A Case History of CFD Support to Accelerator Driven System Plant Design

The Integrated Project EUROTRANS, funded by the European Commission in the VI European framework program, was aimed at providing the advanced design of a multi purpose research oriented Accelerator Driven System (ADS), called eXperimenTal-ADS (XT-ADS), and the preliminary design of an industrial scale ADS, called European Facility for Industrial Transmutation (EFIT). One contribution of CRS4 (Centro di Ricerca, Sviluppo e Studi Superiori in Sardegna) has been to provide support to the overall plant design by means of Computational Fluid Dynamics (CFD) simulations. The simulations were required by the designer either for basic checking or in case of doubts on the validity of some technical options. We present four series of simulations which lead to the detection of unsatisfactory plant behaviour, related design modification and eventually control of the variant behaviour correctness. The first three simulation series deal with the EFIT design while the forth one deals with the XT-ADS design. In the first case, the simulation put in evidence a large recirculation zone under the reactive core that had to be removed for oxygen control concern. The recirculation zone is suppressed by modifying the shape of the core support grid. In the second case, we put in evidence a recirculation zone at the entrance of the pumping system above the core. This recirculation zone can lower the pump efficiency. The entrance shape was modified to eliminate the recirculation zone. In the third case, we check the behaviour of the passive Decay Heat Removal (DHR) heat exchanger. We show that while the primary coolant flow is globally organized as expected, some flow mixing limits the efficiency of the system. The system efficiency is restored by increasing its passive pumping strength. This is performed simply extending the Heat Exchanger shroud a half-meter in the bottom direction. In the last case, we investigate the capability of an external DHR system to withstand a long complete plant shutdown. The simulation encompasses about 6 hours of physical time, enough to understand the critical trends and infers that the DHR system may be not sufficient for its purpose. This result has suggested some modification to the design (i.e. surface treatment to improve metal wall emissivity) as well as to the accident management (i.e. restart primary pumps to eliminate fluid stratification). All these design improvements have been obtained in a reasonable amount of time thanks to the continuous collaboration and exchange of information between the CFD engineer and the designer.