Élaboration d'une Méthode d'Évaluation du Potentiel Énergétique pour les Projets de Renouvellement deParcs Éoliens

For new wind farm investments, including repowering, two key metrics are crucial: the estimated annual energy production and its associated uncertainty. Investors seek low uncertainty, which hinges on accurately estimating the uncertainty itself. This creates two research goals: 1) reducing uncertainty through improved climate and wind farm modeling—current methods still range from 6% to 12% uncertainty, and 2) refining the reliability of uncertainty models with better and faster propagation techniques. State-of-the-art EYA methods are well-accepted in the industry, but reducing uncertainty offers significant financial advantages. Repowering offers a new opportunity for EYA since it benefits from extensive historical data from the existing site, including 10-minute energy production records. This thesis aims to incorporate this data into the EYA process while to improve both accuracy and reliability. Most studies overlook uncertainty or fail to quantify improvements. This thesis proposes a method that achieves reliable AEP estimates with better uncertainty predictions.