【目的】 针对现有多分支配电网故障定位中波头标定可靠性不高和经济性差的问题，提出一种基于直线检测（line segment detector， LSD）算法和模态时间差的多分支配电网故障定位方法。【方法】 首先，考虑到行波抵达母线时，初始故障波头呈现陡斜直线特征，为捕获该特征，将零模、线模行波转换为图像，利用LSD算法识别图像中亚像素级的陡斜直线段，进而实现故障行波波头的精确标定。其次，根据零模、线模行波在短距离配电网线路上传输速度不同的特点，仅需要通过测量主干线路两端初始行波波头的模态时间差，即可在分支线路未配置测量装置的前提下，实现对复杂多分支线路的故障定位，从而显著降低故障定位成本。最后，在MATLAB/Simulink中搭建多分支配电网模型进行仿真分析。【结果】 与传统故障定位方法进行对比，结果表明，所提方法平均误差约为6 m，误差降低约80%，并且在10 dB的强噪声条件下，仍能准确识别波头。【结论】 仿真结果表明所提定位方法可靠性高、经济性好、准确度高。

[Objective] To address the challenges of low reliability and high cost of fault location in multibranch distribution networks using wavefront detection， this study proposes a novel fault location method based on the line segment detector （LSD） algorithm and modal time differences.[Methods] First， considering that the initial fault wavefront exhibited steep linear characteristics when reaching the busbar， the method converted zero- and line-mode traveling waves into images. Using the LSD algorithm， the sub-pixel-level steep line segments in the images were identified to achieve precise wavefront detection. Next， by leveraging the difference in the propagation speeds of the zero- and line-mode traveling waves over short distribution lines， the fault location was determined solely by measuring the modal time difference of the initial traveling wavefronts at both ends of the main feeder. This approach eliminated the need for measurement devices on branch lines， significantly reducing the cost of fault location. Finally， a multibranch distribution network model was developed using MATLAB/Simulink.[Results] Compared with traditional fault location methods， the proposed method achieved an average error of approximately 6 m， representing an error reduction of approximately 80%. Furthermore， it exhibited accurate wavefront identification even under strong noise conditions of 10 dB.[Conclusions] Simulation results demonstrated that the proposed method achieved high reliability， cost efficiency， and accuracy in fault location.