Study of the design choices scalability to higher power level for an astrid-like sodium cooled fast reactor

In the French ASTRID project framework, CEA, AREVA and EDF are currently working together to design an innovative Sodium Fast Reactor. One of the objective of this project is to build a robust safety demonstration for this reactor. Since 2010 the reference concept for ASTRID (600MWe) is the CFV, a low sodium void worth core which presents radial and axial heterogeneities. Some design optimizations are still led on this concept to enhance its performances. Naturally, these optimizations aim to improve the operational and safety performances of the ASTRID reactor. But, as a demonstrator, the ASTRID core design must also be scalable to higher power level, which means that the industrial design inheriting from the demonstrator conception choices must be optimal in terms of safety. Thus we have to ensure that each choice made to enhance the demonstrator operational performances is relevant when scaled to an industrial reactor. This article presents the study of the scalability of an ASTRID-like sodium fast reactor to higher power level (1000MWe) performed at EDF, using an in-house optimization process, called the SDDS method. Our goal is to determine, for each conception choice made to improve the safety, whether its relevance is related to the power level and thus, whether this conception choice is scalable. Two types of transients are considered to evaluate the safety level of a design, the Unprotected Loss of Supply Station Power (ULOSSP) and the Unprotected Control Rod Withdrawal (UCRW). It appears that, without any constraint, the design recommendations, to reach a favorable trade-off between ULOSSP and UCRW behaviors, are the same for both demonstrator and industrial power level (there is a set of optimal fuel pin geometry shared by the demonstrator and the industrial reactor). However, some additional safety considerations and technical constraints must be taken into account. Among others, an adequate number of sub-Assemblies is necessary for safety reasons, whatever the reactor size, while the sub-Assemblies must be large enough to ensure a minimal space between two control rods because of the size of control rod mechanism. Because of these considerations, the optimal design of the demonstrator slightly differs from the industrial one.