The Need of Two-Phase CFD for Nuclear Reactor Safety

The first part of the paper was devoted to the European Commission EUROFASTNET Project (European Project for Future Advances in Sciences and Technology for Nuclear Engineering Thermal-Hydraulics), which regrouped some fifteen organisations in eight countries. The objective of the project was to identify a set of R and D actions to develop better methods and tools in two-phase thermal-hydraulics, in particular to develop new computer models, and new experiments for a more detailed and better qualified description of the physics. The proposals for R and D actions would contribute to ensuring safety of present and future reactors, extending the life span of the current generation of reactors, contributing to the design of the next generation of reactors. Dr Bestion described briefly the programmes of work of three Working Parties of the Project (Review of Industrial Needs, State of the Art, Proposals for Future R and D Work) and elaborated on the following aspects of these programmes: review of industrial needs, prioritisation of needs, analysis of the needs. Dr Bestion then presented the CEA/EDF R and D Programme on Two-Phase Thermal-Hydraulics (NEPTUNE Project), which had the following objectives: - improvement of system codes and component codes, - development of a two-phase CFD type code, - coupling between scales and between thermal-hydraulics and neutronics, mechanics, and fuel, - development of advanced instrumentation techniques, - new experimental programmes, - improvement of numerical schemes. He stressed that it is important to distinguish two types of two-phase 3-D codes: 3-D codes for a porous medium and 3-D codes for an open medium (CFD type codes), and then described current work in the area of two-phase flow modelling (R and D about physical modelling, investigations in progress for 1-D and 3-D in a porous medium and for a 3-D CFD type model in an open medium). He then discussed the advantages and drawbacks for the nuclear community of using commercial CFD codes; his conclusion was that the drawbacks outweigh the advantages, and recommended the development of a CFD code within the nuclear community. Dr Bestion then described investigations in progress in the area of two-phase DNS techniques, the capabilities and limitations of the computation of present codes, and areas for possible improvement. Finally he presented the results of some calculations