Power management company Eaton today announced a $7 million, 24‑month contract from the U.S. Air Force Research Laboratory (AFRL) to apply quantum computing, machine learning and advanced visualization to improve resilience and protection at the power grid.
The effort will advance new quantum-enabled algorithms and hybrid quantum‑classical methods, developed with partners Infleqtion and Penn State, to better detect, visualize and respond to multiple, concurrent physical and cyber threats on the grid.
Today, the North American Electric Reliability Corporation (NERC), a not-for-profit international regulatory authority based in Washington, D.C., requires transmission systems to withstand two sequential failures (N‑2). Eaton's project aims to analyze and prepare for multiple, concurrent and unpredictable threat combinations.
Eaton will address a key security challenge in electrical grid management known as the contingency problem, which involves evaluating countless grid configurations to identify potential vulnerabilities. Eaton will combine its expertise in intelligent power management with the computational power of quantum computing for grid resilience.
“We’re facing unprecedented risks to electric reliability and security from extreme weather, wildfires, physical and cyber threats and need tools that consider many failures at once,” said Sid Suryanarayanan, senior chief engineer, strategic partnerships and innovation at Eaton.
Eaton research will combine specialized quantum hardware from Infleqtion, advanced machine learning algorithms and artificial intelligence (AI) supported by Pennsylvania State University to deliver solutions that enhance infrastructure planning, daily operations and emergency preparedness.
The project includes the development of new quantum algorithms, optimizing circuits for hybrid computation, conducting experiments on multiple quantum hardware platforms, and correcting and mitigating errors.
Results will be shared in a proof-of-concept demonstration showcasing how current quantum hardware combined with new algorithms and machine learning approaches can solve real-world grid challenges.
