Direct numerical simulation and large eddy simulation on a turbulent wall-bounded flow using lattice Boltzmann method and multiple GPUs

Xian Wang, Yanqin Shangguan, Naoyuki Onodera, Hiromichi Kobayashi, Takayuki Aoki

Research output: Contribution to journalArticle

20 Citations (Scopus)

Abstract

Direct numerical simulation (DNS) and large eddy simulation (LES) were performed on the wall-bounded flow at Re τ = 180 using lattice Boltzmann method (LBM) and multiple GPUs (Graphic Processing Units). In the DNS, 8 K20M GPUs were adopted. The maximum number of meshes is 6.7 × 10 7, which results in the nondimensional mesh size of Δ + = 1.41 for the whole solution domain. It took 24 hours for GPU-LBM solver to simulate 3 × 10 6 LBM steps. The aspect ratio of resolution domain was tested to obtain accurate results for DNS. As a result, both the mean velocity and turbulent variables, such as Reynolds stress and velocity fluctuations, perfectly agree with the results of Kim et al. (1987) when the aspect ratios in streamwise and spanwise directions are 8 and 2, respectively. As for the LES, the local grid refinement technique was tested and then used. Using 1.76 × 10 6 grids and Smagorinsky constant (C s) = 0.13, good results were obtained. The ability and validity of LBM on simulating turbulent flow were verified.

Original languageEnglish
Article number742432
JournalMathematical Problems in Engineering
Volume2014
DOIs
Publication statusPublished - 2014

Fingerprint

Wall flow
Large Eddy Simulation
Lattice Boltzmann Method
Direct numerical simulation
Large eddy simulation
Graphics Processing Unit
Aspect ratio
Aspect Ratio
Turbulent flow
Mesh
Grid Refinement
Local Refinement
Reynolds Stress
Turbulent Flow
Direct numerical Simulation
Graphics processing unit
Fluctuations
Grid

ASJC Scopus subject areas

  • Mathematics(all)
  • Engineering(all)

Cite this

Direct numerical simulation and large eddy simulation on a turbulent wall-bounded flow using lattice Boltzmann method and multiple GPUs. / Wang, Xian; Shangguan, Yanqin; Onodera, Naoyuki; Kobayashi, Hiromichi; Aoki, Takayuki.

In: Mathematical Problems in Engineering, Vol. 2014, 742432, 2014.

Research output: Contribution to journalArticle

@article{133b09d0ed2f4477bf243f72fe991833,
title = "Direct numerical simulation and large eddy simulation on a turbulent wall-bounded flow using lattice Boltzmann method and multiple GPUs",
abstract = "Direct numerical simulation (DNS) and large eddy simulation (LES) were performed on the wall-bounded flow at Re τ = 180 using lattice Boltzmann method (LBM) and multiple GPUs (Graphic Processing Units). In the DNS, 8 K20M GPUs were adopted. The maximum number of meshes is 6.7 × 10 7, which results in the nondimensional mesh size of Δ + = 1.41 for the whole solution domain. It took 24 hours for GPU-LBM solver to simulate 3 × 10 6 LBM steps. The aspect ratio of resolution domain was tested to obtain accurate results for DNS. As a result, both the mean velocity and turbulent variables, such as Reynolds stress and velocity fluctuations, perfectly agree with the results of Kim et al. (1987) when the aspect ratios in streamwise and spanwise directions are 8 and 2, respectively. As for the LES, the local grid refinement technique was tested and then used. Using 1.76 × 10 6 grids and Smagorinsky constant (C s) = 0.13, good results were obtained. The ability and validity of LBM on simulating turbulent flow were verified.",
author = "Xian Wang and Yanqin Shangguan and Naoyuki Onodera and Hiromichi Kobayashi and Takayuki Aoki",
year = "2014",
doi = "10.1155/2014/742432",
language = "English",
volume = "2014",
journal = "Mathematical Problems in Engineering",
issn = "1024-123X",
publisher = "Hindawi Publishing Corporation",

}

TY - JOUR

T1 - Direct numerical simulation and large eddy simulation on a turbulent wall-bounded flow using lattice Boltzmann method and multiple GPUs

AU - Wang, Xian

AU - Shangguan, Yanqin

AU - Onodera, Naoyuki

AU - Kobayashi, Hiromichi

AU - Aoki, Takayuki

PY - 2014

Y1 - 2014

N2 - Direct numerical simulation (DNS) and large eddy simulation (LES) were performed on the wall-bounded flow at Re τ = 180 using lattice Boltzmann method (LBM) and multiple GPUs (Graphic Processing Units). In the DNS, 8 K20M GPUs were adopted. The maximum number of meshes is 6.7 × 10 7, which results in the nondimensional mesh size of Δ + = 1.41 for the whole solution domain. It took 24 hours for GPU-LBM solver to simulate 3 × 10 6 LBM steps. The aspect ratio of resolution domain was tested to obtain accurate results for DNS. As a result, both the mean velocity and turbulent variables, such as Reynolds stress and velocity fluctuations, perfectly agree with the results of Kim et al. (1987) when the aspect ratios in streamwise and spanwise directions are 8 and 2, respectively. As for the LES, the local grid refinement technique was tested and then used. Using 1.76 × 10 6 grids and Smagorinsky constant (C s) = 0.13, good results were obtained. The ability and validity of LBM on simulating turbulent flow were verified.

AB - Direct numerical simulation (DNS) and large eddy simulation (LES) were performed on the wall-bounded flow at Re τ = 180 using lattice Boltzmann method (LBM) and multiple GPUs (Graphic Processing Units). In the DNS, 8 K20M GPUs were adopted. The maximum number of meshes is 6.7 × 10 7, which results in the nondimensional mesh size of Δ + = 1.41 for the whole solution domain. It took 24 hours for GPU-LBM solver to simulate 3 × 10 6 LBM steps. The aspect ratio of resolution domain was tested to obtain accurate results for DNS. As a result, both the mean velocity and turbulent variables, such as Reynolds stress and velocity fluctuations, perfectly agree with the results of Kim et al. (1987) when the aspect ratios in streamwise and spanwise directions are 8 and 2, respectively. As for the LES, the local grid refinement technique was tested and then used. Using 1.76 × 10 6 grids and Smagorinsky constant (C s) = 0.13, good results were obtained. The ability and validity of LBM on simulating turbulent flow were verified.

UR - http://www.scopus.com/inward/record.url?scp=84899972630&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84899972630&partnerID=8YFLogxK

U2 - 10.1155/2014/742432

DO - 10.1155/2014/742432

M3 - Article

AN - SCOPUS:84899972630

VL - 2014

JO - Mathematical Problems in Engineering

JF - Mathematical Problems in Engineering

SN - 1024-123X

M1 - 742432

ER -