| In the simulation of turbulent fluid flow and heat transfer at supercritical pressures, substantial difficulties have been encountered in the modeling of turbulence and bounda-ry layer. This is due to significant fluid property variations with respect to the local temperature and pressure, especially in the near-wall region of a heated wall, where large temperature differences occur. The classical turbulence models available in literature were typically developed for constant-property fluids, where an empirical wall function in the high-Re k-epsilon model, and a damping function in the low-Re k-epsilon model were derived based on the constant-property data to solve the boundary layer. As it can be found in the existing literature, large differences have been observed between the experimental and numerical simulation results of the heat transfer coefficient predictions in the en-hanced and deteriorated heat transfer situations for supercritical fluids.;In this thesis, a novel near-wall treatment method is proposed to treat large property variations in the thermal and velocity sub-layers. In the near-wall region, the supercritical fluids can be considered thermal-conductive and viscous forces dominated. The thick-ness of the viscous sub-layer (VSL) and the conduction sub-layer (CSL) can be related to the wall shear stress and local Prandtl number information by using computational CFD models, such as that implemented in the NPHASE-CMFD code. The fluids' bulk and wall temperature information has been obtained from the literature review of experi-mental measurements. The wall temperature and heat transfer coefficient calculated from the k-epsilon model with the proposed wall treatment method have been found to be in good agreement with experimental data for both heat transfer enhancement and deterioration cases for two most widely used fluids: CO2 and water. The proposed model has been applied in the reactor-scale thermal-hydraulic analysis of different flow path designs in Gen-IV supercritical water nuclear reactors (SCWR). The main objective of this study has been to validate the performance of the current approach as a tool for the analysis of large-scale systems. |
