The huge, beautiful treetops of the western US, which can grow dangerously close to power lines, can quickly start devastating wildfires. In fact, 70% of power outages are caused by vegetation, and this number has increased by 19% year over year from 2009-2020. The second largest wildfire in California history, The Dixie Firearose when high-voltage cables came into contact with a pine tree.
Can AI-powered solutions help prevent wildfires before they start by analyzing the tree growth they can cause? Hitachi Energy, the Zurich, Switzerland-based global technology company, says yes.
AI is critical to the future of renewable energy
Hitachi Energy, formerly known as Hitachi ABB Power Grids (the name was changed last October) is currently focused on “powering good for a sustainable energy future”. A concern was how to position itself to serve off-grid customers and help industries that have connectivity resources spanning large geographic areas, including energy, telecommunications, water, gas pipelines and railways. That includes utilities that handle thousands of miles of growing vegetation.
“These industries all have similar issues managing their miles and miles of assets,” Bryan Friehauf, SVP of business software solutions at Hitachi Energy, told VentureBeat. “For example, you need to keep trees away from railroad tracks and roads, but also away from gas pipelines and other critical infrastructure.”
Three trends have made the use of geospatial and AI-powered technology critical, he explained: aging infrastructure, silo systems and climate change. “It can be difficult or dangerous to view or manage assets under these conditions,” he said.
Inspect trees to prevent forest fires and breakdowns
To address these challenges, Hitachi Energy today announced a new AI-powered solution, Hitachi Vegetation Manager, part of the company’s new Lumada Inspections Insights offering. The company claims it is “the first of its kind, a closed-loop vegetation resource planning solution that uses artificial intelligence and advanced analytics to improve the accuracy and effectiveness of an organization’s vegetation operations and planning efforts.”
Using algorithms developed at one of the company’s research and development centers in Japan, the solution creates images of trees and forests from a variety of visual sources, including photos, videos and images from leading Maxar satellites. By combining the images with climate, ecosystem and cutting plan data and machine learning algorithms, Hitachi Vegetation Manager provides utilities with visibility across the network and better insights so organizations can optimize decision making.
Satellites capture images, AI analyzes them
“With satellites capturing images remotely and AI analyzing them, we can better optimize and plan to address problem areas,” says Friehauf. “This will also reduce management program costs and emissions by minimizing truck and helicopter downtime and ultimately minimizing downtime and fires caused by vegetation.”
Using AI to track and analyze vegetation is particularly essential for utilities around the world, which are facing unprecedented climate-related challenges. In 2021, global wildfires generated an estimated 6,450 megatons of CO2 equivalent – about 148% more than total EU fossil fuel emissions in 2020†
According to John Villali, research director at IDC Energy Insights, inspection, planning and monitoring are “among the most critical tasks utilities undertake to maintain network reliability and resilience. Hitachi Energy’s AI-powered solution, he explained, enables utilities to improve decision-making, optimize operations, and “as a result, achieve their reliability, safety and sustainability goals.”
Utilities adopt AI more easily
Historically, as a highly regulated industry, the utilities industry has not been a leader in AI and other emerging technologies, said Phil Gruber, general manager in the energy/industrial utility practice at Hitachi Vantara, Hitachi’s IT services management company. “The utilities sector is often very cautious for good reason and generally isn’t a leader in using technology, but they’re starting to flounder,” he said.
One problem is that organizations often think they don’t have enough good quality data to get started with AI or ML. “A lot of our discussions with clients are about trying to meet them where they are with the data sets they have,” Gruber says. “We often find that they have enough data to really improve their decision making and results.”
But Hitachi Energy’s solution means utilities no longer need arborists to traverse miles of transmission lines to identify each species, Friehauf explained. Once species data is entered into the model, including location and details such as soil quality, the algorithm can take precipitation data about the weather, analyze the growth profile of tree species, and predict where growth will and will not occur.
“Of course the precipitation is not homogeneous, so you can even have areas within the same province that have more precipitation than another,” Friehauf said. “The tool can show that even though you have cut back some vegetation, you may need to do it sooner because it rains a lot, or if you’re in a drought, you need to know how drought-resistant a species is.”
Overall, the Hitachi Vegetation Manager “gives you a very accurate forecast of how that vegetation is growing,” said Friehauf. “This is important not only for utilities dealing with wildfires or blackout risks, but for anyone who needs to manage the vegetation around their linear assets.”