Long-term energy scenario modeling and dispatch simulations are two key stages in a methodology for cost-effective transition to a lowcarbon power system. Although these stages are equally important, they are often performed independently. This decoupled approach can lead to future investment trajectories decided by long-term energy models with no guarantee of generation adequacy. In this respect, TIMES-ANTARES is the result of linking a long-term energy model with a stochastic power system model. The aim of this study is to develop a general methodological framework using a multimodel approach to investigate optimal power mixes which meet generation adequacy requirements. The emphasis of this article is on the methodology, but we also include a case study of power generation planning applied to the French system for 2013-2050. The results show that using TIMES alone exposes the power generation mix for 2030 to a risk of insufficient supply. On the other hand, activating the iterative feed-back loops over capacity credit parameters has the potential to ensure both the economic effectiveness of the mix and the security of the electricity supply criterion set by the French public authorities.
Keywords TIMES, ANTARES, generation adequacy, linking-model approach