Direct Numerical Simulation On Turbulent Convection Of Low Prandtl Number Fluid In Channel

Liquid metals are widely used as first-loop coolants in fourth-generation advanced nuclear energy systems.Liquid metals have a low Prandtl number and the heat transfer characteristics are different from conventional fluids,which requires an in-depth study of the convective heat transfer characteristics of liquid metals.The existing numerical studies have proved that the turbulent Prandtl number of liquid metals is not a constant,and it is difficult to simulate the convective heat transfer process of liquid metals accurately by the conventional RANS method.Direct numerical simulation is a direct approach to solve the control equations and does not use additional turbulence models.DNS has high numerical accuracy with sufficient fine grid and time scale.In this paper,OpenFOAM is used to carry out direct numerical simulation of turbulent heat transfer in liquid metal in a channel.The friction Reynolds numbers of 180 and 395 and the molecular Prandtl numbers of 1,0.71,0.25,0.125,0.05,0.025,0.01,and 0.005 are used for the present simulation.the accuracy of the numerical results is verified by comparing the velocity distribution,turbulence intensity,Reynolds stress transport term,and temperature distribution with the previous numerical results.The numerical results show that the proportion of the log-law region in the cross section decreases as the Prandtl number decreases,and the conventional log-law distribution does not fit well with the temperature distribution of the liquid metal.In this paper,an exponential function is used to calculate the dimensionless temperature distribution within the cross section,and the exponential is fitted as a function of the Prandtl number to obtain the temperature distribution equation for the Prandtl number in the range of 0.005 to 1.Combined with the velocity power law,the Nusselt number correlation is derived on the basis of the temperature distribution equation.The derived equations are in good agreement with the results of this paper and the DNS database.Compared with the formulas obtained by fitting based on experimental data,the derived formula in this paper has a wider range of application for the Prandtl number and are in better agreement with the DNS results in the low Prandtl number fluid region.