01 What happened
A Microsoft Research intern developed a machine-learning system to estimate how space weather could affect power grids across the continental United States.
02 Key details
- The system produces location-specific risk estimates for 66,935 substations; measured inference takes approximately 333 milliseconds.
- In experiments, it produced risk estimates 30–60 minutes before potential impacts. Further validation with utilities and operational data is needed before grid use.
- For geomagnetically induced current risk, detection rates were 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).
03 Why it matters
After further validation with utilities and operational data, the estimates could help operators prioritize engineering reviews and consider targeted protective actions.
04 Who it matters to
Energy specialists, infrastructure engineers and artificial intelligence researchers.
Original source