object,and combined with actual situation,modeling and simulation are carried out.The simulation results show that the proposed method is not affected by the conditions of the fault time,the transition resistance value and the fault location,and it has strong robustness and feasibility.