August 17, 2026

FSU researchers build GridFusionX AI model to forecast power grids across regions

FAMU-FSU engineers built GridFusionX, an AI grid forecasting model that improved accuracy by up to 56% and cut reserve costs by up to 66% in tests.

Electricity pylons carrying transmission lines across open countryside, illustrating the multi-regional power grid addressed by the AI grid forecasting model GridFusionX

Photo by Petr Ganaj on Pexels

Researchers at the FAMU-FSU College of Engineering and Florida State University’s Center for Advanced Power Systems published a new AI grid forecasting model called GridFusionX on August 3, 2026, designed to predict power system conditions across interconnected regions rather than one grid area at a time, according to an FSU News release. In tests across ten European regions, the model improved forecasting accuracy by up to 56 percent and cut reserve costs by as much as 66 percent compared with the researchers’ baseline methods, FSU reported.

The work, titled “GridFusionX: Network-Aware Probabilistic Forecasting for Multi-Regional Power Systems,” was published in IEEE Transactions on Network Science and Engineering. Doctoral student Quoc Bao Phan led the study, with co-authors Tuy Nguyen, Olugbenga Moses Anubi and Abdulrahman Takiddin, all faculty in FSU’s Department of Electrical and Computer Engineering, and associate professor Ravikumar Gelli, according to the university.

How the AI grid forecasting model works

GridFusionX combines multiple data streams, including past electricity demand, renewable generation output, energy market prices and grid network topology, using a graph neural network to model how conditions in one grid region affect neighboring regions connected through shared power lines, weather patterns and energy markets, FSU said. The model produces both forecasts and uncertainty ranges rather than single-point predictions, which the university said helps grid operators judge how much confidence to place in a given forecast.

That distinction matters as grids carry more variable renewable generation. Winss has reported that US renewables reached 30.3 percent of electricity generation in early 2026, a share that increases the forecasting challenge GridFusionX is built to address, since wind and solar output shift faster and less predictably than conventional generation.

GridFusionX test result Figure
Test regions 10 European regions
Forecasting accuracy improvement Up to 56%
Reserve cost reduction Up to 66%
Publishing venue IEEE Transactions on Network Science and Engineering
Lead researcher Quoc Bao Phan, doctoral student
Funding FAMU-FSU College of Engineering
GridFusionX reported improvements over baseline forecasting Reported improvement (%) 56% Forecasting accuracy 66% Reserve cost cut Source: FSU News, August 3, 2026, citing tests across 10 European regions
FSU reported GridFusionX improved forecasting accuracy by up to 56% and cut reserve costs by up to 66% in tests across ten European regions. Source: Florida State University News, August 3, 2026.

Nguyen described the aim of treating the grid as “a dynamic puzzle” in the FSU release. Anubi said the approach could also sharpen utility billing: “Right now, utility companies estimate your usage… With our approach, predictions become more precise, so your bill matches what you truly use,” he said. Gelli framed the broader goal as building “engineering intelligence” into grid operations. FSU’s release states the project was supported by the FAMU-FSU College of Engineering; it does not name an external funding agency or grant number.

FSU’s release does not specify the baseline forecasting method GridFusionX was measured against, and the underlying IEEE paper’s abstract page was not accessible to confirm additional methodological detail beyond what the university disclosed.

About FSU’s Center for Advanced Power Systems

The Center for Advanced Power Systems is a research unit within the FAMU-FSU College of Engineering, a jointly operated engineering school of Florida State University and Florida A&M University in Tallahassee, Florida. The center’s Cyber-Physical Systems Security Group, which produced GridFusionX, focuses on power systems research spanning grid security, forecasting and control. The August 2026 publication extends that work from single-region grid modeling toward forecasting that accounts for how neighboring grid regions influence each other, a direction the researchers say is increasingly relevant as more renewable generation and cross-border power flows make single-region forecasts less reliable on their own.


Sources: Florida State University News; TechXplore; IEEE Transactions on Network Science and Engineering

Featured image: photo by Petr Ganaj on Pexels (free Pexels license).


Become a Sponsor

Our website is the heart of the mission of WINSS – it’s where we share updates, publish research, highlight community impact, and connect with supporters around the world. To keep this essential platform running, updated, and accessible, we rely on the generosity of you, who believe in our work.

We offer the option to sponsor monthly, or just once choosing the amount of your choice. If you run a company, please contact us via info@winssolutions.org.

Select a Donation Option (USD)

Enter Donation Amount (USD)

What do you feel about this?