New tools for generation IV assemblies modelling

Full text of publication follows: In the framework of the development of generation IV concepts, the need of new assembly modelling tools arises. These concepts present more geometrical and spectral heterogeneities (radially and axially). Moreover thermal-hydraulics and neutronics aspects are so closely related that coupled computations are necessary. That raises the need for more precise and flexible tools presenting 3D features. The 3D-coupling of the thermal-hydraulic code FLICA4 with the Monte-Carlo neutronics code TRIPOLI4 was developed in that frame. This new tool enables for the first time to obtain realistic axial and radial power profiles with real feedback effects in an assembly where thermal-hydraulics and neutronics effects are closely related. The BWR is the existing concept presenting the closest heterogeneous characteristics to the various new proposed concepts. This assembly design is thus chosen to compare this new tool, presenting real 3D characteristics, to the existing ones. For design studies, the evaluation of the assembly behavior, currently necessitate a depletion scheme using a 3D thermal-hydraulics assembly calculation coupled with a 1D axial neutronics deterministic calculation (or an axial power profile chosen as a function of the assembly averaged burn-up). The 3D neutronics code (CRONOS2) uses neutronic data built by 2D deterministic assembly calculations without feedback. These cross section libraries enable to take feedbacks into account via parameters such as fuel temperature, moderator density and temperature (history parameters such as void and control rod are not useful in design evaluation). Recently, the libraries build-up has been replaced by on line multi-2D deterministic assembly calculations performed by a cell code (APOLLO2). That avoids interpolation between pre-determined parameters in the cross-section data used by the 1D axial neutronics calculation and enable to give a radial power map to the 3D thermal-hydraulics calculation. The problem is that assembly 2D deterministic schemes are fitted to known concept behavior (self-shielding, cells gathering) in order to obtain realistic computation times. The use of a Monte-Carlo code with point wise cross sections enables to model any kind of concept without previous fitting. This new 3D neutronics/thermal-hydraulics approach is thus perfectly adapted to the evaluation of new concepts. This kind of assembly modelling can however only be used as a reference. The computation time is still too high to envisage a depletion calculation during a whole cycle. It can however be used as reference, in a second phase, to establish new fully deterministic schemes. (authors)