| With the improvement of the distribution network structure,the single-phase grounding fault line selection and location technology of small current system is developing rapidly.In this paper,the existing methods of line selection and location are discussed in detail,the advantages and disadvantages of various methods are discussed,and two methods of fault line selection and location are put forward from different perspectives of signal decomposition and big data processing.In view of the difficulty that the existing algorithm of line selection is difficult to accurately determine the frequency band of zero-sequence current,a new MEEMD algorithm is proposed.Firstly,in order to obtain signals of attenuated oscillations at high frequencies in feeders,Hausdoff threshold is used to improve the traditional EEMD algorithm,and its energy is calculated after the Fault occurring.Finally,line selection is performed by the energy ratio of each feeder and all feeders.In this paper,the EEMD and MEEMD algorithms for zero sequence fault current are compared respectively.the experiment is carried out by using MATLAB/SIMULINK software under different fault conditions,the results indicate that modal aliasing of the traditional algorithm is completely avoided.The new algorithm will quickly and accurately select the fault line with high stability.With the extensive application of big data theory in the engineering field,this work proposes a new method of fault line selection and location based on high dimensional stochastic theory.In this method,the zero sequence current and zero sequence voltage signals are respectively formed into high-dimensional random matrices according to certain rules,at the same time the coincidence degree between the eigenvalue probability density function and M-P law curve is compared.Also,the distribution of singular value points on complex plane is analyzed,and fault line selection and fault location are carried out by using M-P law and Single Ring Theorem as criteria.This method makes up for the defect that the MEEMD energy algorithm cannot accurately determine the fault point.The simulation experiment proves that this method has a high accuracy in fault line selection and location,and the calculation is simpler. |