The Cyber-Physical Systems Security Group members, including (from left) Olugbenga Moses Anubi, Ph.D., Ravikumar Gelli, Ph.D., Abdulrahman Takiddin, Ph.D., doctoral student Quoc Bao Phan, and Tuy Nguyen, Ph.D., pose in the Simulation Lab with the NovaCor Real-Time Simulator at the Center for Advanced Power Systems in Tallahassee, Florida. (Scott Holstein/FAMU-FSU College of Engineering)
Key Points
FAMU-FSU College of Engineering researchers, working with FSU’s Center for Advanced Power Systems, have developed an AI grid forecasting tool called GridFusionX that helps utility operators predict electricity demand and renewable energy generation more accurately.
The tool, described in a study published in IEEE Transactions on Network Science and Engineering, uses a graph neural network to model how power conditions in one region affect neighboring regions.
In tests across 10 European regions, GridFusionX improved forecasting accuracy by up to 56% and cut reserve costs by as much as 66%.
Doctoral student Quoc Bao Phan led the study, working alongside FAMU-FSU faculty members Olugbenga Moses Anubi, Ravikumar Gelli, Tuy Nguyen and Abdulrahman Takiddin. The research is based at the college in Tallahassee, Florida.
Researchers at the FAMU-FSU College of Engineering and Florida State University’s Center for Advanced Power Systems (CAPS) have developed an artificial intelligence tool that could help make electric grids more reliable and reduce operating costs.
The system improves forecasts of electricity demand and renewable energy generation, giving grid operators better information to balance supply and demand.
FAMU-FSU College of Engineering researchers, working with FSU’s Center for Advanced Power Systems, have developed an AI grid forecasting tool called GridFusionX that helps utility operators predict electricity demand and renewable energy generation more accurately.
The tool, described in a study published in IEEE Transactions on Network Science and Engineering, uses a graph neural network to model how power conditions in one region affect neighboring regions.
In tests across 10 European regions, GridFusionX improved forecasting accuracy by up to 56% and cut reserve costs by as much as 66%. Doctoral student Quoc Bao Phan led the study, working alongside FAMU-FSU Department of Electrical & Computer Engineering faculty members Olugbenga Moses Anubi, Ravikumar Gelli, Tuy Nguyen and Abdulrahman Takiddin. The research is based at the college in Tallahassee, Florida.
What problem does GridFusionX solve?
As more renewable energy comes online, today’s electric grid has become increasingly difficult to manage. Uncertainty in forecasts can force operators to keep too much reserve power, raising costs, or too little, increasing the risk of blackouts.
The FSU-developed model addresses that challenge by treating the power grid as a connected network to generate more accurate forecasts. The research was published in IEEE Transactions on Network Science and Engineering.
The forecasting system, called GridFusionX, combines information such as past electricity demand, power generated from renewable sources and energy market prices to generate predictions and confidence intervals. That helps operators run the grid more efficiently with less wasted reserve power.
“Our research looks at smart systems as a dynamic puzzle: every factor, from different energy sources, to shifting demands in cities and changing prices and how they fit together,” said study co-author Tuy Nguyen, an assistant professor in the Department of Electrical and Computer Engineering. “What happens in one area can affect neighboring areas because they share power lines, weather patterns and energy markets. By continuously analyzing these pieces, we can predict energy needs more accurately and adapt to ensure both reliability and cost savings.”
How does GridFusionX work?
The main innovation in the AI tool is multi-modality: the ability to draw on many sources of information in a single estimate.
“Whether we’re predicting power usage, traffic patterns or even weather, reducing uncertainty in our predictions lets us make smarter, more adaptive decisions that benefit daily life,” said Associate Professor Olugbenga Moses Anubi, a co-author of the study.
The system uses a graph neural network, a mathematical representation of connected objects, to map how conditions in one region of the power grid can affect neighboring regions. The tool addresses a gap in existing methods by providing both spatially connected predictions and measures of uncertainty.
The model’s precision allows grid operators to plan reserves more efficiently and respond proactively to sudden changes, such as spikes in demand or drops in renewable generation. In real-world tests across 10 European regions, GridFusionX improved forecasting accuracy by up to 56% and cut reserve costs by as much as 66%, while maintaining reliable service.
Why does more accurate power grid forecasting matter?
Today’s power grids have grown increasingly complex. Regional power nodes are linked not just by electrical lines but by market economics, weather patterns and other factors that affect power usage.
“Right now, utility companies estimate your usage to make sure you never underpay, which almost always means you overpay,” Anubi said. “With our approach, predictions become more precise, so your bill matches what you truly use.”
Knowing how much energy is needed, and when, is fundamentally a problem of predictability.
“The main vision is engineering intelligence,” said Associate Professor Ravikumar Gelli. “We want to help utility operators balance generation and load and help consumers pay less. That’s the vision behind it.”
How is this research training the next generation of energy engineers?
FAMU-FSU faculty are incorporating concepts from this research into courses on cybersecurity for electric grids and artificial intelligence for power systems. Doctoral student Quoc Bao Phan led the study.
“It benefits students to work on a project like this because they have the chance to work with four faculty instead of just one,” Anubi said. “This mentorship serves as a model for other students and other kinds of collaborations. There are good learning and research outcomes for students, and that’s also an important part of this project.”
Assistant Professor Abdulrahman Takiddin was also a co-author of the study, which was supported by the FAMU-FSU College of Engineering.
Visit the FAMU-FSU College of Engineering website for more information about research at the college. Visit the Center for Advanced Power Systems website to learn more about advanced power system engineering research at CAPS.
Editor’s Note: This article was edited with a custom prompt for Claude Sonnet 5, an AI assistant created by Anthropic. The AI optimized the article for SEO/GEO discoverability, improved clarity, structure and readability while preserving the original reporting and factual content. All information and viewpoints remain those of the author and publication. This article was edited and fact-checked by college staff before being published. This disclosure is part of our commitment to transparency in our editorial process. Last edited: 08/03/2026.
RELATED ARTICLES
RIDER Center Partners in $28.7M Grid Resilience Project
Engineering Professor Will Improve Electric Grid Cybersecurity with $2.9M Department of Energy Award
Researcher Controls Voltage Response for Safer Electric Grid
FAQ
GridFusionX is an artificial intelligence forecasting tool developed by researchers at the FAMU-FSU College of Engineering and Florida State University’s Center for Advanced Power Systems. It combines electricity demand data, renewable generation data and energy market prices to predict how much power a grid will need, along with a confidence interval for that prediction.
GridFusionX uses a graph neural network, a machine learning method that models connections between objects, to treat the power grid as a network of linked regions rather than isolated points. This lets it account for how demand, weather and market conditions in one area affect neighboring areas, producing more accurate, spatially aware forecasts with measured uncertainty.
In real-world tests across 10 European regions, GridFusionX improved forecasting accuracy by up to 56% and cut reserve costs by as much as 66%, while maintaining reliable electric service, according to the research team.
GridFusionX was developed by a team at the FAMU-FSU College of Engineering, led by doctoral student Quoc Bao Phan. Co-authors include Associate Professor Olugbenga Moses Anubi, Associate Professor Ravikumar Gelli, Assistant Professor Tuy Nguyen and Assistant Professor Abdulrahman Takiddin, all of the Department of Electrical and Computer Engineering. The research was supported by the FAMU-FSU College of Engineering and conducted through FSU’s Center for Advanced Power Systems.
More accurate forecasting means utilities need to hold less reserve power on standby, which lowers operating costs. Researchers say those savings could translate to bills that more closely reflect what customers actually use, rather than conservative estimates that tend to result in overpayment.
The research was published in IEEE Transactions on Network Science and Engineering, a peer-reviewed engineering journal.