ライブラリ登録: Guest

ISSN Online: 2377-424X

ISBN Print: 978-1-56700-474-8

ISBN Online: 978-1-56700-473-1

International Heat Transfer Conference 16
August, 10-15, 2018, Beijing, China

A STUDY OF THE EFFECTS OF THE PORE SIZE ON TURBULENCE INTENSITY AND TURBULENCE LENGTH SCALE IN FORCED CONVECTION FLOW IN POROUS MEDIA

Get access (open in a dialog) DOI: 10.1615/IHTC16.cov.023328
pages 3351-3358

要約

We investigated the relation between the pore size and turbulence intensity for flows in porous media. Our goal was to address a paradox between turbulence generation by a single solid obstacle and turbulence suppression by multiple solid obstacles. In a clear fluid region, a single obstacle will act as a turbulence enhancer, with the enhancement being proportional to the obstacle size. Unlike around a single obstacle, turbulence in a porous medium is restricted by the surfaces of other obstacles surrounding it. This restriction is expected to be proportional to the distance between the surfaces of the two neighboring obstacles.

We used a representative elementary volume (REV) with 4×4 cylindrical obstacles to represent the infinite porous medium structure. The REV had periodic boundaries in the x, y and z-directions, and a specified mass flow rate in the x-direction. Changing the cylinder diameter, d, under a constant Reynolds number of approximately 5,000, we compared the macroscopic turbulence intensity, 11,macro, the location of maximum turbulence intensity, It,max, and the maximum turbulence length scale, lt,max, for each case. Although the driving force (the applied pressure gradient) acts only in the x-direction, we observed a mean velocity in the y-direction, and a noticeable change in the bulk flow direction for the d/s = 0.8 case, where s is the distance between centers of the obstacles. We found that It,macro increases in the range d/s = 0.1~0.4 with increased d, then slightly decreases in the range d/s = 0.6~0.8. We think this is caused by the turbulence suppression from cylinder walls. We also observed the bulk flow direction deviating from the direction of the applied pressure gradient, which could also contribute to the slight decrease of lt,macro in these cases. The location of It,max changes from near the separation point slightly behind each cylinder for d/s = 0.1~0.2, to the location where the wake comes into contact in the front of each cylinder for d/s = 0.4~0.8. This suggests that the turbulence generated by a cylinder is being suppressed by the surrounding cylinders. The maximum turbulence length scale decreases with increased d throughout the range of d/s = 0.1~0.8. There is a large decrease in the range d/s = 0.1~0.2, which we believe is the result of turbulence structures generated from each cylinder starting to interact with the surrounding cylinders.

Although further confirmation of these results is required, this study provides an estimate on how the pore size may affect the turbulent flow in porous media with large solid obstacles.