Energy and powerNews

Californian vegetation experts have improved powerline inspections with satellites and AI

Californian vegetation experts have improved powerline inspections with satellites and AI

Image courtesy ACRT

In California, two companies with expertise in utility vegetation management teamed up for a pilot project to validate and analyse satellite imagery to identify individual trees at risk of striking electric transmission and distribution lines. The partners say their results are groundbreaking.

Companies Satelytics and ACRT Pacific, automated cloud-based geospatial analytics and utility vegetation management companies respectively, teamed up on the project, which analysed 50cm resolution Pléiades 1A/1B satellite imagery.

According to the partners, the implications of their pilot project are significant for the utility vegetation management industry, enabling field workers to save time during routine inspections by focusing on trees that have already been tagged as problems.

In the project, ACRT Pacific leveraged Satelytics’ vegetation encroachment algorithms to identify 57,000 trees that posed a strike risk to electric lines in an area covering 10 square kilometres.

ACRT Pacific then validated the accuracy of the results with its teams of experienced utility vegetation arborist inspectors.

Fieldwork confirmed that by using an AI-based process, the team was able to pinpoint trees that exhibited the potential to strike the lines. Using stereo pair imagery at the optimal time and angle of collection resulted in up to 90% accuracy in these alerts.

Have you read:
Australia’s SA Power Networks pioneers satellite connectivity for remote workers
Satellite imagery used to identify high-voltage power line faults

Utility vegetation management: actionable intelligence and demand-sagged powerlines

Commenting in a release was Brian Joiner, president of ACRT Pacific: “The primary advantage of using satellite imagery in this analysis is immediacy.

“This provides actionable intelligence quicker than traditional field crews or airborne LiDAR. By prioritising trees with condition changes, it alerts pre-inspectors, allowing workers to promptly verify trees requiring attention on the ground.”

Joiner explained that summer storms are a significant threat to transmission operations. During hot days, when electricity demand is high, the combination of outside temperature and power load makes the lines sag out of their normal orientation.

Add in high winds and the lines start to sway, potentially contacting vegetation that otherwise might be out of their safe buffer proximity. The calculations also consider the swaying of the trees in windy conditions.

Said Sean Donegan, president and CEO of Satelytics: “The stakes in safeguarding utility infrastructure cannot be overstated.

“The potential liability associated with a vegetation-related wildfire – whether the loss of life or damage to assets – is measured in billions of dollars. These high-resolution satellite imagery as well as spatial and location data, can precisely pinpoint these trees, with fieldwork conducted by ACRT confirming a remarkable 90% accuracy rate.

utility vegetation management
UVM collage, courtesy ACRT

“Power outages, wildfires, infrastructure damage and more can all be caused by one neglected tree — you only need to miss one to cause a disaster,” Donegan added.

“Satelytics’ AI-enhanced geospatial analytics provides an immediacy that was previously unthinkable, automatically flagging events and keeping assets and surrounding communities safe in the process.”

According to the companies in a release, Satelytics’ AI algorithms can identify dangers down to the individual tree level through analysis of 50cm resolution stereo pairs captured by the satellites.

ACRT Pacific worked with Satelytics in developing and refining vegetation management algorithms to a level of sophistication unmatched in the market today, they add.

Remotely sensed data, often imagery or LiDAR point clouds, has commonly been used to find tree branches that have extended into the buffer zone around electric lines.

These data sets also help to identify trees that have grown tall enough to strike a line if they were to fall and identify trees that will encroach the compliance zone if not mitigated.

Joiner said that field crews won’t be replaced by this method, but instead, make these teams more efficient during their site visits.