EAGER: Exploratory Research on DNS Modeling of Turbulent Heat Transfer in Porous Media

North Carolina State University, Raleigh NC

Investigators

Abstract

DNS Modeling of Turbulent Heat Transfer in Porous Media One of the most controversial topics in the field of convection in porous media is the issue of macroscopic turbulence, which is the known process of enhancing transport due to high speed flow. It remains unclear whether macroscopic turbulence can occur in a fluid saturated porous medium and, if so, what impact it has on convection heat transfer. It is extremely difficult to carry out velocity measurements within porous media to ascertain if turbulence exists on not in it, as they are typically optically opaque. At the same time, it is now possible to conduct high-definition direct numerical simulation (DNS) to study this phenomenon. In contrast to most existing work on turbulence in porous media, which relies on certain unverified turbulence models, DNS allows one to understand the phenomenon in all its complexities by directly resolving all the scales of transport. Hence, this project examines the processes that take place in porous media at large flow rates (high Reynolds numbers) via DNS, attempting to accurately describe them and analyze whether they can be labeled as true turbulence. The hypothesis here is that the size of the pores determines the maximum size of the turbulent eddies. If the size of turbulent eddies cannot exceed the size of the pores, then turbulent phenomena in porous media may differ from turbulence in clear fluids. Indeed, this size limitation must have an impact on the energy cascade, for in clear (of porous media) fluids the turbulent kinetic energy is predominantly contained within large eddies. The study entails carrying out extensive DNS studies for various 2D and 3D porous media geometries. Porous media with different pore sizes will be investigated to study the effects of solid walls bounding a porous medium filled channel. Turbulent structures that appear in a porous medium will be compared with those in a clear fluid; their effects on convective heat transfer will be analyzed. In particular, the kinetic energy cascade from larger to smaller eddies will be investigated, and the spectrum and size of energy-containing eddies analyzed to uncover the mechanisms of turbulence energy production, transfer, and dissipation in flow through porous media. By modeling temperature as a passive scalar and investigating instantaneous temperature distributions, this project aims at identifying connections between hydrodynamic and thermal processes in turbulent convection. Also, the effects of a large number of solid obstacles, constituting the solid porous matrix, on the amount of kinetic energy in turbulent flow will be unveiled and also how it affects the temperature field. The possibility of a second Reynolds number in a bi-disperse porous medium will be considered, which would manifest through the appearance of turbulent eddies corresponding to a larger pore size. Comparison of the DNS results with published experimental data will be performed to validate the numerical results.