Jiang X., Lai C-H. Numerical Techniques for Direct and Large-Eddy Simulations

CRC, 2009. 276 р. ISBN: 1420075780Compared to the traditional modeling of computational fluid dynamics, direct numerical simulation (DNS) and large-eddy simulation (LES) provide a very detailed solution of the flow field by offering enhanced capability in predicting the unsteady features of the flow field. In many cases, DNS can obtain results that are impossible using any other means while LES can be employed as an advanced tool for practical applications. Focusing on the numerical needs arising from the applications of DNS and LES, Numerical Techniques for Direct and Large-Eddy Simulations covers basic techniques for DNS and LES that can be applied to practical problems of flow, turbulence, and combustion.Acknowledgments xv
AuthorsGoverning Equations: Compressible and Incompressible Formulations
Fundamental Governing Equations for Fluid Flows
Comments on the Governing Equations
Governing Equations for Incompressible Flows
Turbulence and Direct Numerical Simulation
Large-Eddy SimulationNumerical Treatment of Boundary Conditions
Inflow and Outflow Boundary Conditions
Inlow BC for DNS and LES
Outlow BC for DNS and LES
Inlow/Outlow BC Based on NSCBC for Compressible Flows
Inlow/Outlow BC for Incompressible Flows
Wall Boundary Conditions
Classical Wall Models
Zonal and Nonzonal Approaches for LES
Near-Wall Models for Nonisothermal Flows
Other Boundary Conditions
Far-Field and Open Boundary Conditions for Compressible Flows
Periodic and Symmetry Boundary Conditions
Boundary Conditions: Other Relevant IssuesDiscrete Time Integration Methods
High-Order Runge–Kutta (Rk) Methods
Linear Multistep Methods: Adams–Bashforth and Adams–Moulton Methods
The Adams–Bashforth Methods
The Adams–Moulton Methods
Other Time Integration Methods
General Implicit Schemes and Backward Diferentiation Formulae
Other Second-Order SchemesDNS of Incompressible Flows
Ample Results: DNS of Channel Flows
DNS of a Separated Channel Flow with a Smooth Proile
DNS of Turbulent Heat Transfer in Pipe Flows
DNS of Turbulent Channel Flow under Stable Stratiication
DNS of Coherent Structure in Turbulent Open-Channel Flows with Heat Transfer
Numerical Features: DNS of Incompressible Flows
Spatial Discretization Schemes
The Fractional Step MethodDNS of Compressible Flows
Sample Results: DNS of Compressible Jet Flows
DNS of a Compressible Plane Jet
DNS of a Variable-Density Round Jet
DNS of a Transitional Rectangular Jet
DNS of a Swirling Annular Jet Flame
Numerical Features: High-Order Schemes for Spatial Discretization
Centered Padé Schemes
Boundary Closures for High-Order Finite Diference Schemes
Other High-Order Finite Diference Schemes
Finite Volume and Spectral-Volume MethodsLES of Incompressible Flows
Sample Results: LES of Incompressible Flows in Complex Geometries
LES of Reacting Turbulent Flows in Complex Geometries
LES of Separated Flows over a Backward-Facing Step
Subgrid Scale Modeling of Incompressible Flows
Subgrid Scale Modeling
SGS Modeling of Reacting Flows
Numerical Features: LES On Unstructured Grids and Immersed Boundary Technique for Complex Geometries
LES on Unstructured Grids
LES with the Immersed Boundary TechniqueLES of Compressible Flows
Sample Results of LES of Compressible Flows
LES of a Ramjet Combustor
LES of Compressible Flows around Cavities
Subgrid-Scale Modeling of Compressible Flows and Implicit Large-Eddy Simulation (ILES)
Subgrid-Scale Modeling of Compressible Flows
Implicit Large-Eddy Simulation (ILES)Further Topics and Challenges in DNS and LES
Multiscale Flow Simulations
The Concept of Multiscale Modeling
An Example of Multiscale Flow Modeling: Turbulent Atomization
Challenges in DNS and LES: Complex Geometry and SGS Modeling
Challenges in DNS: Complex Geometry
Challenges in LES: SGS Modeling
Hybridization: Detached Eddy Simulation (DES)Appendix: Fortran 90 Routines of the Finite Difference Schemes