Shanxi's 500 kV and above transmission line projects use drone AI for independent acceptance for the first time

Recently, at the Shanxi Shuozhou Shentou switching station-Yantong double-circuit line project site that was connected to the 500 kV line from the 1,000 kV substation in northern Shanxi, Shanxi's 500 kV and above transmission line projects used drone AI for independent acceptance for the first time.

"Traditional quality acceptance is mainly based on manual tower inspection and ground instrument measurement. It is easily affected by terrain, weather, measurement angle, etc., and has visual blind spots, which affects the acceptance effect." The relevant person in charge of the Shanxi Construction Branch of State Grid said , this innovative acceptance uses drone RTK + front-end AI visual navigation, automatic target tracking and locking, adaptive zoom shooting and other technologies. Compared with traditional RTK flight shooting, this innovative acceptance has the full automatic shooting of key parts and a pixel increase of more than 10 times. , automatic data classification management after the job is completed and other advantages. The data formed by AI automatic acceptance can provide data processing personnel with an immersive sensory experience and make it easier to detect minor defects.

This drone inspection captured the real-time installation status of transmission line towers, fittings, insulator strings and other parts, and accurately scanned the distance between ground wires, tree growth in line corridors, geographical environment, crossings and other conditions. These photos will be included in the project acceptance report and digital archives, and will be accumulated and improved to form a typical project defect database, providing the most comprehensive and basic original data for the reliable operation of the power grid, and at the same time providing guarantee for the "zero-defect" operation of the project.

In the next step, the company will further explore the data value of the standardized acceptance typical defect database, use typical neural networks to achieve autonomous classification and identification of collected images, automatically identify and alarm power grid engineering defects and form reports, and promptly report them to construction, supervision and construction personnel. , operation and other participating parties push information to realize process management of defect identification, information reminders, defect rectification reporting, and rectification confirmation, and realize digital and intelligent transformation in the field of power grid infrastructure.

Read more:  https://www.hemeielectricpower.com/