Faculty of Electric Power Engineering, Kunming University of Science and Technology, Kunming 650051, China
This work is supported by National Natural Science Foundation of China (No. 51667010， No. 51807084).
According to the problems of complex topology of the radial distribution network and difficult identify of faulty branch. Based on the definition of the topological backbone, the principal component analysis (PCA)-support vector machine(SVM) model is constructed by PCA and SVM, which is used to extract the curve cluster characteristics of the fault current traveling wave, and the fault section is divided and defined as a cut set. Then, the ranging function is constructed by the relationship between the forward and reverse voltage traveling waves, the distribution characteristics of traveling wave along line are extracted, the invalid mutation points are filtered according to the relationship between the amplitude, polarity, time window and line length of the mutation point, and the effective mutation point is identified. For the extra mutation points that cannot be filtered out, the “post-test simulation” combined with the fault branch discriminant is further filtered to identify the effective mutation points and achieve accurate fault location of the radial distribution network. Finally, the PSCAD simulation software is used to simulate the faults of different branches and the results show that the proposed method can accurately locate the fault points.
SHU Hongchun,LIU Jialu,TIAN Xincui.Fault Location of Radial Distribution Network Based on Sudden Changes of Fault Traveling Wave Along Line and Model Matching[J/OL].Automation of Electric Power Systems,http://doi.org/10.7500/AEPS20190522001.