Researchers at Sandia National Laboratories are tackling power grid instability as the rapid expansion of AI data centers increases the risk of sudden voltage drops that could disrupt critical infrastructure, including military installations.
The Distributed Energy Resource Management System (DERMS), which Sandia enhanced with artificial intelligence, is designed to regulate grid voltage in real time.
Unlike conventional systems that rely on capacitor banks and line regulators, it coordinates smart inverters connected to solar panels and battery storage to balance the grid more dynamically.
The system continuously monitors grid conditions, directing thousands of connected devices to counter voltage fluctuations as they occur.
It also predicts shifts in electricity demand and supply, coordinates distributed energy resources, and automatically responds to disruptions to keep voltage within safe operating limits.
“The way we generate electricity and the loads being placed on the grid are evolving, but the backbone of the grid that connects these is staying the same,” said Rachid Darbali-Zamora, an engineer at Sandia National Laboratories.
“We need more control to ensure everything can be integrated into the grid in a more reliable manner. A key goal is keeping voltage within operating limits as conditions change from second to second.”
Bringing It Into Real-World Tests
The researchers later evaluated DERMS at two sites in Lubbock, Texas: Sandia’s Scaled Wind Farm Technology facility and the Texas Tech University GLEAMM microgrid.
During the trials, the system was connected directly to live grid equipment to assess its performance under rapidly changing operating conditions.
Sandia also deployed DERMS at the Texas Tech GLEAMM microgrid, home to an active data center, where the tool coordinated grid-connected devices in real time.
According to the team, the system reduced site voltage, which typically runs about five percent above normal, bringing it closer to the utility’s target level.
“These demonstrations prove that AI can meaningfully improve how microgrids and distributed resources operate,” said Miguel Jimenez-Aparicio, a researcher at Sandia National Laboratories.
“The field data reinforces what we observed in PHIL testing. This technology can deliver real benefits to utilities, communities and critical infrastructure.”
