摘要

The performance of the turbulent flow model for predicting the buoyancy effect on convective heat transfer of supercritical fluid is severely affected by strongly varying thermal physical properties near the pseudo-critical point. Over-prediction is attributed, at least partly, to the misuse of the constantly turbulent Prandtl number for the turbulent heat flux in the turbulence model. A method that considers the anisotropic turbulent heat flux has been proposed to improve the prediction accuracy of numerical simulation. A buoyancy effect model that accounts for the production of turbulent kinetic energy and a turbulent Prandtl number model accounting for turbulent thermal diffusion, which are both based on the anisotropic turbulent heat flux model, was adopted in the original AKN k-epsilon model. Experimental results and direct numerical simulations (DNS) data were used to validate the performance of the "Modified model." The "Modified model" produced accurate predictions for all heat transfer deterioration cases examined in the present paper. The buoyancy effect model reflects the basic mechanism of heat transfer deterioration and recovery due to accurate predictions of turbulent kinetic energy. The value of Pr-t in the buffer layer obtained with the turbulent Prandtl number model is essential for accurate reproductions of experimental data.