CFD investigation of three-dimensional flow phenomena in a JAEA 127-pin wire-wrapped fuel assembly

Abstract

This paper presents a RANS (Reynolds Averaged Navier-Stokes simulation) based CFD (Computational Fluid Dynamics) investigation of the three-dimensional flow phenomena in a JAEA (Japan Atomic Energy Agency) 127-pin wire-wrapped fuel assembly. Complicated and vortical flow phenomena in the wire-wrapped fuel bundles have been elucidated by a CFD analysis with a high resolution scheme and a shear stress transport (SST) turbulence model, and by a vortex structure identification technique based on the critical point theory. The CFD results show a good agreement with the JAEA experiment results with the 127-pin wire-wrapped fuel assembly, which was implemented using water for validating the pressure drop formulas in the JAEA sub-channel analysis code, ASFRE. The generation and demise of the vortex structures in the edge, corner, and interior sub-channels were periodically synchronized with the angular positions of the wire spacers. The axial and tangential flows in the edge and corner sub-channels are much stronger than those in the interior sub-channels. The vortex structures in the edge sub-channels have higher axial and tangential velocities than those in the corner sub-channels, interior sub-channels, and the wake region generated by wire spacers. The vortex structures in the edge subchannels are a type of longitudinal vortex, and have a larger scale than the other sub-channels. The vortex structures in the corner sub-channels have lower axial and tangential velocities than those in the edge and interior sub-channels. The turbulence intensity in the sub-channels is remarkably dominated by the magnitude of the axial velocity. The vortex structures in the edge, corner, and interior sub-channels have a higher turbulence intensity and lower vorticity than the small-scale wakes near the wire spacers. The driving forces on the wire in the X-, Y-, and Z-directions are not only dependent on the axial velocity, but are also significantly dependent on the relative angular position between the wire spacer and rod, and the relative position between the wire spacer and duct wall.