Microsoft Research develops ML system to forecast space weather risks for 66,935 US substations

Microsoft Research created an end-to-end machine learning pipeline that uses solar-wind information and forecasts of Auroral Electrojet (AE) and Disturbance Storm Time (Dst) indices, combined with local factors like latitude and geology, to generate location-specific space-weather risk estimates for 66,935 substations across the continental United States. The system detected nearly 80% of major space-weather events during the evaluation period and provides grid operators with a 30 to 60 minute advance warning before a specific risk appears.