Weather conditions have a significant impact on the production of renewable electricity sources, whether wind or solar. Atos and Météo-France launched last November a platform for forecasting renewable electricity production.
Based on meteorological data provided by Météo-France, production histories, as well as information on the physical equipment in place, this platform uses Machine Learning algorithms to determine the most accurate forecasts possible, and thus provide a so-called “weather-sensitivity” service.
Intended for electricity producers and aggregators and integrated into a smart grid approach, this platform will make it possible to manage electricity production and consumption more efficiently. It will make it possible to anticipate the system’s equilibrium conditions and, in particular, other production needs, and to integrate renewable energies as effectively as possible into the French and European electricity grid.
Atos will also provide two supercomputers to Météo-France, which will increase its computing power by more than fivefold. The total investment of the project is 144 million euros. The implementation of these two supercomputers will make it possible “to improve the prediction of dangerous phenomena with a gain of 1 to 2 hours on the forecasts, to improve geographical accuracy and therefore to better determine the risks, by going down to an infra-departmental scale, to take into account more observations and new types of observations such as connected objects”.