AI Data Centers Get New Power Architecture
News related to:TerraFlow Energy · 3 min read
HOUSTON, Sept. 28, 2026 /CourierPR/ -- TerraFlow Energy and DG Matrix have announced a commercial agreement to deploy the first integrated vanadium redox flow battery (VRFB) and solid-state transformer (SST) architecture in the United States. The integration of TerraFlow's LDUPS™ long-duration uninterruptible power system with DG Matrix's Interport™ solid-state transformer technology will create a new data center power architecture that manages compute loads while protecting the grid.
The initial deployment will integrate TerraFlow's LDUPS™ long-duration uninterruptible power system with DG Matrix's Interport™ solid-state transformer technology and Dell PowerEdge servers, putting a new data center power architecture into operation against real compute loads. The significance of this integration goes beyond combining two technologies. AI data centers are introducing enormous new loads to the power system, but the challenge is not only how much electricity they consume. High-performance compute can change power demand rapidly, pushing that volatility back through electrical infrastructure built for more predictable loads.
The integrated architecture is designed to manage power between the grid, on-site generation, energy storage, and compute in real time. TerraFlow's LDUPS™ combines the millisecond response required for UPS functionality with hours of energy capacity, allowing it to continuously buffer changes in compute demand. DG Matrix's Interport™ provides programmable solid-state power conversion, routing, and control between power sources and the compute load. Together, the technologies are designed to absorb rapid changes in demand before they reach the grid and other upstream electrical infrastructure.
The commercial agreement establishes a framework for TerraFlow and DG Matrix to bring the integrated LDUPS™ and Interport™ architecture into data center projects beyond the initial deployment. The companies are targeting applications where AI and high-performance computing require greater control over how power is delivered, managed, and exchanged with the grid.
The architecture also addresses a growing issue facing utilities and data center developers: how to connect gigawatts of new compute demand without requiring the power system to respond instantaneously to every change occurring inside the data center. By controlling those changes behind the meter, integrated storage and power distribution can help make large loads more predictable, more flexible, and ultimately easier for the grid to serve.
For TerraFlow and DG Matrix, the initial deployment is the starting point for a commercial architecture designed to scale with the rapidly growing power requirements of AI infrastructure.
TerraFlow Energy is transforming energy storage from backup equipment into critical power infrastructure. Built on long-duration, non-flammable vanadium flow battery technology, the company's battery-in-building architecture combines uninterrupted power, long-duration energy storage, and controllable load management in a single platform that simplifies electrical infrastructure for AI data centers and other energy-intensive facilities. By turning large electrical loads into flexible grid assets, TerraFlow helps customers improve resilience, reduce operating costs, accelerate deployment, and strengthen the electric grid.
DG Matrix, the global leader in solid-state transformer solutions, is unlocking power as the binding constraint on AI. It is removing this critical bottleneck of stranded power by building the intelligent power fabric of the AI data center. Built on the world's first commercially available, programmable multi-port solid-state transformer, the company's AI-enabled Interport™ platform converts, routes, and controls every electron, between the grid, on-site generation, storage, and compute loads through a single, software-defined stage. DG Matrix delivers an integrated and simpler solution for faster deployment, lower energy costs, native 800 VDC readiness, and protection from GPU pulse loads. Because speed to power is speed to inference, DG Matrix helps hyperscale, neocloud, and colocation operators bring AI capacity online in months instead of years, at scale, anywhere in the world.