A Parameter-Free Fault Location Algorithm for Hybrid Transmission Lines Using Double-Ended Data Synchronization and Physics-Informed Neural Networks

Guangjie Yang, Guojun Xu, Ruijing Jiang, Yanfeng Jiang, Xiaolong Chen (), Lirong Sun, Yitong Li and Yihan Gao
Additional contact information
Guangjie Yang: State Grid Handan Electric Power Supply Company, Handan 056000, China
Guojun Xu: State Grid Handan Electric Power Supply Company, Handan 056000, China
Ruijing Jiang: State Grid Handan Electric Power Supply Company, Handan 056000, China
Yanfeng Jiang: State Grid Handan Electric Power Supply Company, Handan 056000, China
Xiaolong Chen: School of Electrical and Information Engineering, Tianjin University, Tianjin 300072, China
Lirong Sun: School of Electrical and Information Engineering, Tianjin University, Tianjin 300072, China
Yitong Li: School of Electrical and Information Engineering, Tianjin University, Tianjin 300072, China
Yihan Gao: School of Electrical and Information Engineering, Tianjin University, Tianjin 300072, China

Energies, 2025, vol. 18, issue 21, 1-17

Abstract: Accurate fault location is crucial for enabling maintenance personnel to quickly reach the fault site for inspection and repair, thereby minimizing power outage duration. To address the low fault location accuracy caused by phase unsynchronization of double-ended recording data and the dependence of traditional algorithms on accurate line parameters, this paper introduces a novel fault location algorithm for hybrid transmission lines. The method integrates a data synchronization approach with a physics-informed neural network (PINN) implemented using a backpropagation (BP) neural network architecture. First, the proposed synchronization algorithm corrects the phase misalignment between double-ended recordings. Second, a distributed-parameter fault location model is developed to derive a location function, which is then used to construct physics-informed input features. This approach reduces the need for large fault datasets, addressing the challenge of the low occurrence of faults in practice. Finally, a BP neural network employing these physics-informed features is utilized to learn the nonlinear mapping to the fault location, allowing for accurate fault location, enabling accurate positioning without requiring precise line parameters. Validation using actual line data confirms the high precision of the synchronization algorithm. Furthermore, simulations show that the proposed fault location algorithm achieves high accuracy and remains robust against variations in fault position, type, transition resistance, inception angle, and load current, making it highly practical for real engineering applications.

Keywords: fault location; hybrid transmission line; double-ended data synchronization; physics-informed neural network; distributed-parameter model (search for similar items in EconPapers)
JEL-codes: Q Q0 Q4 Q40 Q41 Q42 Q43 Q47 Q48 Q49 (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/1996-1073/18/21/5710/pdf (application/pdf)
https://www.mdpi.com/1996-1073/18/21/5710/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jeners:v:18:y:2025:i:21:p:5710-:d:1783318

Access Statistics for this article

Energies is currently edited by Ms. Cassie Shen

More articles in Energies from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().