Short-Term Forecast of Wind Turbine Production with Machine Learning Methods: Direct and Indirect Approach

The Energy Transition Act defined by the French State has precise implications on Renewable Energies, in particular on its remuneration mechanism. Until then, a purchase obligation contract permitted the sale of wind-generated electricity at a fixed rate. From now, it will be necessary to sell this electricity on the Market (at variable rates) before obtaining additional compensation intended to reduce the risk. This sale on the market requires to announce in advance (about 48 h) the production that will be delivered on the market, thus it is very important to predict this production. The objective of the project is to provide, every day, short-term forecasts (48 h horizon) of wind power production. We use two approaches: a direct one that predicts wind generation directly from weather data, and an indirect one that predicts wind from weather data and converts it into production. In order to forecast the production we use different machine learning algorithms and we propose features engineering to improve the forecasts. Our results are very conclusive compared to those in literature.