Fault Current Identification of DC Traction Network Based on EEMD Sample Entropy

LENG Yue,YANG Honggeng,WANG Zhiqi
DOI: https://doi.org/10.11857/j.issn.1674-5124.2016.12.020
2016-01-01
Abstract:Aiming at the frequent malfunctions of relay protection system caused by the oscillation current of Metro DC traction network,an identification approach of the oscillation current and the short-circuit fault current in DC traction network combined ensemble empirical mode decomposition (EEMD) with sample entropy is presented. In this method, the feeder current is decomposed by using EEMD method,and the sum of the sample entropy of each intrinsic mode function (IMF) is figured out. Thus, the feature which involves information on the operation state of DC traction network is acquired. From the analytical calculation of the typical feeder currents,it can be found out that the feature extraction method combined EEMD and sample entropy can effectively distinguish the oscillation current and the short-circuit fault current of DC traction network. Examples verify the effectiveness of the method.
What problem does this paper attempt to address?