Êtes-vous un étudiant de l'EPFL à la recherche d'un projet de semestre?
Travaillez avec nous sur des projets en science des données et en visualisation, et déployez votre projet sous forme d'application sur GraphSearch.
The planning of national power systems is traditionally based on long-term forecasts of electricity demand and fuel prices. However, over such long time horizons (20-50 years), these forecasts often prove to be inaccurate. As an example, wrong projections of low natural gas prices and high electricity demand in Europe in the early 2000s stimulated massive investments in electricity generation, which led to the current situation of overcapacity in the European electricity market. In this work, we first present the issue of overcapacity in Europe and discuss its causes; in particular, we highlight the relationship between overcapacity and errors in forecasts of electricity demand and fuel prices. Then, we apply a novel robust optimization framework to a real-world strategic energy planning problem, and compare it to the standard, deterministic decision-making approach. The results show that the robust investment strategies are up to 58% less likely to require future modifications leading to overcapacity. Overall, this suggests that considering uncertainties in the long-term planning process can reduce the risk of generating overcapacity in national power systems.