A machine learning pipeline developed at Microsoft Research generates location-specific geomagnetically induced current (GIC) risk estimates for 66,935 substations in the continental United States. The system combines solar-wind forecasts with local geological data to provide grid operators with 30 to 60 minutes of advance warning before specific risks appear.
- The pipeline uses a gradient-boosting model to estimate dB/dt, the rate of magnetic-field change associated with GIC risk.
- It achieved detection rates of 76.5% for major events (≥10 nT/min), 81.2% for severe events (≥20 nT/min), and 64.1% for extreme events (≥50 nT/min).
- The AE predictor recorded 410.2 nT RMSE, while the Dst predictor outperformed the Burton equation on 62.2% of peak-activity hours.
- Inference for all substations takes approximately 333 milliseconds, allowing rapid evaluation of multiple scenarios.
This approach enables utilities to prioritize engineering reviews and consider targeted protective actions, such as adjusting reactive-power reserves, based on precise geographic exposure.