Cable fault identification and precise location method based on SSA-BP and ICEEMDAN-NTEO algorithms

Authors

• Fating YUAN, Hubei Power Transmission Line Engineering Technology Research Center, China Three Gorges University, Yichang 443002, China; College of Electrical Engineering and New Energy, China Three Gorges University, Yichang 443002, ChinaFollow
• Haoyue LI, College of Electrical Engineering and New Energy, China Three Gorges University, Yichang 443002, China
• Huqiang LI, College of Electrical Engineering and New Energy, China Three Gorges University, Yichang 443002, China
• Yi YANG, College of Electrical Engineering and New Energy, China Three Gorges University, Yichang 443002, China
• Shengkai JIAN, College of Electrical Engineering and New Energy, China Three Gorges University, Yichang 443002, China
• Yuqing JIANG, College of Electrical Engineering and New Energy, China Three Gorges University, Yichang 443002, China
• Bo TANG, Hubei Power Transmission Line Engineering Technology Research Center, China Three Gorges University, Yichang 443002, China; College of Electrical Engineering and New Energy, China Three Gorges University, Yichang 443002, China

Keywords

distribution lines; double-ended traveling wave location; SSA-BP algorithm; NTEO algorithm; ICEEMDAN algorithm; ATP-EMTP

Abstract

Existing power cable traveling wave ranging methods rely on the accurate identification of the initial traveling wave and suffer from the problem of inaccurate fault location. Based on the electromagnetic transient simulation software ATP-EMTP, this paper establishes a 10 kV power cable transmission line model and proposes a double-ended traveling wave location method based on back propagation (BP) neural network optimized by the sparrow search algorithm (SSA) and novel Teager energy operator (NTEO). Firstly, the power cable transmission line model is established, the fault current waveforms under different working conditions are studied, and the SSA-BP algorithm is utilized to identify the cable fault types. The prediction results of the training set and the test set show that the SSA-BP algorithm is able to accurately and quickly identify the power cable fault types. Then, through the phase-mode transformation of the three-phase current of the cable, appropriate fault components are selected for fault location of the cable according to the different fault types of power cables. The improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) algorithm is used to decompose the fault waveform, filter out the noise interference in the fault signal, enhance the initial traveling wave head characteristics by the NTEO algorithm, and accurately determine the time when the initial traveling wave arrives at the detector, so as to realize the accurate location of power cable faults. The method proposed in this paper achieves the fault identification accuracy of 98.3 % with the consideration of different short-circuit faults, grounding resistance, fault distance, etc. The fault location accuracy of the complete ensemble empirical mode decomposition (CEEMD)-NTEO and wavelet transform algorithms is 99.83 % and 99.67 %, respectively, while that of the proposed method is 99.88 %. The research results provide an important theoretical basis for the accurate identification and location of cable faults.

DOI

10.19781/j.issn.1673-9140.2026.02.012

First Page

127

Last Page

144

Recommended Citation

YUAN, Fating; LI, Haoyue; LI, Huqiang; YANG, Yi; JIAN, Shengkai; JIANG, Yuqing; and TANG, Bo (2026) "Cable fault identification and precise location method based on SSA-BP and ICEEMDAN-NTEO algorithms," Journal of Electric Power Science and Technology: Vol. 41: Iss. 2, Article 12.
DOI: 10.19781/j.issn.1673-9140.2026.02.012
Available at: https://jepst.researchcommons.org/journal/vol41/iss2/12