| Due to its power supply reliability,the small current grounding system has been widely used in China.Moreover,the incidence rate of line fault caused by single-phase grounding account for 80%of small-current grounding faults.However,the existing distribution network there was a ground fault circuit,fault line selection device missed to choose,leakage,etc.,lead to line selection accuracy cannot meet the requirement of the power supply reliability.Therefore,it is of great significance to study how to quickly and accurately identify the fault line in order to reduce the influence of fault on power supply and give full play to the advantages of small current grounding system.To this end,this article starts research on the fault line selection of small current grounding systems,focusing on solving the problems of extracting fault features,determining faulty lines,and improving the accuracy of line selection.The main work of this paper is as follows:Combined with the actual line single-phase grounding fault characteristics,The phase-to-ground fault of the small current grounding system is modeled,and the development process and fault characteristics of the phase-to-ground fault are theoretically derived according to the model.At the same time,the superiority of transient line selection algorithm is proved by comprehensive comparison of fault signal amplitude and fault feature richness.The simulation model of small current grounding system fault is built in MATL AB/Simulink,and various single-phase grounding faults that may occur in the actual distribution network are simulated by the simulation model.The simulation data are obtained to further verify the correctness of the theoretical analysis and provide data support for the selected line model in the following paper.After the zero-sequence current of each outgoing line before and after the fault was obtained,three kinds of IMFs containing different components were obtained by variational modal decomposition,and eigenvectors were formed by calculating the modal energy measure of each line.Input the SVM classifier to classify the fault sample data.Aiming at the SVM parameter optimization problem,the whale swarm optimization algorithm was introduced to get the optimal parameter value.Finally,the test samples were input into the trained ELM,KELM,SVM and WOA-SVM models respectively to get the test results.This paper illustrates the advantages of the proposed line selection model based on VMD feature extraction and WOA-SVM classification. |
