.avif)
Typical Meteorological Year
Download a statistically representative “typical year” file built from 15 + years of satellite data for long-term yield modelling.
Overview
Markets, Portfolios and Grid models
Probabilistic MW-scale forecasts engineered for grid-wide accuracy
_Frame.png)
Complete, end-to-end generation insight, for any asset
Market & Portfolio Models fuse live SCADA, 1–2 km satellite irradiance and NWP ensembles to generate kW/MW time-series at 5- to 60-minute resolution complete with P10/P50/P90 uncertainty bands out to 14 days. Traders, ISOs and utilities rely on these physically informed machine-learning models for optimal dispatch, hedging and DER planning.
Grids & Markets
Behind-the-meter & utility-scale aggregation
Turn hundreds of thousands of rooftop and utility PV systems into a single forecast for any ISO zone, feeder or VPP. Solcast Grid Aggregations blend 1–2 km satellite imagery with advanced PV models to create real-time and 5- to 60-minute outlooks, refreshed as often as every five minutes.
Probabilistic P10/P50/P90 outputs and custom geographic boundaries give operators the visibility to sharpen load forecasts, flag minimum-demand events and cut congestion risk.
Portfolios
Probabilistic fleet forecasts for solar & wind
Our model already supports 2,000+ sites (150 GW) across 20 countries, producing accurate power for individual projects or entire portfolios. Machine-learning and physical models fuse live SCADA feedback with global weather ensembles, delivering deterministic and probabilistic outputs every 5–60 minutes for horizons up to 15 days.
By turning weather uncertainty into probabilistic MW curves, you can bid nearer to true output without breaching risk limits
Load Forecasting
Demand curves you can trade on
DNV’s demand models integrate mesoscale weather forecasts, advanced machine-learning algorithms and power-systems engineering principles to predict load from ISO scale down to individual meters. Outputs refresh as frequently as every five minutes and include calibrated P10/P50/P90 confidence intervals.
Data available via API, SFTP or email, giving DERMS operators, planners and traders the precision needed for coincident-peak management, price-responsive dispatch and risk-aware market bids.
Ready to validate the models on your data?
Leave your details and one of our team will be in touch.
Oops! Something went wrong while submitting the form. Please refresh the page and try again.
Dive straight into our API Toolkit, access the models or test our Historical Time Series, Live, and Forecast data
Frequently Asked Questions
How far ahead do the forecasts extend?
Our live-plus-forecast feeds cover up to 14–15 days ahead, with up to seven days of recent history, depending on the service tier.
How often are the model outputs refreshed?
Update intervals are configurable down to 5 minutes; Solcast grid aggregations refresh every 5–15 minutes, while Forecaster offers frequencies from daily to 5-minute resolution.
What geographic coverage is available?
The aggregation models generate data for virtually any land area worldwide (global, except ocean and polar regions) using 1–2 km satellite imagery and numerical weather-model inputs.
Which variables and uncertainty metrics can I access?
Outputs include aggregated PV power in MW with P10, P50 and P90 probability bands; Forecaster also supplies probabilistic power forecasts for wind, solar or demand.
How are the forecasts delivered?
Data streams via REST API, SFTP or email and are available in JSON or CSV through the Solcast Web Toolkit.
Can I evaluate the models before subscribing
Yes. Solcast lets new users make up to ten free live and forecast API calls, and the Green Data Products team will provide a demo or hindcast file on request.
