Open AccessSimulation Research on High-Speed Railway Dropper Fault Detection and Location Based on Time-Frequency Analysis
Author(s) -
Jin Li,
Xuewu Zhang,
Cheng Zhang,
Tiantian Tian
Publication year - 2020
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 - 10.1088/1742-6596/1631/1/012100
Subject(s) - support vector machine , computer science , pattern recognition (psychology) , string (physics) , position (finance) , artificial intelligence , set (abstract data type) , fault detection and isolation , fault (geology) , identification (biology) , time–frequency analysis , construct (python library) , independent component analysis , computer vision , actuator , mathematics , botany , finance , filter (signal processing) , seismology , economics , mathematical physics , biology , programming language , geology
In this paper, a machine learning detection method, namely, SVM-ICA, which aims to solve the fault identification of droppers in high-speed railway, was proposed based on time-frequency analysis. The proposed SVM-ICA method can be utilized to detect and locate the faulty droppers. In detail, the time-frequency statistical features of the senor data are firstly extracted and the significant features are selected to construct the training set. Secondly, based on the training set, the support vector machine (SVM) fault detection model is trained. Finally, the trained model is used for detecting faulty droppers, and further the position of the break string can be located by using the independent component analysis (ICA) method. Simulation results show that the proposed fault detection method of droppers based on time-frequency analysis can accurately identify if the string breaks, and additionally can locate the position where the failure occurred.