Forecast monitoring and optimization
Ensuring high prediction quality for large numbers of models
With increasing decentralization and the expansion of renewable energy sources, reliably good forecast models of consumption and generation have become a critical factor in the energy sector. However, it is increasingly challenging to keep track of an ever-growing number of forecast models – often several thousand – in terms of quality and timeliness. Even more so as models can quickly become outdated due to rapid changes of influences caused by effects such as the expansion of renewable energy. It is therefore necessary to monitor a large number of models with minimal effort in order to detect negative trends in forecast quality early on and identify the causes of this deterioration.