Robust Single-Ended Fault Location in Double-Circuit Lines with Fault Topology Identification and Uncertainty Quantification
Single-ended fault location on double-circuit transmission lines is complicated by the mutual coupling between the circuits, the remote-end infeed and the decaying dc component of the fault current. This paper proposes a method in which the fault distance and the fault path conductances are estimated jointly from the condition that the remote-bus voltages calculated through the two circuits of an exact six-conductor distributed-parameter model must be equal. Thirty-one fault types, including the faulted circuit, are fitted by a weighted Huber estimator and ranked by the Bayesian information criterion; faults that cannot be located from one end are detected, and confidence intervals are obtained by Monte Carlo propagation. The phasors are estimated by a two-cycle least-squares fit that includes the decaying dc term, and instrument transformer errors and time skew between channels are corrected. The method and a recent single-ended method were applied to the same waveforms of 1658 simulated faults on a 500 kV line of 150–400 km. For the 990 noise-free cases of the main study, the median error was 0.0014 % against 0.270 %, and the fault type was identified correctly for 99.5 % of the tested observable faults. With protection-class instrument transformer errors and certificate correction, the median error was 0.071 % against 0.237 %. The coverage of the 95 % confidence interval was 98.1–100 % in all case sets. Errors in the zero-sequence impedance, an inductive fault impedance and an uncorrected current-to-voltage time skew remain the main error sources.
Authors
- Soma Deb (ORCID: https://orcid.org/0000-0001-7247-5468)
- Abhay Yadav
- V. P Singh
Institutions
- Sharda University (IN)
- Malaviya National Institute of Technology Jaipur (IN)
Publication Details
- Journal
- Research Square
- Published
- 2026-11-18
- DOI
- https://doi.org/10.21203/rs.3.rs-11513737/v1
- Primary Topic
- Power Systems Fault Detection
- Type
- preprint