Series Arc Fault Recognition Based on Multi-feature Extraction and GA_SVM

摘要：

　　To detect the series arc fault under the condition of 115V/400Hz power supply, the even harmonic sum and kurtosis index analysis is performed on the current signal before and after the arc fault in the point contact experiment.The selected method covers the characteristics of the time and frequency domains of the signal.The even harmonic sum and kurtosis values are extracted as the features for discriminating the arc fault, and the arc fault is identified by the support vector machine optimized by genetic algorithm (GA_SVM).Compared with the recognition accuracy of a single feature, the accuracy of the recognition after combining the features into a two-dimensional matrix combination is further improved.The result shows that the correct rate reaches 98.5714%.

关键词： Series fault arc Even harmonic sum Kurtosis index Support vector machine optimized by genetic algorithms

作者: Deshuan Tong Ze Li Ruihua Cui

作者单位: Hebei University of Technology, Tianjin, 300130 Hebei University of Technology, Tianjin, 300130;Electrical Apparatus Institute of Hebei Province, Ti

会议类型: 国际会议

会议名称: The 7th Intrenational Conference on Reliability of Electrical Products and Electrical Contacts(第七届电工产品可靠性与电接触国际会议)

会议地点: 江苏苏州

会议语种:英文

页码: 410-414

在线出版日期: 2019-11-04（万方平台首次上网日期，不代表论文的发表时间）