By GREGORY ZELLER //
Handing control of the national power grid over to artificial intelligence? What could possibly go wrong?
Remain calm, humans. What sounds like the opening of every man-vs.-machine sci-fi epic you’ve ever seen is actually a giant leap toward safety and reliability, according to researchers, who say an AI-driven national energy network represents our most efficient and secure energy future.
And the National Science Foundation agrees, as evidenced by a $5 million “cooperative agreement” that will fund 10 national research teams – led by Stony Brook University – on a mission to develop an AI-enabled, autonomous power grid invulnerable to cyberattacks, systemic faults and disastrous accidents.
Peng Zhang: Natural intelligence.
Stony Brook University Professor Peng Zhang and collaborators from across academia, industry and government are already working hand-in-hand on the “AI-Grid,” under the auspices of the NSF’s Convergence Accelerator Program, which supports basic multidisciplinary research designed to hasten society-improving solutions.
In this case, the solution would be a more resilient national energy network. The work started in 2020 with a Phase 1 NSF award of $1 million and continues with the $5 million add-on announced Sept. 16; Phase 2 will “focus on expanding the solution prototype and building a sustainability plan beyond the NSF funding,” according to SBU.
To date, more than 30 partners – academic institutions, power utilities, local and state governments and others – have contributed to the mission, which has involved everything from deep-learning computers to encrypted controls to fully programmable microgrids.
Now, according to principal investigator Zhang, a SUNY Empire Innovation Professor in the Department of Electrical and Computer Engineering in SBU’s College of Engineering and Applied Sciences, it’s time for the long pants.
“We expect to show that our AI-Grid solution is affordable, lightweight, secure and replicable, thus offering what could be unprecedented flexibility for an approach to transform today’s infrastructures into tomorrow’s autonomous AI-Grid,” Zhang said, trumpeting the AI-Grid’s enormous potential for energy resilience and economic equality.
Looks innocent enough: The AI-Grid prototype, under construction at SBU.
“[Phase 2] will demonstrate AI-Grid’s capability to empower our nation’s digital economic engines, relieve the pains of those communities suffering from high electricity costs and knock out low-energy reliability and poor resilience,” he added.
Stony Brook-based contributors include the university’s Department of Electrical and Computer Engineering and the Department of Computer Science – a “multidisciplinary collaboration (that) strengthens our research enterprise while demonstrating to our students how complicated problems are solved in the modern world,” according College of Engineering and Applied Sciences Interim Dean Jon Longtin.
“The technologies that Peng and his team are developing come at a critical time as the nation pivots toward renewable energy and the inevitable impact it will have on the grid in the coming decade,” added Longtin, also a CEAS professor of mechanical engineering.
Jon Longtin: Critical pivot.
Off-campus Phase 2 collaborators include Brookhaven National Laboratory, New England-based utility Eversource Energy, PSEG Long Island, Hitachi America and several other academic and industrial partners.
With an eye on the future, Zhang’s AI-Grid team recently established end-user agreements with Energy and Innovation Park, a fuel cell-centered, grid-connected energy project based in Connecticut; the Epic Institute, a California-based global climate-solutions organization; and the Chicago-based Bronzeville Community Microgrid project.
Before Phase 2 is over, those end-users are expected to test, demonstrate and, if all goes according to plan, implement the AI-Grid technology – precisely the kind of high-speed, highly productive collaboration the Convergence Accelerator Program envisions, according to Douglas Maughan, who leads the NSF effort.
“A convergence approach is essential to solving large-scale societal challenges, which is why the NSF Convergence Accelerator requires our funded teams to include a wide-range of expertise from academia, nonprofits, industry, government and other communities,” Maughan said in a statement. “The merging of ideas, techniques and approaches combined with human-center design concepts assists our teams in accelerating their ideas toward solutions within three years.”
