Ziming Zhu, Xiaoyong Yu, Ke Zhou, Tanghua Wu
Multi-branch overhead lines; cloud-edge collaboration; fault localiza-tion; traveling-wave-based ranging
Aiming at the problems of insufficient fault localization accuracy of multi-branch overhead lines and the limited real-time performance of existing centralized architectures, this study proposes a hierarchi- cal fault localization method based on cloud-edge collaboration. By integrating the high-frequency signal acquisition capabilities at the edge with the in-depth data analysis advantages of the cloud, this method achieves efficient processing of fault signals and fusion of multi-source data. A fault distance matrix is established based on the principle of dual-terminal traveling wave ranging, and a fault deter- mination matrix is generated in combination with a defined structural distance matrix. The algorithm’s anti-interference capability is en- hanced through differential operations and correction margins. Sim- ulation experiments demonstrate that the proposed method achieves localization errors of less than 20 meters across various fault scenar- ios. Quantitative comparison with conventional methods confirms a significant enhancement in localization performance, with error reduction ranging from approximately 74.4% to 91.4%. Compared with traditional single-ended methods and artificial intelligence algo- rithms, it demonstrates significant improvements in adaptability to complex network topologies, dynamic response efficiency, and fault tolerance, providing a new approach for intelligent fault localization in distribution networks.
Important Links:
Go Back
