A fault location method and system for intelligent high-low voltage electrical line

CN122330587APending Publication Date: 2026-07-03JIANGXI QIANNUO CONSTRUCTION ENGINEERING CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
JIANGXI QIANNUO CONSTRUCTION ENGINEERING CO LTD
Filing Date
2026-05-06
Publication Date
2026-07-03

AI Technical Summary

Technical Problem

Existing fault location methods for high and low voltage electrical lines are difficult to effectively capture faults when their characteristics are weak. There is a disconnect between signal processing and advanced diagnostic models, and the methods lack self-evolution capabilities, resulting in location delays, insufficient accuracy, and poor adaptability.

Method used

A multimodal disturbance observer based on the physical equations of power transmission cables is used to extract low-intensity fault features. Variational Bayesian Kalman filtering and graph attention neural network are combined to perform signal super-resolution reconstruction and fault location probability localization. A dynamic fault knowledge graph is constructed for model self-evolution.

Benefits of technology

It significantly improves the accuracy and anti-interference capability of fault location in complex electromagnetic environments, reduces the risk of false detection and missed detection, enhances the system's adaptability and operational stability, and ensures power supply reliability and safety.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122330587A_ABST
    Figure CN122330587A_ABST
Patent Text Reader

Abstract

The application discloses a kind of intelligent high-low voltage electrical line fault locating method and system.The method comprises: constructing multi-modal disturbance observer based on power transmission physical equation, state space modeling and residual analysis are carried out on the electrical parameters of monitoring point, and low-intensity initial fault characteristic pulse packet is extracted;Using super-resolution anti-interference processing based on variational bayes Kalman filter and low-rank matrix approximation, time-frequency domain joint noise reduction and waveform reconstruction are carried out on fault traveling wave signal.The application realizes end-to-end collaborative optimization of each technical link by constructing comprehensive loss function across modules linkage, solves the problems of existing technology, such as lagging start point of early fault detection, information fault in signal processing and artificial intelligence model, and lack of cross-time closed-loop experience solidification ability of fault diagnosis model, significantly improves fault locating precision, anti-interference ability and system adaptive evolution ability.
Need to check novelty before this filing date? Find Prior Art