Researchers at two US universities are working on an initiative to predict storm-related power outages, reduce restoration times and strengthen the resilience of the electric grid.

The project at the University at Albany in New York and University of Connecticut is called the North American Forecasting Weather, Outage, Load & Damage Initiative, and aims to create a scalable outage-prediction model to forecast system failures across the USA and Canada.

The forecasting model is the latest project to emerge from the Center for Weather Innovation & Smart Energy & Resilience (Wiser). Backed by a grant from the National Science Foundation, Wiser was established in 2023 to apply the research and expertise at UAlbany and UConn to solve problems related to energy systems, now and in the future.

Chris Thorncroft, director of UAlbany’s Atmospheric Sciences Research Center and Emmanouil “Manos” Anagnostou, director of UConn’s Institute of Environment & Energy, are leading Wiser for their respective universities.

“As the climate changes, extreme weather events are becoming increasingly disruptive and destructive,” Thorncroft said. “By combining high-resolution weather forecasts with real-world outage data from utilities, we believe we can create tools that help keep the lights on and improve power grid resilience. This represents a first-of-its-kind collaboration that has the potential to reduce economic impacts from severe weather across the country.”

Anagnostou added: “Anticipating storm impacts on the electric grid and its interdependent infrastructure would allow power utilities to work closely with state and local governments, communities and other stakeholders to take preemptive steps, such as requesting out-of-state assistance, increasing readiness of repair teams, as well as community energy, conservation and coordination, to mitigate the effects of storm impacts,”. 

The North American Forecasting Model will be a multi-year effort. The first phase will focus on local pilot regions, including New England, New York and California. Later phases will expand to regional areas along the east and west coasts, followed by a full North American rollout if additional funding is available.

The model will rely on UAlbany and UConn’s expertise in weather models and AI, combined with publicly available outage data, to improve system outage predictions over entire regions and help utilities more efficiently allocate resources before storms.

“This novel initiative uses AI to unite data from North American energy companies, turning a massive amount of information into a survival guide for the grid,” said Jan Woodcock, industry liaison officer for Wiser. “Learning from the entire continent’s history, our AI model doesn’t just watch the weather, it predicts damage before it happens, allowing us to keep the lights on and improve affordability.”

Tens of millions of Americans are now facing higher utility bills after regulators approved dozens of rate hikes last year. Regulators green-lit 43 rate hikes across the country in 2025, totalling $11.6bn in increases. This is especially true in New York, where electricity bills have spiked 33 per cent on average over the past decade, outpacing national growth.

Researchers involved with the initiative say better outage forecasts could help utilities position crews more precisely, speed up restoration and lower operating costs that ultimately impact rates.

“We have spent more than a decade building and validating power outage prediction models through one-on-one collaborations with utilities,” said Diego Cerrai, interim director of the UConn (uconn.edu) Eversource Energy Center. “The North American Forecasting Model enables us to move beyond individual utility partnerships and, for the first time, to develop a scalable platform with national reach, contributing to improving the energy security of North America.”

Jeff Freedman, research faculty at UAlbany’s Atmospheric Sciences Research Center, added: “Weather affects every aspect of the power grid, especially with more frequent and impactful severe events threatening the reliability of an aging transmission infrastructure. A critical component of maintaining a resilient energy system is our ability to predict, with precision, where and when weather-related power outages will occur. More accurate outage predictions help utilities better position crews in the field, reducing their costs, limiting disruptions to the power grid, and keeping electricity affordable for residential and commercial customers.”

This month, Wiser (www.albany.edu/wwwres/researchinitiatives/wiser/about.html) hosted its third annual industry advisory board meeting at UAlbany (www.albany.edu). During the meeting, the Wiser industry advisory board announced $550,000 in support to launch the North American Forecasting Model, along with seven other research projects that are related to improving the energy industry.