CFD TURBULENCE MODELLING: RANS, LES & DNS

STRUCTURE
The formidable modelling task Overview some RANS models & limitations Show for LES model not that important Outline the things for LES that matter and the order of importance Discuss mixing LES & RANS models RANS model defects and management
2

KEY ROLE OF TURBULENCE

Drag generation Heat transfer Particle dispersion Scalar mixing Sound generation

TURBULENCE
l

y+

Transition

da Vinci - describes the

clouds as scattered and torn

Van Gogh

FORMIDABLE TASK
I am an old man now, and when I die and go to heaven there are two matters on which I hope for enlightenment. One is quantum electrodynamics, and the other is the turbulent motion of fluids. And about the former I am rather optimistic Sir Horace Lamb FRS (1849-1934) 2nd Wrangler Trinity College Turbulence is the last great unsolved problem in classical physics Richard Feynman (Nobel Prize in Physics - quantum electrodynamics ) Do not even know the Karman constant (l = y 0.38 < < 0.45) or if it is a constant!!! Spalart (2006) 2% decrease in gives 1% decrease in predicted aircraft drag

MODEL BASIS
Phenomenological but we do not fully understand the phenomena!!! Spalart & Allmaras (1994) La Recherche Aerospatiale, No 1, 5-21 Abstract A transport equation for turbulent viscosity is assembled based on empiricism and arguments of dimensional analysis
6

DICTIONARY DEFINITION
Empiricism - Philosophy. the doctrine that all knowledge is derived from sense experience. - Undue reliance upon experience, as in medicine; quackery.

SA MODEL BASIS
4 nested models: (I) Free shear flows; (II) log-outer layer; (III) buffer and viscous sublayer & (IV) laminar and trip region Model (I)

D t = C1St + Diffusion Term [ C2 , ] Dt

Shearing for production
8

CALIBRATION
2D mixing layer Wake
max = 0.01( U )2

max = 0.06( U )2

Calibration suggests 0.6<<1; 0.1375< C1<0.1275 & 0.6< C2<0.7 Pick:2/3, 0.1355, 0.622. Acknowledge plane jet spreading rate 38% too high
9

RANS t BASED MODELS

Boussinesq (1877)
U/y

uiuj = tSij

+ QUADRATIC TERMS + CUBIC TERMS

Gatski & Speziale (1993)

Craft, Launder & Suga (1996)

10

BASIC RANS MODELS

Classified by the number of differential equations Realizability - whether the model is constrained so that it does not break basic physical principles k > 0 Hundreds of basic RANS models - indicative of quest for something better.

11

ZERO EQUATION MODELS

Dimensional grounds t Length x Velocity Incomplete model Log layer viscous sublayer Outer BL Mixing layer Round jet Plane jet Wake

l eqn takes different forms for different flow zones.

l = y l=yD l=C l 0.07 x the layer width; l 0.075 x the jet half width; l 0.09 x jet half width and l 0.16 x the wake half width

Numerous l choices and these, in themselves, can be subjected to

further corrections see later. Zero equation models require relatively modest grids - slow iterative convergence. Seldom be expected to provide high accuracy unlikely to massively err.
12

ONE EQUATION TURBULENCE MODELS

Zero eqn. assume turbulence equilibrium. One-eqn allow turbulence transport

D = 2 + S Dt
SA model, Secundov, Baldwin-Barth t ; k-l model = k Grid forgiving frequently used just near walls

13

TWO EQUATION TURBULENCE MODELS

Vast number - most are based on k & k2 t = C Secundov et al. t and l; Kim and Chung k-t; Wilcox k-; Warner et al. k-kl k- - sensitive to the specified free stream turbulence intensity level - zonal Menter (1993) model.
14

RSM MODEL
Around 12 massive differential equations to solve Most exact model Contains substantial empiricism Computational cost is extreme Numerous gradient terms grid demands Separation or buoyancy large scale unsteady structures failure Uncertain near wall performance: modelling dissipation, pressure-strain, ad hoc modifications Wide range of options suggests a weakness and unresolved issues Simpler eddy viscosity models used near walls

15

NON-LINEAR EDDY VISCOSITY MODELS

Half-way-house between costly
RSM and simple linear EVM Extended forms of the Boussinesq approximation: quadratic or cubic Quadratic - anisotropy modelling; cubic - streamline curvature

Marketed as a means of getting RSM type performance at linear eddy viscosity model cost

16

QUALITATIVE PREDICTIVE ACCURACY

Round jet/plane jet anomaly - d/dx opposite traits Ma > 1 d/dx decreases x2 but RANS insensitive Axis switching t based models cant handle

Flow Plane jet Round jet

k- 0.091 0.101

Measured 0.1-0.11 0.086-0.096

17

WALL JETS

RANS 30% over-predict spreading rate

18

URANS

Linear models Non-linear models

OK - has spectral gap - unusual Liu and Tucker (2007) IJNME

19

URANS
T [K]

20

The Resolved Solution in Different Approaches

By Strelets group
21

WHAT IS THE SOLUTION? & WHAT IS LES?

RANS = Resolve time average of flow LES = Resolve all large eddies

Modelled < 2 x

Resolved/solved for

22

L. F. Richardsons (1922) Rhyme & Kolmogorov (1941)

Big whorls have little whorls, which feed on their velocity, and little whorls have lesser whorls, and so on to viscosity (in the molecular sense).
Big whorls

Kolmogorov (1941), smaller whorls or eddies isotropic Energy k-5/3

23

LES HIERARCHY
PROBLEM DEFINITION
BCs, Soln uniqueness

WALL MODELLING
GRID/SOLVER COMPATIBLITY

MODEL

24

ULTIMATE NUMERICAL INFLUENCES FOR LES

NUMEROUS FORMS OF NS EQNS Chow & Moin (JCP) 2003 NUMEROUS DISCRETIZATIONS dissipation Staggered grids, cell centered/vertex, codes with smoothers, Rhie and Chow, axis treatment one legged and two legged GRID TOPOLOGY dissipation

+ turbulence model validity + wall modelling + problem definition + soln uniqueness + = Do not get too hung up on turbulence model

25

HEAT TRANSFER

26

GRID INFLUENCE

Denton code 10-20 x faster than HYDRA

27

LES
Boussinesq (1877)
Comte-Bellot and Corsin

uiuj = sgsSij

+ Non-linear terms

Clark et al. (1979) [+Leonard] Kosovic (1997), Leray, LANS [Geurts & Holm (2005)] Cubic [Lund & Novikov (1992)]

28

LES MODELS TESTED

ui,j = ui/ xj

29

LES MODEL MINIMAL INFLUENCE GOOD RESULTS WITH NO LES MODEL!! x/D = 15 Re = 4 x 103 Circa 1 million cells

30

KEY LES PROBLEM

Resolving streaks Pope (2004), Boeing 777 at cruise 108 streaks
Hinze (1975)

LES Cost Re2.5* Hybrid LES-RANS Cost Re0.5 (Based on outer layer modelling)
y+=90

*Piomelli, AIAA-2008-396
31

WING-FLAP
[Re = 23 x 106, 3.3 million cells] ZONAL ILES-RANS vorticity contours

Model

CL
%

Error

RANS Zonal LES-RANS Zonal ILES-RANS ILES

+24 +10 -5 -16

32

CHEVRON ILES-RANS

33

JET WITH CO-FLOW

x/D=1

Re = 300,000

x/D=2

34

COMPRESSOR/TURBINE LES

35

DES of F-15 Post-Stall

By Forsythe, Wurtzler, Squires, Cobalt

36

Work of L. Hedges, NASA Funded

URANS

SRANS,
Partly Converged
Vorticity Magnitude

DES

37

Generic Heavy Truck in CrossWind

By Wurtzler, Forsythe, Cobalt

38

Electronics

Heat transfer zone

39

LES SIMULATIONS

40

WASHING OUT OF EDDY VISCOSITY

Surface orthogonal flow direction

41

PART 2: RANS MODEL DEFECTS AND MANAGEMENT

42

SIMPLER MODEL PROBLEMS

Cf. Boussinesq - uu=f(U/y), RSM - Duu/Dt= Erroneous predicted turbulence for curved shear layers and adverse pressure gradients; Separation suppression on curved surfaces Failure at separation Excessive turbulence in stagnation zones Wrong behaviour for rotating flows Insensitivity to density gradients Excessive heat transfer at reattachments points Insensitivity to system rotation
43

STAGNATION PROBLEM
Turbulence massively over predicted Suppression of leading edge separation; excessive predicted heat transfer Ad hoc corrections:
Kato and Launder (1993); Yap (1987) & C modifications
44

CURVATURE PROBLEM
Turbulence can be virtually eliminated around convex surfaces Basic Richardson correction (various forms, can be used in various places in same model)

U U Ri = R r
Many other curvature corrections available e.g. modification of the k or equation or C based on contractions of more advanced models
45

STREAMLINE CURVATURE

B. E. Launder Int. J. Heat and Fluid Flow,1989

46

BODY FORCE/SWIRL PROBLEM

Turbomachinery flows often involve rotation/swirl & local swirl d ( rU ) /( rdr ) > 0 Stable Rossby number (different forms)

Ro =

U / y

Many other correction forms - SARC

47

BUOYANCY
T/r > 0 unstable, Rayleigh-Bernard instability Large scale unsteady structures Way beyond even RSM Again a range of different corrections e.g. gradient based hypothesis

48

CONVEX FEATURES PROBLEM

Tucker, Rumsey and Spalart AIAA J. (2005)

wall distances

Eddy viscosity contours

Flow direction

| |=1

| |=1+ 2

=
49

CONVEX FEATURES PROBLEM

Hamilton-Jacobi equation Tucker, Rumsey, Spalart . (AIAA J.)

| |=1
Measured Palliative d function Normal calculation 3.10 2.90 2.77

| |=1+ 2 =

RANS still vital for design

50

TRANSITION MODELLING
Transition modelling is of some importance for LPT, engine inlets Influenced by roughness, acoustics and external disturbances, pressure gradient, freestream velocity change, surface curvature, temperature gradients and rotation Generally associated with the growth of Tollmien-Schlichting (TS) waves. Bypass transition, linear T-S wave process is bypassed, free stream intensities greater than 0.5% Some turbulence models can naturally predict the bypass transition process

51

TRANSITION MODELLING
k Convection = k Diffusion + P - D k convects and diffuses into BL Increases t, increases Pk = t (dU/dy)2 Substantial kave growth until transition

52

TRANSITION MODELLING
Standard transition procedure, estimate the transition point, exp. correlations (Abu-Ghannan and Shaw (1980), Arnal (1992)) Trip model Lots of models, active research area With fine enough grids LES will capture transition
53

REAL GEOMETRY, CAA AND DESIGN OPTIMISATION

54

REAL GEOMETRY

Blisters

55

DESIGN OPTIMISATION

56

SILENT AIRCRAFT

Acoustic waves reflect from upper surface

Shadow region

Sound reflected from underside of wing

Reflection of jet noise for a conventional podded engine installation

57

NOISE SHIELDING

58

Low Noise Design: High-Lift System

59

Combustion Noise

URANS + LES + HighFidelity Models

60

CONCLUSIONS
Vast number of RANS models, choice can have substantial impact - CFD use a specialist activity CFD predict correct deltas To predict exact levels extreme insights into turbulence model and many other CFD aspects + calibration data Rationalism is desirable with every effort being made to place the CFD solution on a solid rational basis Many practical flows are highly three dimensional in which inviscid pressure driven structures occur and then turbulence stresses become less important However, if the 3D structures are unsteady in nature, other challenges arise

61

CONCLUSIONS
URANS can help Zonal RANS-LES & LES with take over but when? Depends on HPC/GPU developments Zonal RANS-LES & LES still need physical insight by analyst

LES Hierarchy?:
Problem definition, solution uniqueness, transition, separation + Near wall LES modelling + Grid structure/quality > LES Model

62