A new dynamic one-equation subgrid-scale model for large eddy simulations

• Published in: International journal for numerical methods in engineering Vol. 81; no. 7; pp. 835 - 865
• Main Authors: Huang, Shenghong, Li, Q. S.
• Format: Journal Article
• Language: English
• Published: Chichester, UK John Wiley & Sons, Ltd 12-02-2010; Wiley
• Subjects: Computational fluid dynamics; Computational techniques; Dynamics; Earth, ocean, space; Exact sciences and technology; External geophysics; Fluid dynamics; Fluid flow; Fundamental areas of phenomenology (including applications); General theory; High Reynolds number; Large eddy simulation; Mathematical methods in physics; Mathematical models; Meteorology; Physics; subgrid scale model; Turbulence; Turbulence simulation and modeling; Turbulent flow; turbulent flows; Turbulent flows, convection, and heat transfer; wind engineering; wind tunnel test; Winds and their effects; Wind; Turbulent flow; Scale models; Adaptability; Combustion; subgrid scale model; wind engineering; Grids; Large Reynolds number; Filters; Large eddy simulation; Energy dissipation; Scaling laws; Turbulent viscosity; Modelling; Kinetic energy; Dynamic model; Mesh generation; Unstructured mesh; Subgrid scale method; computational fluid dynamics; Reynolds number; Wind tunnel test; Aerodynamics; turbulent flows; Eddy viscosity; Energy losses
• ISSN: 0029-5981; 1097-0207
• DOI: 10.1002/nme.2715

A new dynamic one‐equation subgrid‐scale (SGS) model is presented for large eddy simulations of turbulent flows. The new model, combining both advantages of the dynamic one‐equation SGS model (J. Appl. Mech. (ASME) 2006; 73:368–373) and the wall‐adapting local eddy viscosity model (Turbulence Combustion 1999; 62:183–200), has three prominent features: (1) one‐equation model, suitable for relatively coarse grid situations and simulation of high Reynolds number flows; (2) no test‐filtering operation is needed in determination of dynamic parameters, suitable for low‐order numerical discretization and unstructured or hybrid grid situations; and (3) treating the production of SGS kinetic energy and energy loss in grid‐scale (GS) portion due to SGS motion with different dynamic mechanisms, which is considered to be more reasonable than the local and instantaneous dynamic mechanism as adopted by most existing one‐equation dynamic SGS models. Computational examinations have been conducted for flows with Reynolds number ranging from 3000–70000, revealing that there are three promising aspects of the new SGS model: (1) adaptability for a wide range of flow regime; (2) less grid dependence; and (3) potential of wide applications in complex geometries and high Reynolds number turbulent flows. Copyright © 2009 John Wiley & Sons, Ltd.