DOE Names Winners of Digitizing Utilities Prize Round 3

Estimated reading time: 5 minutes · Last updated: 1 September 2026

The U.S. Department of Energy’s Office of Electricity announced 13 Phase 1 winners of the Digitizing Utilities Prize Round 3 on August 31, 2026, with each team receiving $75,000 to develop data-driven tools that improve grid reliability and security, as first reported by the Department of Energy. Winners range from startups focused on load forecasting and outage management to university groups applying AI to distribution models. The prize is structured so selected teams move into Phase 2 demonstrations alongside utility partners and the National Laboratory of the Rockies, with the goal of turning prototype analytics into deployable tools utilities can adopt.

Teaming up to tackle complex grid challenges requires bridging the gap between raw data and practical decision-making,

Catherine Jereza, Assistant Secretary of the Office of Electricity

Who won and what each team is building

Thirteen teams advanced from Phase 1 to Phase 2 of the Digitizing Utilities Prize Round 3. The winners and their locations are AI Power (Princeton, NJ); Reactive Technologies (New York, NY); Analog Garage (Boston, MA); OpenDrawing/City of Elba (Costa Mesa, CA); PowerOutage.com (Orlando, FL); Gridient (Denver, CO); Theta Runaways (New York, NY); Moonshot (Spearfish, SD); GridSense@UH (Houston, TX); Team REC (Fredericksburg, VA); The Resilient Alabama Team (Birmingham, AL); Plentiful.ai (Portland, OR); and The UCF Power Team (Orlando, FL).

Their proposals target distinct operational gaps: AI Power focuses on improved load forecasting, Reactive Technologies on high-resolution edge sensing, Analog Garage on electrical model delivery, and OpenDrawing on digitizing paper maps. Other entries concentrate on outage integration, vegetation analytics, wildfire precursors in smart-meter harmonics, distribution GIS/OMS corrections, sensor-fused digital twins, and zero-shot AI decision support for disaster restoration.

How Phase 2 demonstrations will link teams to utilities and labs

Phase 2 is explicitly a demonstration and refinement stage: teams will work side-by-side with utility partners and the National Laboratory of the Rockies to test their tools on operational data sets and in operational workflows. The Office of Electricity describes this stage as a chance to refine interfaces, data pipelines and decision-support outputs so utility operators can assess reliability and security benefits under realistic conditions.

That structure is important because integration with utility systems and operational practices is frequently the main hurdle for analytics pilots. DOE’s approach pairs small-scale funding at Phase 1 with practical demonstrations at Phase 2 so winners can show measurable improvements — for example, faster fault identification, more accurate outage mapping, or validated CVR (conservation voltage reduction) sequences — before broader adoption is pursued.

Technical focus areas and what they aim to deliver

The competitors fall into several technical clusters: forecasting and resource alignment (AI Power); edge sensing and transient analysis (Reactive Technologies, Theta Runaways); model generation and validation (Analog Garage, GridSense@UH); outage integration and hazard fusion (PowerOutage.com); and vegetation or wildfire analytics (Team REC, The Resilient Alabama Team, Moonshot). Plentiful.ai proposes a sensor-fusable digital twin layer that maps every building into a single operational context, while UCF’s platform targets real-time disaster restoration decision support.

Those technical choices map to concrete operational gains utilities seek. Better forecasting reduces the need for expensive reserves; verified CVR sequences can yield the 1–3% savings Gridient cites; automated GIS and OMS corrections shrink manual model maintenance time; and harmonics-based wildfire precursor detection aims to use data already collected by smart meters rather than deploying new sensors.

What this means for utilities and the path to adoption

For utilities, the practical test is whether a tool reduces outage minutes, lowers operating cost, or eases planning workloads. DOE’s prize funds and the Phase 2 lab-and-utility demonstrations are designed to produce those validation data points so utilities can make procurement decisions with performance evidence rather than pilots alone. If a tool demonstrably shortens fault-isolation times or reduces field crew dispatches, procurement becomes a clearer business case.

There are still hurdles: utilities must permit access to operational data, vendors must adapt tools to diverse system models, and measurable grid benefits must survive seasonality and extreme events. The prize narrows those gaps by emphasizing demonstrations, but wider adoption will depend on follow-on funding, standards for data exchange, and vendor support for integration across different utility IT/OT environments.

How the prize could play out

The bottom line

The Digitizing Utilities Prize Round 3 channels modest Phase 1 awards into operational demonstrations intended to shorten the path from prototype analytics to utility deployment. With 13 teams backed by $75,000 apiece and a Phase 2 plan that pairs innovators with utility operators and the National Laboratory of the Rockies, the program is structured to produce measurable operational metrics. Whether those metrics translate into broad procurement will depend on demonstration results, follow-on funding, and utilities’ willingness to share data and adapt procurement processes. The prize narrows early-stage barriers but does not replace the system-level work needed for large-scale adoption.

What to watch

• Watch for the DOE or National Laboratory of the Rockies to publish the Phase 2 demonstration schedule; no date has been set.
• Watch for public demonstration reports that quantify outage-minute reductions or CVR savings from winning teams; no release date has been set.
• Watch for announcements of follow-on funding or procurement commitments from utility partners; no date has been set.