DOE Funds AI Tools for Faster Grid Planning
Pacific Northwest National Laboratory will lead two projects using artificial intelligence to speed up electric grid planning models.
Updated on Nov. 5, 2026 in Energy
The Department of Energy awarded Genesis Mission funding to Pacific Northwest National Laboratory for two grid planning projects. The projects aim to use artificial intelligence to make complex electric grid models faster to run and easier to review.
Why it matters
Grid models are becoming so large and computationally demanding that they produce more scenarios than human teams can efficiently review. Faster models could help planners study future grid pathways across the United States more quickly.
Capacity expansion models can produce ten thousand different scenarios, while electromagnetic transient models calculate grid physics in microsecond resolution and are computationally heavy. The funding amount and performance benchmarks have not been disclosed, and the project teams have six months to demonstrate their research goals.
The players
Pacific Northwest National Laboratory
A Department of Energy national laboratory in Richland, Washington, that conducts research in energy, national security, and science.
Arizona State University
A public research university in Tempe, Arizona, that leads the second grid planning project in this award.
Marcelo Elizondo
A PNNL researcher who leads the first project using machine learning to approximate electromagnetic transient grid models.
The details
A machine learning algorithm will approximate complex electromagnetic transient calculations in the first project, led by PNNL's Marcelo Elizondo with partners at Argonne, Sandia, and Washington State University. The second project, led by Arizona State University with PNNL participation through Casey Burleyson, plans to build an AI model that selects significant scenarios from capacity expansion model output.
Timeline
Next six months: project teams are expected to demonstrate their research goals.
Big Picture
The awards extend the Department of Energy's Genesis Mission into electric grid planning, where models have grown too large for human teams to review scenario by scenario. The projects reflect a federal push to apply AI to the long-term buildout needed as demand on the grid roughly doubles.
If these tools succeed, grid planners could study far more future scenarios in less time, which may shape the infrastructure decisions behind future electricity reliability. The projects are still in early research, so no immediate change to household power service is described in the stub.
The takeaway
Grid planning is becoming a data problem as much as an engineering one, and AI tools that condense huge model outputs could help utilities and regulators make faster, better-informed decisions. Readers should watch whether these early demonstrations produce results that hold up under comparison with traditional methods.
Further reading
For broader context on power systems and grid planning, see Energy coverage on Wisevoter.
Source note: This article includes information reported by HPCwire.
