Open Access

ARTICLE

Fault Location of Distribution Network Based on Traveling Wave Head Inversion

Guanghua He¹, Jinlong Qi^1,*, Yao Feng¹, Jiayi Han¹, Heng Chen², Baoming Huang³, Jiangtao Li³
1 Wuxi Power Supply Company, State Grid Jiangsu Electric Power Company, Wuxi, 214000, China
2 Shandong Kehui Electric Automation Co., Ltd., Zibo, 255000, China
3 School of Electrical Engineering, Xi’an Jiaotong University, Xi’an, 710049, China
* Corresponding Author: Jinlong Qi. Email:
(This article belongs to the Special Issue: Advanced Analytics on Energy Systems)

Energy Engineering https://doi.org/10.32604/ee.2026.076354

Received 19 November 2025; Accepted 26 December 2025; Published online 14 January 2026

Abstract

The identification of the traveling wave head is an important factor affecting the accuracy of fault traveling wave positioning. In practice, in addition to the attenuation of traveling wave amplitude and rising speed caused by distribution line factors, various traveling wave sensors can also cause transmission distortion of high-frequency traveling wave signals, which in turn affects the calibration of traveling wave arrival time and the accuracy of fault distance measurement. The inversion technology of sensor transmission characteristics using analytical methods has limited ability to reflect factors such as stray capacitance and sensor differences. In comparison, data-driven artificial intelligence modeling inversion technology can better capture the details of sensor transmission characteristics and help improve positioning accuracy. Regarding this issue, a fault location method for distribution lines based on initial traveling wave head inversion is proposed in this paper. Firstly, the mechanism of fault traveling wave propagation distortion, including distribution lines and sensor factors, was analyzed. Furthermore, a traveling wave head inversion method based on a one-dimensional convolutional neural network was proposed. This method constructs an inversion model by training the detail coefficients of the primary and secondary traveling wave heads of the sensor after the wavelet transform as samples. Based on the training model, the wavelet detail coefficients of the secondary side wave after the fault occurs are inverted into the corresponding primary side coefficients, and the arrival time of the wave head is accurately extracted through the wavelet modulus maximum method, thereby improving the positioning accuracy of the double-ended traveling wave method. Experiments show that the average location error of this method does not exceed 2.2% under various fault conditions with a sampling rate of 1 MHz, and demonstrating the robustness and reliability of this method in complex fault scenarios.

---

Keywords

Traveling wave fault location; inversion of traveling wave head; convolutional neural network; voltage sensor; wavelet transform

---