National Grid has turned to artificial intelligence (AI) to support the monitoring of over 7,000 miles of overhead wires and pylons across England and Wales.
For the past two years the organisation has been using six drones equipped with high-res still, video and infrared cameras to assess the pylons – checking steelwork, wear and corrosion and damage to conductors.
This avoids the need for engineers to climb up pylons or use helicopters, but it generates a huge amount of data which then needs to be analysed to determine whether maintenance work is necessary.
As a result, National Grid has started applying machine learning to process the drone data and cut the amount of footage that humans need to review.
Its chief executive John Pettigrew told the Guardian that the technology was an aid to existing staff and was not intended to replace them.
The KAI platform from UK startup Keen Ai “will determine the overall condition of the asset and whether it needs to replaced or repaired,” he explained. “We are just developing this as a prototype. When we talk about digitisation it’s real practical engineering-type stuff we’re doing.”
According to Keen Ai, its technology has cut the time taken to process footage by 66% and reduced the risk of asset failure by removing the backlog of footage waiting to be processed.
National Grid now hopes to hone the system to better identify problems.
“The results improve greatly as we supply more images to the application,” the organisation said.
Disruptive technologies and the next wave of connectivity will have a dramatic impact on the energy and utilities sector. To learn more about the opportunities and challenges that lie ahead, download our Next-generation connectivity report today.