Product · Compound Vision
By Compounding Energy
Wind and solar power forecasting. 7,979 farms and expanding globally. Physics, not black boxes.
Compound Vision produces hourly power forecasts for every public wind and solar farm in CONUS — plus any custom assets you add. The engine is pure physics: Monin-Obukhov atmospheric stability, manufacturer power curves, Perez irradiance transposition, and 30+ documented steps from weather to watts. No ML. No hidden assumptions. Probabilistic P10–P90 bands that widen honestly when data is uncertain.
Wind (1,667) and solar (6,312) farms forecasted individually from USGS databases
Forecast horizon via Open-Meteo (GFS, ECMWF, ICON, GEM)
Probabilistic bands with inference-aware uncertainty widening
Manufacturer power curves for real turbine models
How it works
Weather ingest
Compound Vision pulls hourly NWP data from Open-Meteo — GFS, ECMWF HRES, ICON, and GEM models. Each farm gets weather interpolated to its exact coordinates. A paid Open-Meteo key cuts full-CONUS runtime from hours to under 10 minutes.
Physics-based power conversion
Wind: stability-dependent log-law shear (Monin-Obukhov with Louis 1979 closure), humidity-corrected air density, 35+ manufacturer power curves, Bastankhah-Gaussian wake losses, composite curtailment (icing, thermal, high-wind hysteresis). Solar: PSA solar position, Perez 1990 transposition, Fresnel IAM, Faiman thermal model, spectral correction, HSU soiling with rain-wash, bifacial rear irradiance, inverter clipping.
Probabilistic output
Every forecast includes P10, P25, P50, P75, and P90 bands. When a farm's specifications are inferred from vintage defaults (rather than known from the manufacturer), the bands widen by up to 1.4x — so the uncertainty is honest about what it doesn't know. Results aggregate by ISO region and are viewable on a 3D CesiumJS globe or via the REST API.
Why physics, not ML
ML models learn patterns from historical data. When the pattern changes — a new turbine model, a revised power curve, a farm expansion — the model needs retraining and the forecast degrades silently until it catches up.
Physics models derive power from first principles: wind speed, air density, atmospheric stability, panel temperature, irradiance geometry. When a farm changes, you update its specs and the forecast adapts immediately. No retraining. No drift. No opaque failure mode.
Compound Vision uses 30+ documented physical steps from raw NWP weather to final power output. Every step is individually testable and physically interpretable. When a forecast is wrong, you can trace exactly why — which atmospheric layer, which power-curve segment, which curtailment rule — and fix it at the source.
Capabilities
3D CesiumJS dashboard
Interactive globe with CartoDB basemap, wind particle animation, solar glow, time scrubber with playback controls, and dynamic marker sizing by capacity, capacity factor, or live power output. Viewport culling keeps it smooth at 2,000+ markers.
Customer farm extensibility
Add your own wind and solar assets via the API or the dashboard's 'Add farm' modal. Compound Vision infers missing specs from vintage-weighted defaults and reports a data-quality badge so you know what's measured vs estimated.
Multi-tenant API
FastAPI backend with JWT + API key auth, per-tenant farm scoping, paginated endpoints for farms and forecasts, and a public tenant that every free-tier user can read from. Rate-limited via Redis in production.
EIA Form 860 + USGS refresh
Monthly automated refresh from upstream USGS USWTDB and USPVDB releases with diff-and-upsert. EIA Form 860 bulk ingest covers assets that USGS misses. Both paths merge before forecasting — one unified physics run per cycle.
Compound Vision + GridSight
GridSight forecasts European electricity prices, demand, and dispatch. Compound Vision forecasts US wind and solar generation from the physics up. Together they represent Compounding Energy’s approach: transparent, auditable forecasting grounded in first principles, covering both sides of the Atlantic.