Detection of DC Series Arc Fault Based on VMD and ELM. Issue 6 (April 2020)

• Record Type: Journal Article
• Authors: Ma, Tao; Tian, Ersheng; Liu, Zhenxing; Liu, Shuxin; Guo, Tianhong; Wang, Taowei; Fu, Long
• Is Part Of: Journal of physics. Volume 1486:Issue 6(2020)
• Publication Date: 2020-04
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/1486/6/062037
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: With the increase of domestic electrical equipment, the incidence of electrical fires has also increased, and research on fault arc detection has become a hot topic today. In this paper, a method combining variational mode decomposition (VMD), and extreme learning machine (ELM) is proposed to detect arc faults accurately. The characteristic signals of the resistance, capacitance and inductive load under normal conditions and arc fault conditions were collected by experiments. Then, the current data was processed by variational mode decomposition (VMD). Due to the different spectral characteristics of normal mode, arc fault mode and switching transient mode, the intrinsic mode function (IMF) under arc fault mode can be selected. Finally, according to the characteristic of determined IMF components, a new arc fault criterion was proposed for general DC arc detection. The experimental results verified that the proposed method can detect arc faults accurately.