Y. Goude (EDF R&D)

Join work with Pierre Gaillard, Raphaël Nédéllec, Gilles Stoltz Short-term electricity forecasting has been studied for years at EDF and different forecasting models were developed from various fields of statisticsor machine learning (functional data analysis, time series, non-parametric regression). We are interested in the forecasting of energy data at different scales (national electricity load, substations, heat demand) based on these different approaches. We investigate in different empirical studies how to use them to improve prediction accuracy. First, we show how combining members of the original set of forecasts can lead to a significant improvement. Second, we explore how to build various (possibly large numbers) and heterogeneous forecasts from these models and analyze how we can aggregate them to get even better predictions.