aerodynamic and static aeroelastic numerical simulations for the 6th - - PowerPoint PPT Presentation

aerodynamic and static aeroelastic numerical simulations for the 6th aiaa cfd drag prediction workshop

6th AIAA CFD Drag Prediction Workshop - 2016

Rodrigo Felix de Souza Murilo C. Mestriner Maximiliano A. F. de Souza Marcello Areal Ferrari Carlos Breviglieri Cleber Spode June 16, 2016

Embraer S/A

Outline

Introduction Numerical simulations results Case 1 Case 2 Case 3 Case 5 Bonus track

1

introduction

Considered Cases and solver setup

Considered four series of computations:

∙ Case 1: Verification study; ∙ Case 2: CRM Nacelle-Pylon Drag Increment; ∙ Case 3: CRM WB Static Aero-Elastic Effect; ∙ Case 5: CRM WB Coupled Aero-Structural Simulation;

Solver setup

Solver : CFD++ (14.1.1) Formulation : Pre-conditioned compressible RANS, perfect gas Turbulence model : SST and SA (case 1 only) Time integration : Point-implicit (SGS) / Algebraic multigrid Spatial discretization : Finite volume, 2nd order Polynomial type : Nodal-based TVD limiter : Minmod compression 2

3

Grids

Grids @ Tiny level

(a) CommonHybrid (b) CustomHexa

4

Grids

Grids @ Tiny level

(c) CustomHybrid-I (d) CustomHybrid-A

5

Grids

Grid sizes in million (WB-AE275): Tiny Fine Nodes Cells Nodes Cells CommonHybrid 20.5 83.6 66.2 271.2 CustomHexa 20.3 20.0 70.3 69.6 CustomHybrid-A 9.4 19.7 23.9 72.6 CustomHybrid-I 8.0 20.3 25.7 68.6 We matched the gridding guidelines for number of cells!

6

numerical simulations results

Case 1: Verification Study

Flow condition: M = 0.15 Re = 6 million AOA = 10 degrees Grid: NACA 0012 TMBWG Family II Turbulence model: SA and SST (CFD++)

8

Case 1: Verification Study

Grid convergence: Lift

1.060 1.070 1.080 1.090 1.100 1.110 0.000 0.002 0.004 0.006 0.008 0.010 CL h=sqrt(1/N) CFD++ - SST CFD++ - SA CFL3D - SA FUN3D - SA 9

Case 1: Verification Study

Grid convergence: Drag

0.012 0.013 0.014 0.015 0.016 0.017 0.018 0.000 0.002 0.004 0.006 0.008 0.010 CD h=sqrt(1/N) CFD++ - SST CFD++ - SA CFL3D - SA FUN3D - SA 9

Case 1: Verification Study

Grid convergence: Drag (pressure component)

0.0040 0.0050 0.0060 0.0070 0.0080 0.0090 0.0100 0.0110 0.0120 0.000 0.002 0.004 0.006 0.008 0.010 CDP h=sqrt(1/N) CFD++ - SST CFD++ - SA CFL3D - SA FUN3D - SA 9

Case 1: Verification Study

Grid convergence: Drag (viscous component)

0.0057 0.0057 0.0058 0.0058 0.0059 0.0059 0.0060 0.0060 0.0061 0.0061 0.0062 0.0062 0.000 0.002 0.004 0.006 0.008 0.010 CDV h=sqrt(1/N) CFD++ - SST CFD++ - SA CFL3D - SA FUN3D - SA 9

Case 1: Verification Study

Pressure distribution @ finest grid

• 7.0
• 6.0
• 5.0
• 4.0
• 3.0
• 2.0
• 1.0

0.0 1.0 2.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 CP x/c CFD++ - SST CFD++ - SA CFL3D - SA FUN3D - SA

10

Case 1: Verification Study

Pressure distribution @ finest grid

• 7.0
• 6.0
• 5.0
• 4.0
• 3.0
• 2.0
• 1.0

0.0 1.0 2.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 CP x/c CFD++ - SST CFD++ - SA CFL3D - SA FUN3D - SA

• 6.0
• 5.5
• 5.0
• 4.5
• 4.0

0.00 0.01 Cp @ L.E.

10

Case 1: Verification Study

Skin friction @ finest grid

• 0.010
• 0.005

0.000 0.005 0.010 0.015 0.020 0.025 0.030 0.035 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 CFX x/c CFD++ - SST CFD++ - SA CFL3D - SA FUN3D - SA

11

Case 2: CRM Nacelle-Pylon Drag Increment

Flow condition: M = 0.85 Re = 5 million CL = 0.50 Turbulence model: SST (CFD++) Grids: CommonHybrid : T,C,M,F CustomHexa : T,C,M,F,X,U (WB only) CustomHybrid-I : T,C,M,F CustomHybrid-A : T,C,M,F Geometry: Aeroelastic deflection at the angle-of-attack 2.75 degrees

12

Case 2: CRM Nacelle-Pylon Drag Increment

Total drag convergence @ WB

0.0250 0.0251 0.0252 0.0253 0.0254 0.0255 0.0256 0.0257 0.0258 0.0259 0.0260 0.0e+00 5.0e-06 1.0e-05 1.5e-05 2.0e-05 2.5e-05 3.0e-05

CDTOT GRIDFAC = 1/GRIDSIZE(2/3)

CommonHybrid CustomHexa CustomHybrid-A CustomHybrid- I 20M 100M 8M 26M CommonHybrid CustomHexa CustomHybrid-A CustomHybrid-I

⎯ ⎯

↕

1 Drag Count

WB ➪ ➪ ➪ ➪

13

Case 2: CRM Nacelle-Pylon Drag Increment

Total drag convergence @ WBNP

0.0274 0.0275 0.0276 0.0277 0.0278 0.0279 0.0280 0.0281 0.0282 0.0283 0.0284 0.0e+00 5.0e-06 1.0e-05 1.5e-05 2.0e-05 2.5e-05 3.0e-05

CDTOT GRIDFAC = 1/GRIDSIZE(2/3)

CommonHybrid CustomHybrid-A CustomHybrid- I 28M 91M 11M 36M CommonHybrid CustomHybrid-A CustomHybrid-I

⎯ ⎯

↕

1 Drag Count

WBNP ➪ ➪ ➪

13

Case 2: CRM Nacelle-Pylon Drag Increment

Delta NP drag convergence (WBNP-WB)

0.0018 0.0019 0.0020 0.0021 0.0022 0.0023 0.0024 0.0025 0.0026 0.0027 0.0028 0.0e+00 5.0e-06 1.0e-05 1.5e-05 2.0e-05 2.5e-05 3.0e-05

∆CDTOT

GRIDFAC = 1/GRIDSIZE(2/3)

CommonHybrid CustomHybrid-A CustomHybrid- I

TINY FINE TINY FINE

CommonHybrid CustomHybrid-A CustomHybrid-I

⎯ ⎯

↕

1 Drag Count

∆NP

TINY FINE

➪ ➪ ➪

13

Case 2: CRM Nacelle-Pylon Drag Increment

Angle of attack for each grid @ CL 0.50

2.00 2.10 2.20 2.30 2.40 2.50 2.60 2.70 2.80 2.90 3.00 0.0e+00 5.0e-06 1.0e-05 1.5e-05 2.0e-05 2.5e-05 3.0e-05

ALPHA [ ° ] GRIDFAC = 1/GRIDSIZE(2/3)

CommonHybrid CustomHexa CustomHybrid-A CustomHybrid- I 20M 100M 8M 26M CommonHybrid CustomHexa CustomHybrid-A CustomHybrid-I

⎯ ⎯

↕

0.1 °

ALPHA WB ➪ ➪ ➪ ➪

13

Case 2: CRM Nacelle-Pylon Drag Increment

Pressure distribution @ wing section 11 (Fine grid)

• 1.40
• 1.20
• 1.00
• 0.80
• 0.60
• 0.40
• 0.20

0.00 0.20 0.40 0.60 0.80 1.00 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 Cp x/c CommonHybrid CustomHexa CustomHybrid-A CustomHybrid-I

13

Case 2: CRM Nacelle-Pylon Drag Increment

Side-of-body separation (SOB): CommonHybrid (T) Nodes : ≈ 20 mi Cells : ≈ 84 mi CustomHexa (T) Nodes : ≈ 20 mi Cells : ≈ 20 mi CustomHybrid-I-A (F) Nodes : ≈ 24 mi Cells : ≈ 73 mi

14

Case 3: CRM WB Static Aero-Elastic Effect

Flow condition: M = 0.85 Re = 5 million AOA = 2.50 | 2.75 | 3.00 | 3.25 | 3.50 | 3.75 | 4.00 degrees AE = 2.50 | 2.75 | 3.00 | 3.25 | 3.50 | 3.75 | 4.00 degrees Grids: CommonHybrid @ Medium level (45M nodes) CustomHexa @ Medium level (45M nodes) CustomHybrid-A @ Medium level (17M nodes) CustomHybrid-I @ Medium level (17M nodes) Turbulence model: SST (CFD++)

15

Case 3: CRM WB Static Aero-Elastic Effect

Wing deformation effect on lift and drag polar CustomHexa

0.48 0.52 0.56 0.60 0.64 2.50 2.75 3.00 3.25 3.50 3.75 4.00 CL AOA [deg] AE2.50 AE2.75 AE3.00 AE3.25 AE3.50 AE3.75 AE4.00

(a) Lift

0.0290 0.0330 0.0370 0.0410 0.56 0.60 0.64 CD CL AE2.50 AE2.75 AE3.00 AE3.25 AE3.50 AE3.75 AE4.00

(b) Drag Polar

Increased AE Decreased tip incidence Decreased lift slope

16

Case 3: CRM WB Static Aero-Elastic Effect

Wing deformation effect on lift and drag polar CustomHexa

0.48 0.52 0.56 0.60 0.64 2.50 2.75 3.00 3.25 3.50 3.75 4.00 CL AOA [deg] AE2.50 AE4.00 AEswp

(a) Lift

0.0290 0.0330 0.0370 0.0410 0.56 0.60 0.64 CD CL AE2.50 AE4.00 AEswp

(b) Drag Polar

Increased AE → Decreased tip incidence → Decreased lift slope

16

Case 3: CRM WB Static Aero-Elastic Effect

Mesh effect on lift and drag

0.480 0.540 0.600 0.660 2.50 2.75 3.00 3.25 3.50 3.75 4.00 CL AE [deg] CommonHybrid CustomHexa CustomHybrid-A CustomHybrid-I

(a) Lift

0.0240 0.0280 0.0320 0.0360 0.0400 0.0440 2.50 2.75 3.00 3.25 3.50 3.75 4.00 CD AE [deg] CommonHybrid CustomHexa CustomHybrid-A CustomHybrid-I

(b) Drag

17

Case 3: CRM WB Static Aero-Elastic Effect

Mesh and wing deformation effect on Cp distribution

• 1.40
• 1.20
• 1.00
• 0.80
• 0.60
• 0.40
• 0.20

0.00 0.20 0.40 0.60 0.80 1.00 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 Cp x/c CommonHybrid CustomHexa CustomHybrid-A CustomHybrid-I

(a) Wing Section 11 - AE2.50 - AOA=2.5o

• 1.40
• 1.20
• 1.00
• 0.80
• 0.60
• 0.40
• 0.20

0.00 0.20 0.40 0.60 0.80 1.00 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 Cp x/c CommonHybrid CustomHexa CustomHybrid-A CustomHybrid-I

(b) Wing Section 11 - AE4.00 - AOA=4.0o

18

Conclusion - Case 1

∙ Results for CFD++ SA are similar to CFL3D and FUN3D except for a shift in total drag; ∙ The difference comes mainly from the pressure drag component; ∙ The SST turbulence model generates different results in comparison to SA;

19

Conclusion - Case 2 and 3

∙ Despite the difference in grid cells:

∙ The CustomHybrid-A is only 1 to 2 dc away from CommonHybrib (WB and WBNP); ∙ On the other hand, CustomHybrid-I generates differences in drag up to 5 dc for the same number of elements; ∙ This result highlights the importance of the way the elements are distributed;

∙ The SOB separation seems to be more related to gridding strategy than to the grid size; ∙ The grid has a significant influence on predicting CD, CL and CP for higher AE deflections;

20

Case 5: WB Coupled Aero-Structural Sim.

∙ AERO Suite, CMSoft Inc.

∙ Coupled fluid/structral/thermal analysis; ∙ Euler, RANS, DES and LES; ∙ ALE (Arbitrary Lagrangian Eulerian) 2nd-order Finite Volume solver; ∙ Linear/non-linear structural analysis;

∙ Modal base obtained with MSC Nastran (linear structure); ∙ SA QCR-2013 up to wall; ∙ Quasi-static CL-driven fluid-structure simulation;

21

Case 5: WB Coupled Aero-Structural Sim.

WT solid FEM model 0.027:1 (10 modes are considered)

(a) 3rd mode: 39.4 Hz (b) 4th mode: 41.0 Hz (c) 1st mode: 39.4 Hz (d) 2nd mode: 41.0 Hz

22

Case 5: WB Coupled Aero-Structural Sim.

CFD medium (level 3) mirrored custom-hexa Mesh motion linear torsional spring analogy min(vol) = O(10−21)m3 Dressing mesh 20k tri AE0.0 loft

23

Case 5: WB Coupled Aero-Structural Sim.

Numerical convergence of coupled system. Wall-clock 3 hours for 1500 partitions.

24

Case 5: WB Coupled Aero-Structural Sim.

CRM NTF 0.027:1 M 0.85 RE 5.0×106 α 2.5971◦ CL 0.499989 CD 0.025412 Q 63.215 kPa, 1320 psf

25

Case 5: WB Coupled Aero-Structural Sim.

Displacement Z

26

Case 5: WB Coupled Aero-Structural Sim.

Displacement Y

26

Case 5: WB Coupled Aero-Structural Sim.

Cp contours

26

Case 5: WB Coupled Aero-Structural Sim.

CL vs span

26

Case 5: WB Coupled Aero-Structural Sim.

Twist vs span

26

Case 5: WB Coupled Aero-Structural Sim.

Wing “shortening”

26

Conclusion - Case 5

∙ FEM is particular to the wind tunnel configuration; ∙ Flexibility effects play a major role in forces and moments prediction; ∙ Secondary geometric effects (“shortening”) are accounted for in coupled simulations; ∙ FSI methods are ready and feasible for complex flight-shape analysis, i.e. transonic flows, practical geometries and meshes.

27

Recommendations - Case 5

∙ Quantify wing bending vs wing twist effects; ∙ Shorten the span for provided static aero-elastic geometries; ∙ Improve FEM analysis header. Excess of 6GB for Nastran F06 file with only 10 modes; ∙ Improve case 5 website description/requirements: loft scale, FEM - wingbox vs WTT; ∙ Experiment mesh motion methods with hybrid tetra+prism meshes.

28

bonus track

The GMA – Automatic Mesh Generator

Replay file .tcl written on ICEMCFD: ∙ Read geometry ∙ Clean up ∙ Generate support entities (points and curves) ∙ Perform block cutting ∙ Calculate bunching ∙ Output grid

30

The GMA – Features

Features: ∙ Fast mesh generation ∙ High quality elements ∙ Consistency Adequate for: ∙ Refinement studies – whole airplane or particular region ∙ Accurate comparison between different geometries

31

The GMA – Read Geometry

32

The GMA – Support Curves

33

The GMA – Support Points

34

The GMA – Block Cutting

35

The GMA – Block Cutting - Edges Only

36

The GMA – Grid Generation

37

The GMA – Grid Generation

37

The GMA – Nose Detail

38

The GMA – Nose Detail

39

The GMA – Wing Fairing Intersection (LE)

40

The GMA – Wing Fairing Intersection (LE)

41

The GMA – Wing Tip (LE)

42

The GMA – Wing Tip (LE)

43

The GMA – Wing Tip (TE)

44

The GMA – Wing Tip (TE)

45

The GMA – Wing Tip (TE)

46

The GMA – Tail Cone

47

The GMA – Tail Cone

48

The GMA – Refinement Studies

The next slides show the grids for the cases: ∙ Tiny (20 M) ∙ Coarse (30 M) ∙ Medium (45 M) ∙ Fine (70 M) ∙ Extra Fine (100 M) ∙ Ultra Fine (150 M) Wing Deflection = 2.75 Deg

49

Tiny – 19,957,518 vol. cells

50

Coarse – 29,985,790 vol. cells

51

Medium – 44,930,430 vol. cells

52

Fine – 69,994,449 vol. cells

53

Extra Fine – 100,984,026 vol. cells

54

Ultra Fine – 149,991,722 vol. cells

55

The GMA – Wing Deflecion Study

The next slides show the grids for the deflections ∙ 0.00 deg ∙ 2.50 deg ∙ 2.75 deg ∙ 3.00 deg ∙ 3.25 deg ∙ 3.50 deg ∙ 3.75 deg ∙ 4.00 deg

56

The GMA – Space Blocking – 0.00 deg

57

The GMA – Space Blocking – 2.50 deg

58

The GMA – Space Blocking – 2.75 deg

59

The GMA – Space Blocking – 3.00 deg

60

The GMA – Space Blocking – 3.25 deg

61

The GMA – Space Blocking – 3.50 deg

62

The GMA – Space Blocking – 3.75 deg

63

The GMA – Space Blocking – 4.00 deg

64

The GMA – General Comments

The grids requires basically two sizes for generation: ∙ BL first element ∙ Element at the leading edge (wing root) All the rest of the grid is defined by prescribing growth ratios. Grids

• f other sizes can easily be obtained by changing these Grid

parameters.

65

The GMA – Uniform Grid Elements at Surface

Investigation on how the number of uniform grid elements on surface influences drag calculation

66

Insert Elements on BL

67

Insert Elements on BL

68

Insert Elements on BL

69

Insert Elements on BL

70

Insert Elements on BL

71

Insert Elements on BL

72

1 Elts @ BL – insert elmt – Wing Fus Intersection

73

2 Elts @ BL – insert elmt – Wing Fus Intersection

74

3 Elts @ BL – insert elmt – Wing Fus Intersection

75

4 Elts @ BL – insert elmt – Wing Fus Intersection

76

5 Elts @ BL – insert elmt – Wing Fus Intersection

77

Drag Calculation - Inserting Layers

0.0250 0.0251 0.0252 0.0253 0.0254 0.0255 0.0256 0.0257 0.0258 0.0259 0.0260 1 2 3 4 5 CD Number of uniform layers AE2.75 IL AE4.00 IL

78

Uniform Grid Elements at Surface

∙ Uniform elements do influence calculation of absolute Drag (≈1 dc for 5 elements); ∙ Sistematic way of generating grids yields smoothness and allows comparison between runs; ∙ Delta drags do not differ for different aeroelastic deformations (differences match almost perfectly);

79

Next steps

∙ Extend GMA for WBPN

80

Thank you! (Obrigado)

81