The main goal with this project is to develop new day-ahead statistical forecasting models to assess the uncertainty associated with wind power generation. These models are based on meteorological ensembles (multiple meteorological forecasts) generated by the global physical model of the European Centre for Medium Range Weather Forecasts (ECMF).

The researchers involved in the project will develop a new statistical technique that combines quantile linear regression (regression models) in the Reproducing Kernel Hilbert Space (RKHS) with parallel computing and real-time learning techniques. The end result will be a probabilistic forecasting model that will be operated and validated for two wind parks in Romania.

Besides CPES, the UPM - Universidad Politécnica de Madrid will also participate in this project. The INESC TEC team involved in this project is composed of Ricardo Bessa (project leader), Laura Cavalcante, Carla Gonçalves. Cristóbal Gallego will be representing UPM.

The INESC TEC researchers mentioned in this news piece are solely associated with INESC TEC.

INESC TEC, June 2016