Forecasting space weather risks on power grids
Microsoft’s prototype forecasts substation-level geomagnetic risk 30 to 60 minutes ahead across 66,935 U.S. sites.
The pipeline combines solar-wind data, AE and Dst forecasts, latitude, geology, conductivity, and grid data to estimate where geomagnetically induced current risk may rise. In evaluation, it detected 76.5% of major events, 81.2% of severe events, and 64.1% of extreme events, with more false alarms at higher severity. Microsoft says the model can generate continental-scale estimates in about 333 milliseconds, but further validation with utilities and operational data is needed before grid use. Microsoft Research AI's note
The pipeline combines solar-wind data, AE and Dst forecasts, latitude, geology, conductivity, and grid data to estimate where geomagnetically induced current risk may rise. In evaluation, it detected 76.5% of major events, 81.2% of severe events, and 64.1% of extreme events, with more false alarms at higher severity. Microsoft says the model can generate continental-scale estimates in about 333 milliseconds, but further validation with utilities and operational data is needed before grid use. Microsoft Research AI's note
score 4