The minimal channel: a fast and direct method for characterising roughness

2nd Multiflow Summer School on Turbulence

Macdonald, Michael^1 ; Chung, Daniel^1 ; Hutchins, Nicholas^1 ; Chan, Leon^1 ; Ooi, Andrew^1 ; García-Mayoral, Ricardo^2

Department of Mechanical Engineering, University of Melbourne, VIC

3010, Australia^1

Department of Engineering, University of Cambridge, CB2 1PZ, United Kingdom^2

关键词: Computational time;    Hydraulic behaviour;    Improve performance;    Near-wall flows;    Near-wall region;    Number of Grids;    Rough surfaces;    Stream-wise velocities;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/708/1/012010/pdf
DOI  :  10.1088/1742-6596/708/1/012010

来源: IOP

【 摘 要 】

Roughness only alters the near-wall region of turbulent flow and leaves the outer-layer unaffected, making it a prime candidate for the minimal-span channel framework which only captures the near-wall flow. Recently, Chung et al. (J. Fluid Mech., vol. 773, 2015, pp. 418-431) showed that the minimal-span channel can accurately characterise the hydraulic behaviour of roughness. Following on from this, we aim to further optimise the minimal-span channel framework by primarily noting that the outer layer it produces is inherently incorrect, and as such modifications to this region can be made to improve performance. Firstly, a half-height channel with slip wall is shown to reproduce the near-wall behaviour seen in a standard channel, but with half the number of grid points. Next, a forcing model is introduced into the outer layer of a half-height channel. This reduces the high streamwise velocity associated with the minimal channel and allows for a larger computational time step. The streamwise length of the channel is also investigated independent of the previous improvements, and suggests the minimum length should be at least 3 times the spanwise width and also 1000 viscous-units long, whichever is longer. Finally, an investigation is conducted to see if varying the roughness Reynolds number with time is a feasible method for obtaining the full hydraulic behaviour of a rough surface, instead of running multiple simulations at fixed roughness Reynolds numbers.