Some lessons learned from the use of two-phase CFD for nuclear reactor thermalhydraulics

Two-phase Computational Fluid Dynamics (2-phase CFD) is now increasingly applied to some Nuclear Reactor thermalhydraulic investigations. The main purpose of this paper is to give some guidance to two-phase CFD potential users, based on previous experience of these tools. This experience comes first from a Writing Group (WG3) of the OECD-CSNI-GAMA on the “extension of CFD to two-phase safety issues” which has identified a list of Nuclear Reactor Safety issues for which the use of 2-phase CFD can bring a real benefit and which proposed a general multi-step methodology. Then experience was gained with the NEPTUNE-CFD code which is being developed by EDF and CEA and sponsored by IRSN and AREVA-NP for nuclear application. Also the NURESIM Integrated Project of the European Commission 6th Framework Program has been using 2-phase CFD for Direct Contact Condensation, Pressurized Thermal Shock, boiling flow, DNB and Dry-Out investigations. Although some progress has been made on all these applications, several difficulties are encountered and are here identified. Based on this experience, this paper intends to update the state of the art, to review the main modeling difficulties and to make recommendations for future developments, validation and application of two-phase CFD. The choice of a basic model, of space and time resolution, the deterministic or statistical treatment of interfaces, the characterization of the interfacial structure, and the automatic recognition of the local flow structure are discussed. Transport of interfacial area, modeling of polydispersion and some closure problems for turbulent, wall, and interfacial transfers are also reviewed.