TY - JOUR
T1 - A POD-Based Variational Multiscale Method for Large Eddy Simulation of Turbulent Channel Flows
AU - Chen, L.
AU - Hu, Xu Qu
PY - 2016
Y1 - 2016
N2 - A POD-based variational multiscale method (VMM) for large eddy simulation (LES) of turbulent channel flows is proposed. The incompressible Navier–Stokes equations are solved using a residual-based VMM technique, and the residual-based model coefficient is adopted from Taylor et al. [1998] “Finite element modeling of blood flow in arteries,” Comput. Methods in Appl. Mech. Eng. 158(1–2), 155–196. A generalized-(Formula presented.) method is applied for the temporal integration. Numerical results show that four or higher wall-normal modes need to be applied for reproducing most of the turbulent characteristics. Results of large vortex structures and the low velocity streaks prove that the turbulent channel flows are well estimated by the proposed approach. The statistical analysis of the turbulent velocities confirms its reliability in turbulent channel flows.
AB - A POD-based variational multiscale method (VMM) for large eddy simulation (LES) of turbulent channel flows is proposed. The incompressible Navier–Stokes equations are solved using a residual-based VMM technique, and the residual-based model coefficient is adopted from Taylor et al. [1998] “Finite element modeling of blood flow in arteries,” Comput. Methods in Appl. Mech. Eng. 158(1–2), 155–196. A generalized-(Formula presented.) method is applied for the temporal integration. Numerical results show that four or higher wall-normal modes need to be applied for reproducing most of the turbulent characteristics. Results of large vortex structures and the low velocity streaks prove that the turbulent channel flows are well estimated by the proposed approach. The statistical analysis of the turbulent velocities confirms its reliability in turbulent channel flows.
KW - proper orthogonal decomposition
KW - turbulent channel flow
KW - Variational multiscale method
UR - http://www.scopus.com/inward/record.url?scp=84994131716&partnerID=8YFLogxK
U2 - 10.1142/S0219876217500499
DO - 10.1142/S0219876217500499
M3 - Article
AN - SCOPUS:84994131716
SN - 0219-8762
VL - 14
JO - International Journal of Computational Methods
JF - International Journal of Computational Methods
IS - 5
ER -