Researchers in Germany are building better forecasting models to improve the integration of renewable energy into the power grid.

Wind and solar power are sustainable, but they're not always reliable. They're dependent on the weather. Power grids must supply power when the power is needed; they're beholden to demand, not the weather.

Understanding when renewable sources will and won't be able to provide significant amounts of power is key to integrating wind and solar power into the power grid.

As part of the EWeLiNE project, researchers at the Fraunhofer Institute for Wind Energy and Energy System Technology have partnered with Germany's National Meteorological Service to develop math models that offer more precise wind and solar power output forecasts.

Based on improved weather forecasting, the models predict how much electricity Germany's photovoltaic and wind-farm facilities will produce in the coming hours and days.

"It's crucial for us to interconnect both worlds -- forecasts of weather and power -- more closely than before, tailoring them better to the requirements of the transmission system operators," project manager Malte Siefert, a researcher with IWES, said in a news release.

The researchers have been perfecting their models over the last four years and are now allowing power grid managers to test them live using a demonstration platform called EnergyForecaster.

Scientists at IWES and and the German Meteorological Service took the models currently used by power grid managers and improved them using updated weather forecasting models and the latest real-world data describing sustainable energy production.

"It's important to forecast how much renewable power will be generated, because that tells us how much conventional generation capacity -- whether nuclear, gas, or coal -- needs to be brought online," added Siefert. "At the same time, the forecast is necessary for calculations to keep the power grid stable and for trading electricity."