UConn Eversource Energy Center’s Dr. Diego Cerrai on AI for storm and damage prediction

Photo by American Public Power Association on Unsplash

The increasing frequency and intensity of storms pose significant challenges to electric utilities, emphasizing the need for technological solutions to optimize storm response. Could AI be part of the solution to those problems?

Dr. Diego Cerrai, Associate Director for Storm Preparedness and Emergency Response at the University of Connecticut (UConn) Eversource Energy Center, will speak on this topic at DTECH (formerly known as DISTRIBUTECH) March 24-27, 2025 in Dallas, Texas.

Cerrai will participate in the session, AI for proactive storm response at Exelon and Oncor, taking place March 25, from 1 PM – 1:50 PM.

Accurate storm outage and damage prediction is crucial for mitigating the adverse impacts of power disruptions, enhancing grid resilience, and ensuring timely restoration of services. Emerging in-field and AI technology allows utilities to proactively allocate resources, optimize crew deployments, and communicate more effectively with customers and stakeholders. However, challenges remain, including data quality and availability, integration with existing systems, upstream process improvement, and change management. 

Exelon has teamed up with UConn to develop an outage prediction model for its area. As part of the partnership, UConn is developing four machine learning-based models for rain/windstorms, tropical storms, snow/ice storms, and thunderstorms. Panelists from Exelon, Oncor, and UConn will discuss their use of AI to enhance the tools and procedures for managing storms within their utilities. They will also address the existing challenges in securing stakeholder acceptance and trust and implementing these solutions in an operational environment.

Dr. Diego Cerrai is an Assistant Professor at the School of Civil and Environmental Engineering at the University of Connecticut, and Associate Director for Storm Preparedness and Emergency Response at the UConn Eversource Energy Center. He is leading the UConn Outage Prediction Modeling (OPM) team, focusing on the prediction of weather-related power outages using machine learning and statistical models. He developed the current version of the OPM, in collaboration with Eversource Energy, Avangrid, Dominion Energy, and Exelon. He is also leading projects related to grid resilience assessment, renewable energy resources integration in the electric grid, wildfire ignition prediction and storm tree damage prediction, and energy justice.


wind turbines in front of an orange sunset

Renewables permitting has been ‘paralyzed’ by Trump – This Week in Cleantech

This Week in Cleantech is a weekly podcast covering the most impactful stories in cleantech and climate in 15 minutes or less.

Redefining microgrids: A vision for dynamic grid control

Erik Amundson, vice president of microgrid engineering and chief technology officer at OATI, provided insights into the evolving role of microgrids during the OATI Energy…