JuMP powers foundational models for the electric grid
open-energy-modeling ·Author: Oscar Dowson
Microsoft Research recently published GridSFM, a small foundation model for the electric grid. We were pleased to notice that the training pipeline is powered by JuMP and PowerModels.jl.
To generate the training data for GridSFM, the Microsoft team solved hundreds of thousands of AC optimal power flow (AC-OPF) problems across 150+ real grid topologies. They did this using PowerModels.jl, a Julia package for power network optimization built on JuMP, with Ipopt as the underlying solver. Their code is open source on GitHub at microsoft/GridSFM.
A related project is the GridFM DataKit, which similarly uses JuMP and PowerModels.jl to generate training data for grid foundation models. The team behind it have published an arXiv paper describing their work. An interesting aspect of GridFM DataKit is that the front-end is a Python package, but it calls Julia and PowerModels via juliacall.
It’s great to see JuMP and the broader Julia ecosystem being used at this scale.
You can learn more about foundation models of the electric grid by watching the recordings of the 5th Workshop on Foundation Models of the Electric Grid.
If your work uses JuMP, we’d love to hear about it. The best way to tell us is by opening a GitHub issue.