008 - KHI Contribution to HiLiftPW-3008 - KHI Contribution to HiLiftPW-3 Hidemasa Yasuda *, Taku...

17KT008677 KHI proprietary

008 - KHI Contribution to HiLiftPW-3

Hidemasa Yasuda*, Taku Nagata*, Atsushi Tajima†, Akio Ochi*

* Kawasaki Heavy Industries, Ltd.† Kawaju Gifu Engineering Co., Ltd.

3rt High-Lift Prediction WorkshopDenver, CO June 3-4, 2017

© 2017 Kawasaki Heavy Industries, Ltd. All Rights Reserved 2KHI proprietary

Outline

Summary of cases completed Introduction of “Cflow” Overview of grid systems and numerical methods Results of HL-CRM Results of JSM Summary and future works


Summary of cases completed

CaseAlpha=8, Fully turb, grid study

Alpha=16, Fully turb,grid study

1a (full gap) yes yes

1b (full gap w adaption) yes yes

1c (partial seal) yes yes

1d (partial seal w adaption) yes yes

CasePolar,

Fully turbPolar,

specifiedtransition

Polar, w transition prediction

2a (no nacelle) yes no no

2b (no nacelle w adaption) yes no no

2c (with nacelle) yes no no

2d (with nacelle w adaption) yes no no

Case 2D

Verification study

3 yes

HL-CRM

JSM

DSMA661 Airfoil

Topics of this presentation


Introduction of “Cflow”

Cflow = Grid Generator + Flow Solver

Cflow has been validated in various workshops.

Non-orthogonal Octree AMR＋layered grid

unsteadyhighly complicated large-scale

Kawasaki original CFD tool

2013HiLift-PW2

2010-2016BANC I-IV

2016DPW6

Unsteady flow Steady flow


Overview of grid systems

Case Grid System Solver Comment1a, 1c B3-HLCRM_UnstrHexPrismPyrTet_PW

CflowSolver

1b, 1d CflowGrid (Unstructured Hexahedra) Single grid is applied to all AoA.

2a, 2c D-JSM_UnstrMixed_JAXA

2b, 2d CflowGrid (Unstructured Hexahedra) Single grid is applied to all AoA.

Sectional GridHL-CRM full gap, eta=0.552, AoA=8[deg]

CflowGrid (HL-CRM full gap) Details about Cflow Grid were presented in GMGW-1.

Flow Adapted

Boundary-fitted Layer

Non-orthogonal Octree


Summary of code and numerics used

Reference for Cflow details1. Nagata, T., Ueno, Y., and Ochi, A., “Validation of new CFD tool using Non-orthogonal Octree with Boundary-fitted

Layer Unstructured Grid,” 50th AIAA Aerospace Sciences Meeting, (AIAA 2012-1259).2. Ueno, Y., Nagata, T., and Ochi, A., “Aeroacoustic Analysis of the Rudimentary Landing Gear Using Octree

Unstructured Grid with Boundary-fitted Layer,” 18th AIAA/CEAS Aeroacoustics Conference, (AIAA 2012-2284).3. Yasuhiro Ito, Mitsuhiro Murayama, Atsushi Hashimoto, Takashi Ishida, Kazuomi Yamamoto, Takashi Aoyama,

Kentaro Tanaka, Kenji Hayashi, Keiji Ueshima, Taku Nagata and Akio Ochi, “TAS Code, FaSTAR and Cflow Results for the Sixth Drag Prediction Workshop,” 55th AIAA Aerospace Sciences Meeting, AIAA SciTech, (AIAA 2017-0959).

4. Atsushi Hashimoto, Takashi Aoyama, Yuichi Matsuo, Makoto Ueno, Kazuyuki Nakakita, Shigeru Hamamoto, Keisuke Sawada, Kisa Matsushima, Taro Imamura, Akio Ochi, and Minoru Yoshimoto. "Summary of First Aerodynamics Prediction Challenge (APC-I)", 54th AIAA Aerospace Sciences Meeting, AIAA SciTech, (AIAA 2016-1780).

Solver methodsGoverning Equations RANS

Spatial Discretization Cell-centerd finite volume methodwith 2nd-order accurate reconstruction based on MUSCL

Inviscid Flux SLAU (Simple Low-dissipation AUSM scheme)

Viscous Flux 2nd-order accurate central difference

Time Integration MFGS implicit method with local time stepping

Turbulence Model SA-noft2


Case 1a, 1b– HL-CRM (full gap)

Grid Convergence

HL-CRM full gap, CflowGrid (Medium), AoA=8[deg], Total Pressure ratio


Grid size

3 sizes of grid is computed for both B3 and Cflow Grid.

Cflow

B3

0100200300400500

Coarse Medium Fine

Num

ber o

f Cel

ls [x

106 ]

Grid Level

Cflow(Case 1b)

B3(Case 1a)

Coarse Medium FineSectional Grid @ η=0.552

HL-CRMfull gap

Cflow Grid is flow-adapted.


Grid convergence – CD-CDi, CL

1.5

1.6

1.7

1.8

1.9

2.0

2.1

2.2

2.3

2.4

2.5

0.0E+00 5.0E-06 1.0E-05 1.5E-05 2.0E-05CL

1/(N^2/3)

CL

AoA=16[deg]

AoA=8[deg]

B3Cflow

Cflow

B3

0.0500

0.0550

0.0600

0.0650

0.0700

0.0750

0.0800

0.0850

0.0900

0.0950

0.1000

0.0E+00 5.0E-06 1.0E-05 1.5E-05 2.0E-05

CD-C

Di

1/(N^2/3)

CD-CDi

AoA=16[deg]

AoA=8[deg]

* CDi=CL2/(πAe), A=9.0, e=1

Cflow

B3

Cflow

B3

Fine Medium Coarse

65 cnts

30 cnts

CflowGrid has lower drag and higher lift than B3.

Discrepancy between B3 and Cflow seems to come from pressure drag (next page).

HL-CRMfull gap


0.0400

0.0450

0.0500

0.0550

0.0600

0.0650

0.0700

0.0750

0.0800

0.0850

0.0900

0.0E+00 5.0E-06 1.0E-05 1.5E-05 2.0E-05CD

p -C

Di

1/(N^2/3)

0.0000

0.0050

0.0100

0.0150

0.0200

0.0250

0.0300

0.0350

0.0400

0.0450

0.0500

0.0E+00 5.0E-06 1.0E-05 1.5E-05 2.0E-05

CDv

1/(N^2/3)

Grid convergence – CDv, CDp

Friction drag coefficient (CDv) and pressure drag coefficient (CDp).

Difference between B3 and Cflowon skin frinction (CDv) is small relative to that of pressure drag.

CDv CDp-CDi

Cflow B3

solid line : AoA=8[deg]dashed line: AoA=16[deg]

Difference of CDp - CDiis dominant to the discrepancy of CD-CDi.

Cflow

B3

Cflow

B3

AoA=16[deg]

AoA=8[deg]

HL-CRMfull gap

65 cnts

30 cnts


Surface streamlines @ AoA=16[deg]Surface : Cf Cflow (Coarse)B3 (Fine)

Surface Grid

HL-CRMfull gap

Separation near flap gap and outer flap edge is larger than Cflow.

Grid around flap gap and outer flap edge is coarse.

Wing surface mesh density is similar to each other.


Case 2a-2d – JSM (Nacelle ON/OFF)


Computational mesh

JAXA Grid

@Y=-0.80 @Y=-0.80

Cflow Grid

JSMNacelle-ON

Unstructure Mixed Unstructure Hexahedra


Lift curve

1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

2.8

3.0

0 5 10 15 20 25

CL

AoA[deg]

JAXA Grid Cflow Grid

1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

2.8

3.0

0 5 10 15 20 25CL

AoA[deg]

1.21.41.61.82.02.22.42.62.83.0

0 5 10 15 20 25

CL

AoA[deg]

WTT (Naccelle ON) - corrected

WTT (Nacelle OFF) - corrected

CFD (Nacelle-ON)

CFD (Nacelle-OFF)

Mach 0.172

Re[-] 1.93 x 106

CL of CflowGrid drops ealierthan JAXA grid.

JSMNacelle-OFF

JSMNacelle-ON


Surface flow AoA=10.47[deg]

JAXA Grid

12

3

45

67

8

12

3

45

67

8

Skin friction coefficient Cf

EXP

JAXA Grid

Cflow Grid

Cflow Grid

EXP

JSMNacelle-OFF

Oilflow

Separation at downstream side of #8 bracket.

12

3

45

67

8

Slat Bracket No.


Surface flow AoA=18.58[deg]

EXP

12

3

45

67

8

12

3

45

67

8

EXP

JAXA Grid

Cflow Grid

JAXA Grid

Skin friction coefficient Cf

Cflow Grid

Largely separated at downstream side of #6 bracket.

JSMNacelle-OFF

Oilflow

12

3

45

67

8


Wing surface mesh


Wing surface mesh density is similar to JAXA Grid.

12

3

45

67

8

12

3

45

67

8

Surface mesh is finer than JAXA Grid around wing LE.

Lower surface

Upper surface


Total pressure ratio @ No.6 bracket


Y-const. plane

No.6 bracket

Wing upper surface

Wing lower surface

Total pressure ratio0.999

0.97

Volume mesh density is finer than JAXA Grid.

Volume mesh density is finer than JAXA Grid.


Effects of initial condition

Effects of initial condition were investigated at AoA greater than 18.54[deg]. Two results are compared.

• Impulsive start• start from the results of AoA=14.54[deg].

1.2

1.4

1.6

1.8

2

2.2

2.4

2.6

2.8

3

0 5 10 15 20 25

CL

AoA[deg]

EXP

JAXA Grid (start from AoA=14.54[deg])

Cflow Grid (start from AoA=14.54[deg])

JSMNacelle-OFF

JAXA Grid (Impulsive)

Cflow Grid (Impulsive)

JAXA Grid CL was greatly improved.

Cflow Grid CL was much lower than impulsive start.


Effects of initial condition on surface flow

ImpulsiveStart

12

34

5

67

8

12

34

5

67

8

EXP

12

34

5

67

8

12

34

5

67

8

12

34

56

78

Separation was supressedin JAXA Grid.

Flow was separated at another bracket in Cflow Grid.


Start fromAoA=14.54[deg]

JSMNacelle-OFF


Summary and Future Works

HL-CRM full gap (Grid convergence) case Results of Cflow Solver using B3 Grid and Cflow Grid were

presented. Grid convergence was obtained in both grids. B3 grid had higher drag than Cflow. Lower grid density of B3

near flap gap and outer flap edge might induce flow separation.

JSM case Results of Cflow Solver using JAXA Grid and Cflow Grid for both

Nacelle-ON/OFF configurations were presented. CL of Cflow Grid droped earlier than experiment. This was

caused by separation near wing tip. Effects of initial condition was investigated. Starting lower AoA

result improved flow field in JAXA Grid but not in Cflow Grid. The separation issue seems to be grid dependency. This detail

causes will be investigated in future work.


Case 1a,b – HL-CRM

Grid Convergence


Grid Convergence – CL, Cm

-0.400

-0.395

-0.390

-0.385

-0.380

-0.375

-0.370

-0.365

-0.360

-0.355

-0.350

0.0E+00 5.0E-06 1.0E-05 1.5E-05 2.0E-05Cm

1/(N^2/3)

1.5

1.6

1.7

1.8

1.9

2.0

2.1

2.2

2.3

2.4

2.5

0.0E+00 5.0E-06 1.0E-05 1.5E-05 2.0E-05

CL

1/(N^2/3)

CL Cm

AoA=16[deg]

AoA=8[deg] AoA=16[deg]

AoA=8[deg]B3

Cflow

Cflow

B3

B3Cflow

B3

Cflow

HL-CRMfull gap


Grid Convergence – CD, CD-CDi

0.1600

0.1800

0.2000

0.2200

0.2400

0.2600

0.2800

0.3000

0.0E+00 5.0E-06 1.0E-05 1.5E-05 2.0E-05

CD

1/(N^2/3)

0.0500

0.0550

0.0600

0.0650

0.0700

0.0750

0.0800

0.0850

0.0900

0.0950

0.1000

0.0E+00 5.0E-06 1.0E-05 1.5E-05 2.0E-05CD

-CDi

1/(N^2/3)

CD CD-CDi

AoA=16[deg]

AoA=8[deg]AoA=8[deg]

AoA=16[deg]

* CDi=CL2/(πAe), A=9.0, e=1

Cflow B3

Cflow

B3

Cflow

B3

Cflow

B3

HL-CRMfull gap


Total Pressure Ratio @ AoA=16[deg]

Cflow

B3

Coarse Medium Fine

Artificial total pressure loss decreases in finer mesh.

HL-CRMfull gap

Separation at wing root seems large relative to Cflow Grid.


Surface streamlines with Cp @ AoA=16[deg]

Cflow

B3

Coarse Medium Fine


Surface streamlines with Cf @ AoA=16[deg]

Cflow

B3

Coarse Medium Fine

Cf


Total Pressure Ratio @ α=16°(CflowGrid, Medium)

eta=0.552

HL-CRMfull gap


Total Pressure Ratio @ α=16°(B3, Medium)

eta=0.552

Wake seems to be disipated around the flap upper surface.

HL-CRMfull gap


Surface Cp around Flap Gap (Medium)

B3 CflowGrid

HL-CRMfull gap


Surface Cp around Root Gap (Medium)

B3 CflowGrid

HL-CRMfull gap


Case 1c,d – HL-CRM

Partially Sealed Flap


-0.0070

-0.0060

-0.0050

-0.0040

-0.0030

-0.0020

-0.0010

0.0000

8 16

ΔCDp

AoA[deg]

Δ(gapped)

Δ(sealed)

Δ = (Cflow Grid) – (B3 Grid), Medium

0.000

0.005

0.010

0.015

0.020

8 16

ΔCL

AoA[deg]

Δ(gapped)

Δ(sealed)

-0.007

-0.006

-0.005

-0.004

-0.003

-0.002

-0.001

0.000

0.001

8 16

ΔCm

AoA[deg]

Δ(gapped)

Δ(sealed)

black line : ΔＣＤ (Total)

HL-CRMpartial seal


Δ = (gapped-flaps) – (sealed-flaps), Medium

0.000

0.010

0.020

0.030

0.040

0.050

B3 CflowGrid B3 CflowGrid

8 16

ΔCL

AoA[deg]

0.0000

0.0010

0.0020

0.0030

0.0040

0.0050


8 16AoA[deg]

0.0000

0.0010

0.0020

0.0030

0.0040

0.0050


8 16

ΔCDp

AoA[deg]

-0.015

-0.010

-0.005

0.000

0.005


8 16

ΔCm

AoA[deg]

black line : ΔＣＤ (Total)

HL-CRMpartial seal


Case 2a-d JSM


CL - CD

1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

2.8

3.0

0.1000 0.2000 0.3000 0.4000 0.5000 0.6000

CL

AoA[deg]

D-JSM_UnstrMixed_JAXA

1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

2.8

3.0

0.1000 0.2000 0.3000 0.4000 0.5000 0.6000CL

CD

1.21.41.61.82.02.22.42.62.83.0

0 5 10 15 20 25

CL

AoA[deg]



CFD (Nacelle-ON)

CFD (Nacelle-OFF)

CflowGrid

Mach 0.172

Re[-] 1.93 x 106


CL - Cm

D-JSM_UnstrMixed_JAXA

1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

2.8

3.0

-0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0

CL

Cm

1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

2.8

3.0

-0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0CL

Cm

1.21.41.61.82.02.22.42.62.83.0

0 5 10 15 20 25

CL

AoA[deg]



CFD (Nacelle-ON)

CFD (Nacelle-OFF)

CflowGrid

Mach 0.172

Re[-] 1.93 x 106


Case 3 – 2D Verification Case


Grid Convergence – Cl, Cd

Cl Cd

Cdp Cdv

Finer

5 counts

5 counts

1 count

Cflow

Cflow

CflowCflow


Wake (U-velocity )5th Finest 4th Finest 3rd Finest

2nd Finest Finest

All CFL3D results are Finest Grid.

008 - KHI Contribution to HiLiftPW-3008 - KHI Contribution to HiLiftPW-3 Hidemasa Yasuda *, Taku...

Documents

Transcript of 008 - KHI Contribution to HiLiftPW-3008 - KHI Contribution to HiLiftPW-3 Hidemasa Yasuda *, Taku...