0
Research Papers: Multiphase Flows

Flow Modulation by Finite-Size Neutrally Buoyant Particles in a Turbulent Channel Flow

[+] Author and Article Information
Lian-Ping Wang

Department of Mechanical Engineering,
University of Delaware,
Newark, DE 19716-3140
e-mail: lwang@udel.edu

Cheng Peng

Department of Mechanical Engineering,
University of Delaware,
Newark, DE 19716-3140
e-mail: cpengxpp@udel.edu

Zhaoli Guo

National Laboratory of Coal Combustion,
Huazhong University of Science and Technology,
Wuhan 430074, China
e-mail: zlguo@hust.edu.cn

Zhaosheng Yu

Department of Mechanics,
Zhejiang University,
Hangzhou 310027, China
e-mail: yuzhaosheng@zju.edu.cn

1Corresponding author.

Contributed by the Fluids Engineering Division of ASME for publication in the JOURNAL OF FLUIDS ENGINEERING. Manuscript received January 11, 2015; final manuscript received August 22, 2015; published online December 8, 2015. Assoc. Editor: E. E. Michaelides.

J. Fluids Eng 138(4), 041306 (Dec 08, 2015) (15 pages) Paper No: FE-15-1016; doi: 10.1115/1.4031691 History: Received January 11, 2015; Revised August 22, 2015

A fully mesoscopic, multiple-relaxation-time (MRT) lattice Boltzmann method (LBM) is developed to perform particle-resolved direct numerical simulation (DNS) of wall-bounded turbulent particle-laden flows. The fluid–solid particle interfaces are treated as sharp interfaces with no-slip and no-penetration conditions. The force and torque acting on a solid particle are computed by a local Galilean-invariant momentum exchange method. The first objective of the paper is to demonstrate that the approach yields accurate results for both single-phase and particle-laden turbulent channel flows, by comparing the LBM results to the published benchmark results and a full-macroscopic finite-difference direct-forcing (FDDF) approach. The second objective is to study turbulence modulations by finite-size solid particles in a turbulent channel flow and to demonstrate the effects of particle size. Neutrally buoyant particles with diameters 10% and 5% the channel width and a volume fraction of about 7% are considered. We found that the mean flow speed was reduced due to the presence of the solid particles, but the local phase-averaged flow dissipation was increased. The effects of finite particle size are reflected in the level and location of flow modulation, as well as in the volume fraction distribution and particle slip velocity near the wall.

Copyright © 2016 by ASME
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Fig. 1

Sketches of (a) the coordinate system used for the channel flow simulation and (b) the two-dimensional domain decomposition for message passing interface parallel implementation

Grahic Jump Location
Fig. 2

Turbulent Reynolds shear stress profiles in half of the channel. All quantities are normalized by u*2. The thin diagonal line denotes the total shear stress, so the difference between this straight line and the data represents the viscous shear stress due to the mean flow.

Grahic Jump Location
Fig. 3

Comparison of the mean flow velocity profiles of the simulated single-phase turbulent channel flows

Grahic Jump Location
Fig. 4

Comparison of the RMS velocity profiles of the simulated single-phase turbulent channel flows

Grahic Jump Location
Fig. 5

The time evolution of mean flow speed (averaged over y) in the LBM simulations. The two vertical dash lines indicate the stationary stage (32.2<t*<56.1) used to obtain average statistics for LBM–PLL. The two blue vertical lines indicate the stationary stage (31.9<t*<53.9) used to obtain average statistics for LBM–PLS. The three horizontal lines mark 15.74, 15.02, and 14.82, respectively.

Grahic Jump Location
Fig. 6

The mean velocity profiles from both the LBM and FDDF simulations with a domain size of 4H×2H×2H : (a) log–linear plot and (b) linear–linear plot

Grahic Jump Location
Fig. 7

The relative change in the mean velocity of the flow due to the presence of particles: (a) linear–linear plot and (b) log–linear plot

Grahic Jump Location
Fig. 8

The mean velocity profiles from the LBM simulations and fitting coefficients in the inertial sublayer by the logarithmic law. The two vertical lines mark the region where linear regression is performed to obtain a fit of the form U+=lny+/k+A. The two vertical lines mark the region (30<y+<130) used to perform the logarithmic fitting.

Grahic Jump Location
Fig. 9

The RMS velocity profiles in the flows simulated by LBM: (a) linear–linear plot and (b) log–linear plot

Grahic Jump Location
Fig. 10

The RMS velocity profiles in the flows simulated by the FDDF approach: (a) linear–linear plot and (b) log–linear plot

Grahic Jump Location
Fig. 11

The relative changes of RMS velocity fluctuations due to the presence of solid particles in the LBM simulations, relative to the single-phase flow simulated by the same method using a same domain size: (a) linear–linear plot and (b) log–linear plot

Grahic Jump Location
Fig. 12

The relative changes of RMS velocity fluctuations due to the presence of solid particles in the FDDF simulations, relative to the single-phase flow simulated by the same method using a same domain size: (a) linear–linear plot and (b) log–linear plot

Grahic Jump Location
Fig. 13

The average local volume fraction Ψp of the particulate phase as a function of y+: (a) linear–linear plot and (b) log–linear plot. The profile from Picano et al. [59] at ϕV=0.1 and ap/H=1/18=0.056 is also shown for comparison. The vertical lines mark the location of y = ap.

Grahic Jump Location
Fig. 14

Replot of Fig. 13 as a function of y/ap: (a) log–linear plot and (b) linear–linear plot. The profiles from Uhlmann [16] and Picano et al. [59] are also shown for comparison.

Grahic Jump Location
Fig. 15

The average forces normalized by [ρf(u∗)2dp2], acting on the particle along the transverse direction as a function of (a) y+ and (b) y/ap

Grahic Jump Location
Fig. 16

Comparison of the phase-partitioned mean velocity profiles: (a) Log-linear plot and (b) linear-linear plot

Grahic Jump Location
Fig. 17

Visualization of instantaneous spanwise vorticity ωz, normalized by u*2/ν, on an xy slice: (a) LBM–SP, (b) LBM–PLL, and (c) LBM–PLS

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In