optimized off-design performance of flexible wings with continuous trailing edge … · 2015. 2....

Optimized Off-Design Performance of Flexible Wings with Continuous Trailing-Edge Flaps

!

!!!

David L. Rodriguez Michael Aftosmis Marian Nemec George Anderson

!Applied Modeling and Simulation Branch, NASA Ames Research Center

!!

AMS Seminar Series - February 17, 2015

3/2/15 DR

Modern Wing Structures Technology

• Many new concepts have high aspect ratio, light, very-flexible, composite wings

• Wing shape varies greatly throughout mission profile

• Boeing 787 wing tips deflect 10 feet at cruise!

2

3/2/15 DR

Wing Morphing Technology• Wing morphing has been used

since the beginning of human flight

• Basic concept is to actively reshape the wing in flight to improve performance and/or control

• One concept currently being researched is the VCCTEF

3

3/2/15 DR

Goal: Evaluate VCCTEF Concept• Variable Camber Continuous Trailing Edge Flaps

• Evaluate the maximum potential benefit of VCCTEF on a generic transport model (GTM) aircraft at cruise • aerodynamic evaluation sufficient

• other design features neglected or held constant (structural weight and layout, trim, actuator weight, viscous effects)

• for simplicity, work with wing and fuselage only

• Must include aeroelastic effects in analysis • conventionally stiff wing

• modern, highly flexible wing

• Develop methodology for designing (optimizing) transport wings while addressing aeroelastic effects

4

3/2/15 DR

VCCTEF Layout on GTM• Flaps over most of the span of the wing

• 1 large inboard flap, 14 smaller outboard flaps, 1 aileron

• 3 segments per flap (camber)

• Elastomer material between flaps to seal gaps

• Tailors spanwise lift distribution throughout mission

5

3/2/15 DR

Modeling the VCCTEF

6

3/2/15 DR

Modeling the VCCTEF

7

• Flap deflections controlled by Blender “armature,” which is analagous to a skeleton

• Surface triangulation is bound to “bones”

• Bones can only rotate about hinge lines

• Sequential flaps bones linked to each other

• Blended transition between flaps to mimic elastomer material

3/2/15 DR

Modeling the VCCTEF with Blender

8

3/2/15 DR

Goal: Evaluate VCCTEF Concept• Variable Camber Continuous Trailing Edge Flaps

• Evaluate the maximum potential benefit of VCCTEF on a generic transport model (GTM) aircraft at cruise • aerodynamic evaluation sufficient

• other design features neglected or held constant (structural weight and layout, trim, actuator weight, viscous effects)

• for simplicity, work with wing and fuselage only

• Must include aeroelastic effects in analysis • conventionally “stiff” wing

• modern, highly flexible, “soft” wing

• Develop methodology for designing (optimizing) transport wings while addressing aeroelastic effects

9

3/2/15 DR

Aerodynamic Analysis Load Transfer

Structural Analysis Deform Geometry

Deformed GeometryConverged?

Stop

Yes

No

Baseline Geometry

AeroelasticIteration

Deformer

AeroelasticAnalysis

Static Aeroelastic Analysis Architecture

10

Cart3D BEAM Blender

3/2/15 DR

Aerodynamic Shape Optimization Architecture

11

AeroelasticAnalysis

Aeroelastic Deformation Converged?

Stop

Optimized Undeformed Geometry

Aerodynamic Optimization

Baseline Undeformed Geometry

Yes

No

3/2/15 DR

Performance of Optimization Method• Alternating aeroelastic

analyses and aerodynamic optimizations • aeroelastic analysis required

5 iterations

• typical optimization required 60-80 design space samples

• Typical aeroelastic optimization • converges in 3-4 iterations

• one iteration ≈ 1 day of wall clock time on 64-cpus of endeavour

12

Optimization Iteration

Tip

Defle

ctio

n (fe

et)

Drag

Coe

ffici

ent

0 1 2 30

1

2

3

4

5

6

7

8

0.0146

0.0148

0.0150

0.0152

0.0154

0.0156

0.0158

0.0160

0.0162

AeroelasticAnalysis

Aeroelastic Deformation Converged?

Stop

Optimized Undeformed Geometry

Aerodynamic Optimization

Baseline Undeformed Geometry

Yes

No

3/2/15 DR

Problem Setup

1. Establish a new baseline for mid-cruise • aerodynamic optimization of GTM wing

with VCCTEF

• include aeroelastic effects

• disregard other disciplines

2. Re-design for off-design • repeat optimization at begin and end cruise

• determines best possible performance at off-design conditions

3. Adapt flap system on baseline for off-design • optimize only flaps while maintaining baseline twist

• compare results with best possible performance from step 2

13

Takeoff Landing

Climb Descent

Cruise @ 36,000 feet

Mach 0.797

Loiter

Mid-Cruise

50% fuel

3/2/15 DR

Problem Setup


with VCCTEF







14

Takeoff Landing

Climb Descent


Mach 0.797

Loiter

Mid-Cruise

50% fuelBegin-Cruise

80% fuelEnd-Cruise

20% Fuel

3/2/15 DR

Problem Setup


with VCCTEF







15

Takeoff Landing

Climb Descent


Mach 0.797

Loiter

Mid-Cruise

50% fuelBegin-Cruise

80% fuelEnd-Cruise

20% Fuel

3/2/15 DR

Load Distributions

16

Verti

cal L

oad

(pou

nds/

foot

)

-1500

-1000

-500

0

500

Distance Along Elastic Axis (feet)0 10 20 30 40 50 60 70

Stiff Wing StructureSoft Wing StructureFuel @ Begin-CruiseFuel @ Mid-CruiseFuel @ End-Cruise

Center Fuel Tank Outboard Fuel Tank

EngineElastic Axis

3/2/15 DR

Design Optimization Problem

17

• Minimize total drag (inviscid)

• Lift is held constant

• Design variables include wing twist and VCCTEF deflections

3/2/15 DR

Design Variables

18

• Angle of attack

• Wing twist distribution • modeled as perturbation to original

• Blender module

• VCCTEF deflections • circular deflection

• Bernstein polynomials

• inboard flap and aileron separate

3/2/15 DR

Design Variables• Angle of attack


• Blender module

• VCCTEF deflections • circular deflection

• Bernstein polynomials


19

fixed root section

twist perturbation stations

3/2/15 DR

• Angle of attack


• Blender module

• VCCTEF deflections • link segments via “circular deflection”

• Bernstein polynomials for outboard flaps


Design Variables

20

Inboard Flap

Aileron

Outboard Flaps

∆2 = 2∆1, ∆3 = 3∆1

∆3 = 15°∆2 = 10°∆1 = 5°

Flap

Defl

ectio

n /

Unit

Desig

n Va

riabl

e

0.0

0.2

0.4

0.6

0.8

1.0

Flap Number (Outboard)1 2 3 4 5 6 7 8 9 10 11 12 13 14

B1B2 B3

B4

3/2/15 DR

Stiff Wing Optimization and Analysis (Mid-Cruise)

• Start with original GTM

• Optimize twist

• Optimize flaps (fixed twist)

• Optimize twist and flaps

21

Spanwise Fraction

Sect

ion

Lift

0 0.2 0.4 0.6 0.8 10

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Elliptic DistributionOriginal GTMOptimized TwistOptimized FlapsOptimized Twist+Flaps

Invi

scid

Dra

g (co

unts

) Original GTM 124.7

Optimize Twist & Flaps

3/2/15 DR



• Optimize twist



22

Spanwise Fraction

Sect

ion

Lift

0 0.2 0.4 0.6 0.8 10

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8


Invi

scid

Dra

g (co

unts

) Original GTMOptimize

Twist

124.7123.8


3/2/15 DR



• Optimize twist



23

Spanwise Fraction

Sect

ion

Lift

0 0.2 0.4 0.6 0.8 10

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8


Invi

scid

Dra

g (co

unts

) Original GTM

120.4

Optimize Twist

124.7123.8

Optimize Flaps


3/2/15 DR



• Optimize twist



24

Spanwise Fraction

Sect

ion

Lift

0 0.2 0.4 0.6 0.8 10

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8


Invi

scid

Dra

g (co

unts

) Original GTM

120.4

Optimize Twist

124.7123.8

120.0

Optimize Flaps


3/2/15 DR

Stiff Wing GTM - Mid-Cruise Optimized

25

Cha

nge

in T

wis

t (de

gree

s)

-5-4-3-2-1012345

Spanwise Location (feet)0 10 20 30 40 50 60 70

Flap

Defl

ectio

n (d

egre

es)

-0.6

-0.5

-0.4

-0.3

-0.2

-0.1

0

Inboard Flap Outboard Flaps Aileron

3/2/15 DR

Stiff Wing Off-Design Analysis and Optimization

• Analyze wing optimized for mid-cruise at off-design conditions • begin-cruise (80% max fuel)

• end-cruise (20% max fuel)

• Re-optimize wing for the off-design conditions

• Quantify penalty for flying mid-cruise optimized wing at off-design

26

50

100

150

200

Original GTM Mid-Cruise Opt. Begin-Cruise Opt.

148.7149.9156.2

Stiff Wing GTM CL = 0.565!80% fuel

Invi

scid

Dra

g C

ount

s

50

100

150

200

Original GTM Mid-Cruise Opt. End-Cruise Opt.

96.197.499.8

Stiff Wing GTM CL = 0.455!20% fuel

Invi

scid

Dra

g C

ount

s

3/2/15 DR

Stiff GTM Wing - VCCTEF Adaptation• Start with wing

designed for mid-cruise

27

Begin-Cruise

End-Cruise

Mid-Cruise Design

Invi

scid

Dra

g

(counts

)

149.9

Invi

scid

Dra

g

(co

unts

) Mid-Cruise Design 97.4

Best Possible

3/2/15 DR



• Compare with wing optimized for off-design condition

28

Begin-Cruise

End-Cruise

Mid-Cruise Design

Invi

scid

Dra

g

(counts

)

148.7

149.9

Best Possible

Invi

scid

Dra

g

(co

unts


96.1Best Possible

3/2/15 DR




• Optimize VCCTEF deflections to recover lost performance

• Improvement in both cases

29

Begin-Cruise

End-Cruise

Mid-Cruise Design

Invi

scid

Dra

g

(counts

)

148.7

149.9

149.2Optimize Flaps

Best Possible

AdaptedVCCTEF

Invi

scid

Dra

g

(co

unts

)

Optimize Flaps

Mid-Cruise Design 97.4

96.1Best Possible96.2

AdaptedVCCTEF

3/2/15 DR

Soft Wing Definition• Stiff wing used structural model similar to that of actual

transport

• Soft wing is defined by halving the bending and torsional stiffness distribution of stiff wing

30

3/2/15 DR

Soft Wing Optimization and Analysis (Mid-Cruise)


• Optimize twist



31

Spanwise Fraction

Sect

ion

Lift

0 0.2 0.4 0.6 0.8 10

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8


Invi

scid

Dra

g (co

unts

)


Original GTM123.4

3/2/15 DR



• Optimize twist



32

Spanwise Fraction

Sect

ion

Lift

0 0.2 0.4 0.6 0.8 10

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8


Invi

scid

Dra

g (co

unts

)


Original GTM

Optimize Twist

123.4

118.2

3/2/15 DR



• Optimize twist



33

Spanwise Fraction

Sect

ion

Lift

0 0.2 0.4 0.6 0.8 10

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8


Invi

scid

Dra

g (co

unts

)


Original GTM

114.9

Optimize Twist

123.4

118.2

Optimize Flaps

3/2/15 DR



• Optimize twist



34

Invi

scid

Dra

g (co

unts

)


Original GTM

114.9

Optimize Twist

123.4

118.2

114.5

Optimize Flaps

Spanwise Fraction

Sect

ion

Lift

0 0.2 0.4 0.6 0.8 10

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8


3/2/15 DR

Flap

Defl

ectio

n (d

egre

es)

-0.6

-0.5

-0.4

-0.3

-0.2

-0.1

0

Inboard Flap Outboard Flaps Aileron

Cha

nge

in T

wis

t (de

gree

s)

-5-4-3-2-1012345

Spanwise Location (feet)0 10 20 30 40 50 60 70

Soft Wing GTM - Mid-Cruise Optimized

35

3/2/15 DR

Soft Wing Off-Design Analysis and Optimization

• Analyze wing optimized for mid-cruise at off-design conditions • begin-cruise (80% max fuel)

• end-cruise (20% max fuel)

• Re-optimize wing for the off-design conditions

• Quantify penalty for flying mid-cruise optimized wing at off-design

36

50

100

150

200

Original GTM Mid-Cruise Opt. Begin-Cruise Opt.

143.1143.5154.5

Soft Wing GTM CL = 0.552!80% fuel

Invi

scid

Dra

g C

ount

s

50

100

150

200

Original GTM Mid-Cruise Opt. End-Cruise Opt.

91.993.7100.6

Soft Wing GTM CL = 0.442!20% fuel

Invi

scid

Dra

g C

ount

s

3/2/15 DR

Soft GTM Wing - VCCTEF Adaptation• Start with wing


37

Invi

scid

Dra

g

(co

unts


Invi

scid

Dra

g

(counts


91.9

Begin-Cruise

End-Cruise

3/2/15 DR




38

Invi

scid

Dra

g

(counts


91.9Best Possible

Invi

scid

Dra

g

(co

unts

) Mid-Cruise Design 143.5Best Possible

143.1

Begin-Cruise

End-Cruise

3/2/15 DR




• Optimize VCCTEF deflections to recover lost performance

• Improvement in both cases

39

Begin-Cruise

Invi

scid

Dra

g

(co

unts

)

Optimize Flaps

Mid-Cruise Design 143.5Best Possible

143.1AdaptedVCCTEF

End-CruiseIn

visc

id D

rag

(counts

)

Optimize Flaps

Mid-Cruise Design 93.7

91.9Best Possible 92.1

AdaptedVCCTEF

3/2/15 DR

Conclusions• Fast iterative method developed to aerodynamically optimize

transport wings while addressing aeroelastic effects • VCCTEF system was evaluated on GTM wing (stiff and soft) as a

means to improve off-design cruise performance • achieved near optimal performance

• results suggest wave drag could be actively reduced

• flap system could reshape a wing with constant airfoil section (ease of manufacturing) to a more optimal design for any given flight condition

• Results similar on conventional (stiff) and highly flexible (soft) wings • Designer of an aircraft with VCCTEF could assume near-optimal

performance throughout cruise

40

3/2/15 DR

Future Work• Another off-design case - over-speed

• another common off-design case is flying faster (to keep a schedule)

• can the VCCTEF improve cruise performance at a higher Mach number?

• repeat evaluation at Mach 0.827 at mid-cruise

• Evaluate VCCTEF on other transport aircraft designs • Truss-Braced Wing

• Common Research Model (higher aspect ratio)

41

optimized off-design performance of flexible wings with continuous trailing edge … · 2015. 2....

Documents