Power distribution line defect identification and anti-external damage method and system based on unmanned aerial vehicle image

CN122157036APending Publication Date: 2026-06-05STATE GRID LIAONING SHENYANG ELECTRIC POWER SUPPLY COMPANY

Patent Information

Authority / Receiving Office: CN · China
Patent Type: Applications(China)
Current Assignee / Owner: STATE GRID LIAONING SHENYANG ELECTRIC POWER SUPPLY COMPANY
Filing Date: 2026-02-09
Publication Date: 2026-06-05

AI Technical Summary

Technical Problem

Existing power distribution line defect identification algorithms lack accuracy and robustness, external force damage risk monitoring is passive and lagging with low automation, and there is a lack of proactive early warning systems based on drone automatic inspection and AI real-time judgment.

Method used

Deploy drone airports, plan refined flight routes, combine multi-model fusion algorithms with CNN and Transformer models for defect identification, monitor and warn of external damage risks in real time, and achieve all-weather automated inspection through drones.

Benefits of technology

It enables all-weather intelligent monitoring and early warning of power distribution lines, improves the accuracy of defect identification and operation and maintenance efficiency, reduces costs, and improves power supply reliability and safety.

✦ Generated by Eureka AI based on patent content.

Smart Images

Figure CN122157036A_ABST

Patent Text Reader

Abstract

The application provides a power distribution line defect identification and anti-external damage method and system based on unmanned aerial vehicle images, which comprises the following steps: deploying an unmanned aerial vehicle airport at a strategic position in a power distribution network operation and maintenance area; preliminarily planning and verifying an unmanned aerial vehicle route; formulating and executing an unmanned aerial vehicle inspection and patrol task; the unmanned aerial vehicle automatically completes the collection of power distribution line inspection images and real-time continuous video streams of underground cable areas in the inspection and patrol task; the power distribution line inspection images are processed by a power distribution line defect identification algorithm to output power distribution line defects; the continuous video streams are analyzed by a special excavator identification model; once an excavator is identified and it is confirmed that the excavator falls into a safety protection area electronic fence, an early warning containing geographic coordinates and a field video clip is generated to notify a duty officer to check. The power distribution line inspection is fully automated, intelligent, unmanned and remotely operable; the real-time automatic identification and positioning of the excavator in the underground cable passage are realized, and passive discovery is changed into active early warning.

Need to check novelty before this filing date? Find Prior Art