The Influence Mechanism of Incident-End Impedance Mismatch on Multi-Branch Positioning and the Design of Coupling Network
Aiming at the problem that insulation fault location in multi-branch cable networks of train DC medium-voltage power supply systems is disturbed by impedance mismatch at the injection end, this paper proposes an impedance matching method based on a resistance–capacitance composite coupling network to suppress false reflections and improve location accuracy. Firstly, the mechanism of multiple false reflections caused by injection-end mismatch is theoretically analyzed, and the expression of reflected waves and their influence on waveform superposition are derived. On this basis, an RC coupling network is designed to achieve effective matching between the measurement end and the cable characteristic impedance (about 97 Ω) within the 8–12 MHz frequency band, suppressing the amplitude of false reflections to less than 2% of the incident wave. Verification through MATLAB R2022b/ANSYS Q3D 2024R2 co-simulation and a 1:8 scaled experimental platform shows that the proposed coupling network reduces the absolute fault location error in a multi-branch network from 6.936 m to 0.188 m, decreases waveform distortion by about 40.8% and lowers the equivalent noise enhancement factor by about 55.2%. This study provides a reliable front-end matching solution for accurate fault location in complex cable networks, with clear value for engineering applications.
Authors
- Yuhua Wang (ORCID: https://orcid.org/0000-0003-3138-1312)
- Yu Han (ORCID: https://orcid.org/0000-0003-1462-1118)
- Jianqiong Zhang
- Hua‐Long Zheng (ORCID: https://orcid.org/0000-0002-2749-0113)
- Xiangqiang Li
- Yanming Han
- Qingfeng Wang
Institutions
- Central Committee of the Communist Party of China (CN)
- Southwest Jiaotong University (CN)
- Changchun University (CN)
Publication Details
- Journal
- Electronics
- Published
- 2026-03-17
- DOI
- https://doi.org/10.3390/electronics15061246
- Primary Topic
- Electrical Fault Detection and Protection
- Type
- article
- Field-Weighted Citation Impact
- 0.00