scholarly journals Direct Numerical Simulation of natural, mixed and forced convection in liquid metals: selected results

2022 ◽  
Vol 389 ◽  
pp. 111597
Author(s):  
Andrea Fregni ◽  
Diego Angeli ◽  
Andrea Cimarelli ◽  
Enrico Stalio
Author(s):  
Haomin Yuan ◽  
Elia Merzari

The flow characteristic of fluid at low Prandtl number is of continued interest in the nuclear industry because liquid metals are to be used in the next-generation nuclear power reactors. In this work we performed direct numerical simulation (DNS) for turbulent channel flow with fluid of low Prandtl number. The Prandtl number was set to 0.025, which is representative of the behavior of liquid metals. Constant heat flux was imposed on the walls to study heat transfer behavior, with different boundary conditions for temperature fluctuation. The bulk Reynolds number was set as high as 50,000, with a corresponding friction Reynolds number of 1,200, which is closer to the situation in a reactor or a heat exchanger than used in normally available databases. Budgets for turbulent variables were computed and compared with predictions from several RANS turbulence models. In particular, the Algebraic Heat Flux Model (AHFM) has been the focus of this comparison with DNS data. The comparisons highlight some shortcomings of AHFM along with potential improvements.


Author(s):  
Xu Chu ◽  
Eckart Laurien

For fluids at supercritical pressure, the phase change from liquid to gas does not exist. Meanwhile, the fluid properties change drastically in a narrow temperature range. With supercritical fluid as working fluid in a heated pipe, heat-transfer deterioration and recovery have been observed, which corresponds to the turbulent flow relaminarization and recovery. Direct numerical simulation (DNS) of supercritical carbon dioxide flow in a heated vertical circular pipe is developed with the open-source code OpenFOAM in this study. Forced-convection and mixed-convection cases including upward and downward flow have been considered in the simulation. In the forced convection, flow turbulence is attenuated due to acceleration from thermal expansion, which leads to a peak of the wall temperature. However, buoyancy shows a stronger impact on the flow. In the upward flow, the average streamwise velocity distribution turns into an M-shaped profile because of the external effect of buoyancy. Besides that, negative buoyancy production caused by the density variation reduces the Reynolds shear stress to almost zero, which means that the flow is relaminarized. Further downstream, turbulence is recovered. This behavior of flow turbulence is confirmed by visualization of turbulent streaks and vortex structures.


Sign in / Sign up

Export Citation Format

Share Document