Source terms modeling for spacer grids with mixing vanes for CFD simulations in nuclear reactors Academic Article uri icon

abstract

  • © 2015 Elsevier Ltd. In the present study, a numerical method was developed to simulate the presence of spacer grids with mixing vanes in nuclear reactors fuel assemblies. These mixing devices usually have a complex morphology that results in difficult mesh procedures and a high computational simulation cost. The presence of spacers and vanes was simulated using momentum sources to overcome the computational power problem from the perspective of a quarter of the full reactor core CFD (Computational Fluid Dynamic) simulations. Several approaches were tested using RANS models. The starting point is calculation with body-fitted mesh of one grid span of a fuel assembly with spacer grids featuring dimples, springs and mixing vanes. Velocities and Reynolds Stresses are extracted from these mesh, then converted to source terms. A new computational domain is created with a coarser mesh and without the presence of dimples, springs, and vanes, but source terms are added to the momentum and Reynolds Stress transport equations to force the solution as the detailed geometry computation. The forcing is imposed in only a few volumes of the domain where dimples, spring and vanes are located. The approach was numerically stable and relatively easy to implement in open source codes Code_Saturne (EDF) and commercial codes Star-ccm+ (Cd-Adapco). Best practice was defined and grid sensitivity analysis on different quantities was performed. The robustness of the numerical method was demonstrated. This new methodology creates an intermediate approach that provides higher spatial resolution than sub-channel codes and reduced computational cost compared to detailed CFD simulation. Development of the method was based on anisotropic second order closure models, due to the swirling flow generated by the mixing devices; nevertheless, this novel approach applies also to one and two equation turbulence models, without lack of generality.

author list (cited authors)

  • Capone, L., Benhamadouche, S., & Hassan, Y. A.

citation count

  • 13

publication date

  • March 2016