A multi-physics computational tool based on CFD and GEM chemical equilibrium solver for modeling coolant chemistry in nuclear reactors

    Research outputpeer-review

    Abstract

    The development of integrated numerical codes based on Computational Fluid Dynamics (CFD) and high-fidelity chemical reaction modules, offers the opportunity of an accurate simulation of turbulent reactive transport in nuclear reactors. This is invaluable in the optimization of coolant chemistry for the safe operation of nuclear reactors. The aim of this article is to develop a multi-physics simulation tool, where the chemical equilibrium solver and database of the commercial HSC-Chemistry software are integrated with the widely used CFD code ANSYS FLUENT. The assumption of local chemical equilibrium is used for a coupled simulation. The coupling procedure between codes is performed by a sequential operator-splitting approach. In this framework, the HSC equilibrium solver is called sequentially from each cell of the FLUENT computational domain whose composition has changed from equilibrium. Both iterative and non-iterative time marching algorithms are implemented and can be selected, depending on the problem under investigation. Communication between the codes is handled by a custom-designed object-oriented coupler class. A compiled-linked-in integration between FLUENT, the coupler class and HSC is designed, in order to minimize CPU time by exchanging data directly in the main memory. The coupled simulator is verified and preliminary validated with examples of chemical systems relevant for Gen. IV nuclear reactors cooled by lead-bismuth eutectic (LBE). The tool shows good performances in terms of numerical accuracy and computational speed. Its complete functionality is then illustrated with an application related to localized-corrosion modeling in flowing LBE. This application showed that the integrated code based on FLUENT and HSC is a promising tool for a wide range of applications involving reactive transport simulations.
    Original languageEnglish
    Article number103190
    Pages (from-to)1-14
    Number of pages14
    JournalProgress in Nuclear Energy
    Volume120
    DOIs
    StatePublished - 1 Feb 2020

    Cite this