Open AccessSingle-phase-to-ground Fault Line Selection Method Based on STOA-SVM
Author(s) -
Yicen Liu,
Xiaojiang Liu,
Songhai Fan,
Xiaomin Ma,
Sijing Deng
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/2095/1/012029
Subject(s) - hilbert–huang transform , symmetrical components , support vector machine , fault (geology) , algorithm , computer science , nonlinear system , sequence (biology) , line (geometry) , pattern recognition (psychology) , artificial intelligence , mathematics , engineering , voltage , physics , geology , telecommunications , geometry , quantum mechanics , seismology , biology , electrical engineering , genetics , transformer , white noise
In view of the complex characteristics of nonlinearity and non-stableness of the zero-order current of each line after the single-phase ground fault of the distribution network, a distribution network fault selection method based on Sooty Tern Optimization Algorithm(STOA) and the combination of support vector machine is proposed. At first, the zero-sequence current before and after fault is obtained, then five kinds of IMFs including different components are obtained by ensemble empirical mode decomposition, and the energy entropy of the fault transient zero-sequence current is obtained by Hilbert transform, the results of training and testing are obtained by inputting the feature vector. The simulation results show that the accuracy of the proposed line selection model is 97.5%.