High Impedance Fault Detection Method Based on Variational Mode Decomposition and Teager–Kaiser Energy Operators for Distribution Network
Xiaowei Wang,Jie Gao,Xiangxiang Wei,Guobing Song,Lei Wu,Jingwei Liu,Zhihui Zeng,Mostafa Kheshti
DOI: https://doi.org/10.1109/tsg.2019.2895634
IF: 10.275
2019-11-01
IEEE Transactions on Smart Grid
Abstract:The focus of the paper is the difficulty of high impedance fault (HIF) detection in distribution network, and its ease to be confused with capacitor switching (CS) and load switching (LS). Based on the intermittent reignition and extinction characteristics of HIF current, this paper proposes a novel HIF detection method, which combines variational mode decomposition (VMD) and Teager–Kaiser energy operators (TKEOs). The HIF detection method is as follows: First, perform the VMD on transient zero sequence currents to obtain the intrinsic mode functions (IMFs) and select the IMFs with the largest kurtosis value as the characteristic IMFs. Second, calculate the characteristic IMFs to obtain TKEOs and divide into subintervals of TKEOs waveform to calculate the time entropy values. Finally, construct HIF detection criterion as follows: when time entropy value is 0, it is judged as CS or LS. When the entropy value is not 0, it is judged as HIF. A large number of simulations and field data tests show that the method is accurate and stable, and under the interference of 1 dB strong noise, it can accurately judge. Compared with other methods, the method has higher feature extraction accuracy, less calculation time, and better judgment accuracy.
engineering, electrical & electronic