Make forecasts comparable
Forecast accuracy depends on when a prediction is made, how far ahead it looks and what it is compared against.
About GraphSight
Solar forecasting research, focused on market operations.
GraphSight develops solar production forecasts across assets, portfolios and European bidding zones. Our work connects graph-based machine learning with the requirements of short-term power markets.
From research to service
Technical depth, directed at an operating decision.
GraphSight uses graph-based machine learning to model how energy generation varies across locations, over time and under different operating conditions. The focus is on producing accurate, reliable forecasts that can be integrated into real-world energy systems.
How we work
Use complexity deliberately
The most complex model is useful only when the data and decision justify it.
Model operational reality
Weather drives solar potential; operations determine what reaches the market.
Earn trust through evaluation
Customers should be able to test GraphSight on the data and workflow that matter to them.
Bring us the forecasting problem behind your market position.
Tell us the assets, portfolio or zone you need to forecast. We will review the available data and propose a fair evaluation scope.