090°Open data
Open datasets, fully documented — searchable here, and readable by any LLM.
1 datasets
US Grid Disruption Intelligence (DOE OE-417)
Monthly grid-disruption panel for the 50 US states, DC and Puerto Rico plus a US national aggregate (2002-01 .. 2023-12), built from the Department of Energy's Form OE-417 major electric disturbance event reports (3,874 events). Each (month, state) row carries event counts by cause category (severe weather, physical attack/vandalism, cyber, equipment failure, fuel supply, system operations, wildfire, other), customers affected, demand loss in MW, fatalities/injuries, and average event duration in hours, plus derived features: a 0-100 severity score (each geography scored against its own history), a 12-month pressure ratio of customers affected, a cross-state event heat rank, and the dominant cause. Units are counts, customers, and MW per calendar month. Caveat: OE-417 covers only major disturbances meeting DOE reporting thresholds (e.g. 50,000+ customers for 1+ hour), so small outages are invisible; reporting completeness improved after 2011; annual summaries lag, so 2023 is the latest year. A coincident disruption signal for demand, churn, and lead-scoring models — not a forecast.
Once today's free allowance is used up, AI features can run on your own provider account.
Kept in this browser tab only (cleared when you close it) and sent with each AI request. Our servers use it for that request and never store or log it.