Solar energy is subject to considerable temporal and spatial fluctuations. With the increasing share of solar energy in our energy supply system, the requirements for solar forecasts are also increasing. These forecasts form the basis for the cost-optimized use of various energy management options such as storage, control energy and controllable loads - all with the aim of integrating solar energy into our energy system at low cost. Today, PV forecasting plays an important role for trading with PV electricity and scheduling of power plants.
At Fraunhofer ISE, we are working on the development of forecasting models for the reliable prediction of the solar power for time scales from a few minutes up to several days and areas from single locations to regions of many kilometers. Different models and input data such as cloud camera, satellite-based and numerical weather predictions are integrated.
The focus of our research activities is the optimization of the forecasts for different applications in close cooperation with the users, as well as the development of probabilistic forecasts for the reliable description of forecast uncertainties and the reduction of the associated costs.