As California’s largest power utility declares bankruptcy amidst a sweep of lawsuits for its role in recent wildfires, utilities nationwide are looking for methods to inspect powerlines with greater efficiency and ease. Currently the industry standard is to check all powerlines only once per year, but for companies like Pacific Gas & Electric (PG&E), whose lines traverse hundreds of thousands of miles, even this standard can fall out of reach.
Typically, inspections are made by helicopter, with workers hovering a mere 10 feet away from high-voltage lines that can transmit as much as 765 kilovolts of electricity. These inspections come at significant expense to the companies and some measure of danger to the employees. Each year about 22 American power line workers die in related accidents accordingly to the U.S. Department of Labor.
As Stanford students Kaitlyn Albertoli, ’19, and Vik Chaudhry, MS ’17, learned more about the process, they began to wonder if there was a better way. What if utilities could do powerline inspections monthly instead of yearly—and with zero risk of death by electrocution? Better, right?
Machine learning, human challenges
The duo met through the Stanford course CEE 246: Venture Creation for the Real Economy. It was spring 2017 and Chaudhry, about to complete his master’s, was eager to leverage an artificial intelligence platform he had developed for analyzing wind energy. Albertoli, though only a sophomore at the time, had an entrepreneurial heart and two previous businesses under her belt.
Initially they had planned to use Chaudhry’s platform to advise wind farm owners how to better orient their turbines, but as they studied the antiquated state of the modern electrical grid in the United States, they pivoted. They felt it was a more important problem to be solved.
“I was shocked to learn that thousands of maintenance requests go unanswered every year. There’s just not enough time to deal with them—until they become a bigger problem,” says Albertoli.
During the class, they reached out to 92 utilities and interviewed all 30 of those that replied. They would ask, “What is your biggest challenge?” And they heard the same responses repeatedly: geographic distance to cover, the expense of the inspection process, the challenge of weather.
With that framework in place, Buzz Solutions took flight and applied for an Innovation Transfer Grant from the TomKat Center that summer. After commencement, Chaudhry became the venture’s first full-time employee.
Send in the drones
Some powerline issues are glaringly obvious to the naked eye—a kudzu vine overtaking a tower or sagging transmission lines that nearly touch the dirt. Other tiny faults are easier to find with cameras; thermal imaging can detect heat, a tell-tale sign of defects in the lines, or the glow of coronal discharges, pinpointing other transmission line deficiencies.
“Our value proposition is in the software,” says Albertoli. Buzz Solution’s imaging and analytic platform is predictive, allowing utilities to use cameras to get out in front of equipment failures and brush fires. “We provide the where and when of an outage before it happens.”
Understanding their value was tied to their platform, and not the inspection vehicle, Chaudhry and Albertoli began to explore new approaches. They realized that drones—nimble, autonomous, and far less expensive to operate for long miles—are better suited to the job than helicopters. Plus, Buzz Solutions can reduce capital costs by simply teaming up with drones available for hire with FAA-certified pilots rather than needing to build up a fleet themselves.
The selling point for utilities? Fifty percent cost savings on annual inspections, and the Buzz Solutions team anticipates that monthly inspections could become a reality using their agile approach.
At the time of publication, drones outfitted with Buzz Solutions technology were just finishing digitally capturing and analyzing 10,000 miles of powerlines across Ontario, Canada, as part of the venture’s first pilot program.
“We’re constantly exploring new markets and learning how we can better assess power companies’ priorities and needs. That’s our biggest goal right now,” says Albertoli. The venture is currently collaborating with four Canadian and U.S. utilities.
Meanwhile, the team is leveraging every last resource available to Stanford students before Albertoli graduates. From the Cardinal Ventures accelerator to the Stanford Venture Studio, she says she will be sad to lose these student resources, but relieved to focus on Buzz Solutions exclusively.
After years of juggling school and startup, she expects one constant will continue after graduation. “There’s never a dull moment.”