GridFM
PulseAugur coverage of GridFM — every cluster mentioning GridFM across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
AI model distilled for real-time edge power grid inference
Researchers have developed a method to distill a large grid foundation model (GridFM) into a significantly smaller version suitable for real-time edge inference. This process reduces the model's parameter count by 32.5 …
-
GENCO neural solver unifies grid analysis tasks, offering major speedups
Researchers have developed GENCO, a unified neural solver designed for steady-state analysis of electrical transmission grids. This new architecture can handle power flow, optimal power flow, and state estimation tasks …
-
New MxGPS model tackles topology overfitting in power grid AI
Researchers have developed MxGPS, a novel multiplex graph transformer designed to address topology overfitting in power grid foundation models. This new model architecture, which uses K task-specialized branches over a …