LES–DPS of the effect of wall roughness on dispersed-phase transport in particle-laden turbulent channel flow

2006 ◽  
Vol 27 (4) ◽  
pp. 619-626 ◽  
Author(s):  
Kyle D. Squires ◽  
Olivier Simonin
Author(s):  
Neng-Tsung Chang ◽  
Chih-Hung Hsu ◽  
Keh-Chin Chang

Particle-laden turbulent channel flow at Reτ = 644, loaded with binary mixture of particles, is numerically studied using the Lagrangian particle tracking method coupled with large eddy simulation. Turbulence statistics of different particle groups are analyzed. Two particle-wall models are applied to this study with / without considering wall roughness. Taking into considerations of rough wall model, the effect of wall roughness in the computations strengthens the wall-normal particle velocity fluctuations. As a result, particles tend to move from the near-wall region to the central core region. It leads to decrement of particle accumulation in the near-wall region as compared to the case considering the smooth wall model. The wall-normal particle mixing capability is enhanced which results in the redistribution of particles in the channel. The behavior of particle motion in the turbulent channel flow should be, thus, dependent on not only the value of Stokes number but also the wall roughness level.


2004 ◽  
Vol 127 (2) ◽  
pp. 250 ◽  
Author(s):  
Michael Benson ◽  
Tomohiko Tanaka ◽  
John K. Eaton

Author(s):  
Marion W. Vance ◽  
Kazuyasu Sugiyama ◽  
Shu Takagi ◽  
Kyle D. Squires

Microbubble transport in fully developed turbulent channel flow is investigated using an Eulerian-Lagrangian approach. The carrier-phase flow is computed using Direct Numerical Simulation (DNS) or Large Eddy Simulation (LES) of the incompressible Navier-Stokes equations. Lagrangian particle tracking is employed for a dispersed phase comprised of small, rigid spheres of negligible density compared to the carrier-phase flow and obeying an equation of motion in which the forces used to predict the motion of the bubble are drag, pressure gradient, and added mass. In general, DNS and LES yield similar predictions of the carrier phase flow and dispersed-phase properties. The bubble Stokes number is varied over a range for which the dispersed phase essentially follows the carrier flow to larger values for which strong segregation of the microbubbles into coherent vortical structures occurs. In general, simulation results show that microbubble response is not a monotonic function of the Stokes number. The most significant structure in the concentration field occurs for Stokes numbers close to the turbulence timescales in the buffer layer. More than 2/3 of the microbubble population in the buffer layer resides in coherent structures that occupy approximately 1/3 of the computational volume.


Sign in / Sign up

Export Citation Format

Share Document