Application of Smart Materials for Adaptive Airfoil Shape Control

American Institute of Aeronautics and Astronautics092407

1

Application of Smart Materials for Adaptive Airfoil Control

Ermira J. Abdullah*, Cees Bil† and Simon Watkins.‡

School of Aerospace, Mechanical and Manufacturing Engineering, Royal Melbourne Institute ofTechnology, Melbourne, Australia.

The development of adaptive airfoil control can potentially improve flight performanceby optimizing the maximum lift-to-drag ratio throughout all flight regimes. Improved flightperformance translates into weight and fuel savings. Smart material is a suitable candidatefor adaptive airfoil design as it can be activated to alter the shape of the airfoil. This paperpresents an overview of smart material application for adaptive airfoil control with focus onshape memory alloy actuator and flexible skin for variable camber airfoil. Thisinvestigation is the first step in the development of an adaptive airfoil that is able to beutilized on an Uninhibited Aerial Vehicle. The influence of changing airfoil’s maximumcamber on aerodynamic performance was also explored and the analysis is presented.

Nomenclatureα = angle of attackCD = drag coefficientCDmin = minimum drag coefficientCL = lift coefficientCLmax = maximum lift coefficientD = drag force∆ = changeδ = deflection of trailing edgeL = lift forceL/D = lift-to-drag ratioL/Dmax = maximum lift-to-drag ratioM = Mach numberq = dynamic pressureRe = Reynolds numberS = wing planform area

I. Introductionne of the most important considerations in airline operation is aircraft efficiency. Fuel costs can approach up to50 percent of airline operating expense for some modern, wide-body, long-range transports.1 A 3-percent

reduction of fuel consumption can produce savings of as much as $300,000 per year for each aircraft.2 An analyticalstudy conducted by NASA on the benefits of variable-camber capability reveals that drag can be significantlyreduced if all wing trailing edge surfaces are available for optimization such as in the case of flight with variablecamber capability. 1

Due to the potential benefits of employing adaptive airfoil, there has been an intensive attempt by researchers indeveloping a working model. With the advancement of materials, many are now considering using smart materials

* Graduate Research Student, School of Aerospace, Mechanical and Manufacturing Engineering, RMIT, Melbourne,VIC, Australia.† Associate Professor, School of Aerospace, Mechanical and Manufacturing Engineering, RMIT, Melbourne, VIC,Australia and Senior Member AIAA.‡ Professor, School of Aerospace, Mechanical and Manufacturing Engineering, RMIT, Melbourne, VIC, Australia.

O

47th AIAA Aerospace Sciences Meeting Including The New Horizons Forum and Aerospace Exposition5 - 8 January 2009, Orlando, Florida

AIAA 2009-1359

Copyright © 2009 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved.


2

to produce airfoil with variable camber capability. Campanille et al developed an active camber control using thebelt-rib concept which allows camber changes within given limits while the in-plane stiffness properties of thesection remain generally unchanged.3 The proposed model is actuated mechanically by cables, but with thedevelopment of smart material, it was suggested to be replaced by multifunctional actuator such as shape memorywire.

Application of shape memory alloy spring as an actuator for an adaptive airfoil has been tested.4 SMA springswith the help of stop structures are used to actuate accurately certain points on the skins to approach the targetairfoil. From the simulation and measured results, it was discovered that the skin actuated by SMA springs onspecific discrete points could obtain good actuating results near these points. There were errors between simulationvalue, measured value and target value, at the positions far away from the points actuated. The error was the biggestat points which are far away from both the actuated points and the constraint points, caused by the differencebetween the successive deformation character of rigid body and the singular character of the target shape. Thismeans average distribution of actuated points along the chord was favourable to approach the target shape better.

Hutapea et al has developed a prototype of a smart actuation system for an adaptive airfoil by controlling theflaps.5 SMA springs were fixed at one end to the wing box toward the leading edge of the airfoil while the other endwas attached tangentially to a rotating cylinder fixed to the flap. In order to produce rotation of the flap in both theupward and downward directions, the springs were arranged in an upper and a lower layer. An applied current wasused to produce heat which controlled the spring actuators. The prototype developed demonstrated strong potentialfor future application based on the experimental and theoretical analysis.

The objective of this research is to develop an adaptive airfoil control system using shape memory alloyactuators which will be implemented on a UAV. At this stage, the focus is to investigate the use of smart material asactuator and flexible material for the skin. The potential benefits of utilizing adaptive airfoil control for UAV is alsoexamined by investigating the effects of changing the airfoil camber on the aerodynamic performance, in particularthe lift to drag ratio.

II. Shape Memory Alloy ActuatorsSmart material can be tailored to create a specific response to a combination of inputs.6 These materials include

piezoelectrics and electrostrictives, and shape memory alloy. In the case of adaptive airfoil, Fontanazza et alconcluded that the ideal material should respond quickly to external stimuli, be capable of large and recoverable freestrains, transform effectively the input energy to mechanical energy, and not be affected by fatigue issues.7 Theysuggested that the benefits of using smart material compared to pneumatic or hydraulic actuators are reducedcomplexity and improved reliability of the system.

Table 1 lists the most common characteristics of some smart materials which include maximum free strain,maximum stress, deformation energy density, efficiency, and relative speed of response.8 Among all the smartmaterials, SMAs appear to have superior capability in producing large plastic deformations. In recent years, interestin SMAs applications for adaptive structures has been increasing not only due to this unique quality, but alsobecause of their high power-to-weight ratio and low driving voltages. SMAs are thermomechanical materialstypically comprise of a mixture of nickel and titanium, which changes shape when heated or cooled.6 When theyare cooled to below a critical temperature their crystal structure enters the martensitic phase, where alloy is plasticand can easily be manipulated through very large strain ranges with little change in the material stress. However,when heated, above the critical temperature, the phase changes to the austenitic phase, where the alloy resumes theshape that it formally had at the higher temperature.

Table 1 The characteristics of smart materials.8

Material Max. strain Max. stress Elastic energy Max. effic. Relative(%) (MPa) density(J/g) (%) speed

ElectrostrictorPolymer P (VDF-TrFE) 4 15 0.17 - FastPiezoelectric Ceramic (PZT) 0.2 110 0.013 >90 FastSingle Crystal (PZN-PT) 1.7 131 0.13 >90 FastPolymer (PVDF) 0.1 4.8 0.0013 n/a FastSMA (TiNi) >5 >200 >15 <10 Slow


3

Nickel titanium is the most commonly used SMA to which copper is sometimes added to aid in the strainrecovery process. The process of shape change or creating movement comprise of a five-step procedure that occurswithin the material in which the shape memory effect is developed. Figure 1 shows the entire process.9,10 The firststep is the parent austenitic phase which occurs at a high temperature with zero stress and strain. In order to createtwinned martensite, the parent austenitic structure is cooled in the absence of both stress and strain. Next, thetwinning process is reversed by stressing the material which causes the now detwinned martensite to developinelastic strains. While still maintaining its detwinned form with the elastic strain, the load is then released. Finally,the material returns to its original shape and composition when all inelastic strains are recovered by heating theSMA to its parent austenitic start temperature.

Figure 1. Schematic of temperature-stress-strain for SMA crystallographic phasetransformation.9, 10

There are some drawbacks in using SMAs such as nonlinear response of the strain to input current and hysteresischaracteristic as a result of which their control is inaccurate and complicated. The accuracy of the mathematicalmodel is critical as the efficiency of an SMA actuator depends on the preciseness of its control. Due to thecomplexity of modelling SMA actuators, there has been a number of studies dedicated to modelling and control ofthe SMA actuator.11-15 The methods proposed to reduce the complexity include continuous-time model which fitsdifferential equation to experimental data, the Preisach model which is used to model the hysteresis, feedbacklinearization and variable structure control. However each of this method has its limitations.

It is a daunting task to achieve precise control by using feedback of temperature, resistance and so forth eventhough some constitutive models can represent the mechanics behaviour of shape memory alloy under the conditionof multifield coupling. For the adaptive airfoil with SMA springs, a simple locating structure can accomplish preciseposition control by combining electric and mechanism methods and the loss of part of the actuating or deformationability of the SMA springs.4

Another method of modelling SMA is using the physics of the process where the Fermi–Dirac statistical modelis used to represent the two-state process.15 Based on this model, two controllers were developed and implementedexperimentally: a gain-scheduled controller based on LQR optimization and a loop-shaping controller. Thesimulation for the gain-scheduled PI controller and SMA shown in Fig. 2 was carried out using MATLAB. Fig. 3shows an accurate tracking of the reference for an input consisting of step. In this model the rate of heating of theSMA can be controlled while the rate of cooling of the SMA wire, which happens through natural convection,cannot be controlled. The simulation and experimental results demonstrate show excellent tracking response for theSMA without the presence of perturbations, thus validating both the model and the control scheme. The resultsproved that the model for describing the transformation between martensite and austenite phases is viable.


4

Figure 2. Controller of the SMA actuator.15

Figure 3. Closed-loop SMA response to a step reference input.15

III. Candidate for Skin MaterialThill et al presented a comprehensive overview of morphing concepts with an emphasis on morphing skin.16

According to the study, in spite of numerous morphing aircraft concepts being put forth, limited focus has beengiven to the problems relating to the skin of morphing structure which requires smooth and continuous deformationwhile carrying load. The task of combining properties like flexibility and stiffness in one structure is proving to bea very complex process. Anisotropic and variable stiffness structures have the potential to alter shape and producesmall increase in area of the wing. Stiffness in the chordwise direction can either be tailored or actively controlledto obtain desirable shape changes. Some possible structures for morphing wing include corrugated structures,reinforced elastometers or flexible matrix composite tubes embedded in a low modulus membrane.

A highly anisotropic skin is the ideal skin for a morphing aircraft according to Gandhi et al because it willprevent skin sections between supports from undergoing local bending deformations and sections of the skin incompressive loading from undergoing buckling.17 It was observed that airfoil with a low skin axial stiffness is ableto camber easily. However, undesirable global camber deformation under the external aerodynamics load will occurif the skin axial stiffness is reduced below a certain limit. So the best solution is to reduce the skin axial stiffness nofurther than the point beyond which the aerodynamics deflections are no longer acceptable. This will produce alarger camber deformation under actuation force.

Selection of a suitable skin material is vital in producing a smooth wing surface when morphing takes place.Different types of skin material will produce different effects on the surface of the morphing wing which leads tovarying aerodynamics performance. For morphing wing, stiffened metallic, typically aluminum alloy, is the mostcommonly use panels as an outer skin. In terms of strength to density ratio, aluminum alloy can be superior to steel,though not to titanium alloy, but it is better than both in respect of stiffness criteria.18

Elastometers or rubbers are a class of polymer with a low density of cross-links with the ability to undergo largeelastic deformations without permanently changing the shape.19 It was employed in the DARPA Smart Wing projectwhere the skin was made of high strain-to-failure silicone, over honeycomb core.20 It performed effectively and metthe program requirements in terms of deflection and shape integrity. However, it was noted that other concept mayprove to be more superior such as a coreless semi-rigid skin, which needs to be investigated.

Acrylonitrile butadiene styrene plastic or ABS is also used as skin material for morphing structures. It is adurable, high strength modelling material that can be machined, sanded, drilled, painted and glued after the model isbuilt. It was used in the second generation prototype wind tunnel model for reconfigurable wing at Texas A&M


5

University College Station.21 Using ABS, allowed the drilling of pressure ports, the adhesion of pressure tubes to theskin, and the sanding of the skin to achieve a smooth aerodynamic surface. In addition to that the thickness could bevaried throughout the model, as oppose to steel skin which had constant thickness. This allowed tailoring of stiffnessfor precise area on the wing. Another example is the Scythe UAV which was developed from SBIR funding.It was constructed with thermoplastic materials utilizing a double monocoque structure with both the structural skinand the large honeycomb cores are made from ABS.§

Research conducted by Reich et al focused on developing an analytical framework for design of a morphing skinthat is able to overcome problems associated with conflicting requirements such as flexibility and stiffness, and atthe same time able to maintain aerodynamic shape with moderate power requirements.22 The proposed solution is adesigner skin capable of spatial flexibility across the wing by creating a composite system made up of a combinationof elastometer or shape memory alloy with smaller reinforcing mechanisms. This type of skin solution has beenemployed in the Northrop Grumman Smart Wing and the NextGen Aeronautics batwing design which featuredpolyurethane membrane skins with internal flexible or mechanize structure as support.

IV. Influence of Shape on AerodynamicLift coefficient (CL) and drag coefficient (CD) are complex functions of profile shape, angle of attack (α), wing

planform (S) Mach number (M) and Reynolds number (Re)1, which can be defined asqSLCL /= (1)

andqSDCD /= . (2)

These functions may be obtained from computation, wind tunnel testing or flight testing. The aerodynamicresults are usually presented as graphs of

)(αfCL = (3)

)(αfCD = (4)

and)( LD CfC = . (5)

Typical curves of these functions for low-speed (no shock wave) flight are shown in Fig. 4 - 6. It can be seen fromthe graphs that the curves of equations (4) and (5) have parabolic shape in the region where the CL variation with αis approximately linear.

The maximum achievable lift to drag ratio (L/D) in cruise flight is a very important performance parameter. Itcan be defined as

DL CCDL // = . (6)

It can be plotted as a function of CL as shown in Fig. 7. In performance optimization, L/D is maximized forall flight cruise conditions.

Figure 4. Change of lift coefficient with angle of attack .1

§ http://www.microuav.com/index.html


6

Figure 5. Change of drag coefficient with angle of attack.1

Figure 6. Schematic of a typical polar of an aircraft.1

Figure 7. E = CL/CD as a function of lift coefficient.1

The change in camber produces varying effect on the aerodynamic performance depending on the modification.In the subsonic region, as camber increases, less α is required for a fixed CL, or CL increases for a constant angle ofattack.23 Increasing camber also increases the linear region of CL as a function of angle of attack, to a larger CL andthe maximum CL:

)()0()(maxmaxmax

δδδ LLL CCC ∆+== (7)

The minimum CD increases by the relation:)()0()(

minminminδδδ DDD CCC ∆+== (8)


7

Different results can be seen for alteration in percent camber and position of maximum camber of the airfoil. Aninitial investigation was carried out to further understand the effect of different configuration of airfoil in order todetermine the best design concept for the adaptive airfoil. Simulation was carried out using XFOIL which is asubsonic computational fluid dynamics software that operates using a modified panel method code to compute bothviscous and inviscid flow about a 2-dimensional body. 24

The analysis was done for a range of angle of attack between 0 degree and 10 degrees at Reynolds number 1 x106. To simplify the process, a symmetrical airfoil NACA 0012 is chosen as the base airfoil. Firstly the value ofmaximum camber was kept constant at 4% while position of maximum camber is varied. Then the value ofmaximum camber was varied while keeping the position of maximum camber at 40% of chord. The analysis wascarried out for Mach number 0, 0.3 and 0.5. Eventually, the placement of the actuator on the airfoil will depend onthe range of shape changes that the airfoil is required to make.

V. Results and DiscussionThe first aerodynamic analysis was carried out for six airfoils with different position of maximum camber. The

position of the airfoil’s maximum camber was varied from 0 to 0.5 of chord. Fig. 8 shows the lift coefficient contourfor angle of attack and change of camber location for different Mach numbers. For Mach numbers 0, 0.3 and 0.5,the lift coefficient increases with angle of attack and Mach numbers. Lift coefficient also increases with increase ofcamber location in percent of chord for angle of attack less than 7, however this limit increases for lower Machnumbers. The airfoil with maximum camber located at 20% of chord has the highest lift coefficient at 10 degreesangle of attack for all Mach numbers. For M = 0, the highest lift coefficient is 1.4748, for M = 0.3, the highest liftcoefficient is 1.5497 and for M = 0.5, the highest lift coefficient is 1.6671.

Figure 8. Lift coefficient contour for angle of attack and change of camber location fordifferent Mach numbers.

Fig. 9 shows the L/D contour for angle of attack and change of camber location for different Mach numbers. Forangle of attack less than 4 degrees, L/D decreases with Mach number but increases with angle of attack. Howeverthis limit increases for lower Mach numbers. As the location of maximum camber increases from 10% to 50% ofchord, L/D increases at low angle of attack. For all Mach numbers L/Dmax increases as the maximum camberlocation increases from 0.1 to 0.5 of chord. The airfoil with maximum camber located at 50% of chord has thehighest value of L/Dmax at 3 degrees angle of attack for all Mach numbers. For M = 0, L/Dmax is 138.363, for M =0.3, L/Dmax is 137.527 and for M = 0.5, L/Dmax is 134.241.


8

Figure 9. L/D contour for angle of attack and increase of camber for different Machnumbers.

Fig. 10 shows the curves for L/D of variable camber airfoil for changes in lift coefficient and maximum camberlocation. L/D increases as maximum camber location increases from 10% to 50% of chord. For the same liftcoefficient and maximum camber, L/D at M = 0 is larger than that of M = 0.5.

Figure 10. L/D plotted against CL for different Mach numbers.

The second aerodynamic analysis was carried out for six airfoils with different percentage of maximum camber.The airfoil’s maximum camber was increased from 0% to 5% of chord. Fig. 11 shows the lift coefficient contour forangle of attack and increase of camber for different Mach numbers. The lift coefficient is linearly proportional toboth angle of attack and maximum camber. For the same maximum camber and angle of attack, the lift coefficientat M = 0.5 is larger than that of M = 0. The airfoil with 5% maximum camber has the highest lift coefficient at 10degrees angle of attack. For M = 0, the highest lift coefficient is 1.5085, for M = 0.3, the highest lift coefficient is1.5273 and for M = 0.5, the highest lift coefficient is 1.533.


9

Figure 11. Lift coefficient contour for angle of attack and increase of camber for differentMach numbers.

Fig. 12 shows the L/D contour for angle of attack and change of camber location for different Mach numbers.For increasing maximum camber, the contours show less α is required for a fixed L/D at low angle of attack or L/Dincreases for a constant angle of attack, α. For the same maximum camber and angle of attack, L/D at M = 0.5 islarger than that of M = 0. The airfoil with 5% maximum camber has the highest value of L/Dmax for all Machnumbers. For M = 0, L/Dmax is 139.5265 at 6 degrees angle of attack, for M = 0.3, L/Dmax is 136.9885 at 5 degreesangle of attack and for M = 0.5, L/Dmax is 129.6215 at 5 degrees angle of attack.

Figure 12. L/D contour for angle of attack and increase of camber for different Machnumbers.

Fig. 13 shows the curves for L/D of variable camber airfoil for changes in lift coefficient and maximum camber.L/D increases with increasing maximum camber. For the same lift coefficient and maximum camber, L/D at M = 0is larger than that of M = 0.5.


10

Figure 13. L/D plotted against CL for different Mach numbers.

It is essential that UAV cruises close to the best lift to drag ratio (L/D) which means flying at constant angle ofattack. In order to maintain flying at the best L/D requires finding a balance between weight, altitude, speed and/orwing area since a failure to do so may cause the L/D to be lower than the best L/D and the range will becorrespondingly less. Variable camber wing may provide a solution to this problem. Adaptive airfoil control allowsthe UAV to change its lift coefficient by changing the camber during cruise in order to operate at optimum L/D forany given lift coefficient and at constant angle of attack. Fig. 10 and 13 can be used to determine the camberneeded to ensure the maximum L/D is achieved during cruise condition which is subject to decreasing liftcoefficient.

VI. ConclusionIn designing an adaptive airfoil control, a few factors have to be taken into consideration. This paper gives a

review of the important elements which includes the smart material for the actuator, flexible material for the skinand the shape of the airfoil. Based on preliminary findings, shape memory alloy is found to be an excellentcandidate for actuator due to its efficiency and large energy storage capacity. Among all SMA available, Nitinol isthe best choice as it is easily available and it is superior compared to other SMA. For the flexible skin, ABSprovides a good solution as it is easy to be manipulated.

Another important element to be considered in the adaptive airfoil control design is the influence of changingairfoil shape on aerodynamic performance. From this analysis, it can be seen that both the value of maximumcamber and its position have significant effect on the L/D ratio. At low angle of attack, L/Dmax increases asmaximum camber position is moved from 0.1 to 0.5 of chord and as maximum camber is increased from 1% to 5%.The results can be used in designing the adaptive airfoil for a range of cruise condition.

The next step in the design process is to develop the wing structure of the UAV that will be implemented withthe adaptive airfoil control system. Once the model is developed, the actuator will be incorporated. Placement of theactuator is critical in obtaining the desired change of the airfoil camber.

References1Bolonkin, A. and Gilyard, G. B., “Estimated Benefits of Variable-Geometry Wing Camber Control for Transport Aircraft,”

NASA TM 1999-206586, 1999.2Gilyard, G. B., Georgie, J., and Barnicki, J. S., “Flight Test of an Adaptive Configuration Optimization System for

Transport Aircraft,” NASA TM-1999-206569, 1999.3Campanille, L. F., and Sachau, D., “The Belt-Rib Concept: A Structronic Approach to Variable Camber,” Journal of

Intelligent Material Systems and Structures, Vol. 11, No. 3, 2000, pp. 215-224.4Dong, Y., Boming, Z., and Jun, L., “A Changeable Aerofoil Actuated by Shape Memory Alloy Springs,” Materials Science

and Engineering A, Vol. 485, 2008, pp. 243–250.5Hutapea, P., Kim, J., Guion, A., Hanna, C., and Heulitt, N., “Development of a Smart Wing,” Aircraft Engineering and

Aerospace Technology: An International Journal, Vol. 80, No. 4, 2008, pp. 439–444.


11

6Culshaw, B., Smart Structures and Materials, Artech House, Inc., Boston, 1996.7Fontanazza, F., Talling, R., Jackson, M., Dashwood, R., Dye D., and Iannucci, L., “Morphing Wing Technologies

Research”, Seas DTC First Conference, 2006.8Bar-Cohen, E., Electroactive Polymer (EAP) Actuators as Artificial Muscles: Reality, Potential, and Challenges, 2nd Edition

Vol. PM136, SPIE Press, Bellingham, WA, 2004.9Nishiyama, Z., Martensitic Transformations, Academic Press, San Diego, CA, 1978.10Kaufman, L., and Cohen, M., “Martensitic Transformations”, Progress in Metal Physics, Vol. 7, 1958, pp. 165-246.11Arai, K., Aramaki, S., and Yanagisawa, K., “Continuous System Modeling of Shape Memory Alloy (SMA) for Control

Analysis,” Proceedings of 5th International Symposium Micro Machine Human Science, 1994, pp. 97–100.12Majima, S., Kodama, K., and Hasegawa, T., “Modeling of Shape Memory Alloy Actuator and Tracking Control System

with the Model,” IEEE Trans. Control Syst. Technol., Vol. 9, No. 1, 2001, pp. 54–59.13Moallem, M., and Lu, J., “Application of Shape Memory Alloy Actuators for Flexure Control: Theory and Experiments,”

IEEE/ASME Trans. Mechatronics, Vol. 10, No. 5, 2005, pp. 495–501.14Elahinia, M., and Ashrafiuon, H., “Nonlinear Control of a Shape Memory Alloy Actuated Manipulator,” Journal of

Vibration Acoustic., Vol. 124, 2002, pp. 566–575.15Jayender, J., “Modelling and Control of Shape Memory Alloy Actuators,” IEEE Transactions on Control Systems

Technology, Vol. 16, No. 2, 2008, pp. 279-287.16Thill, C., Etches, J., Bond, I., Potter, K., and Weaver, P., “Morphing Skins,” The Aeronautical Journal, Vol. 112, No. 1129,

2008, pp. 117-139.17Gandhi, F., and Anusonti-Inthra, P., “Skin Design Studies for Variable Camber Morphing Airfoils,” Smart Materials and

Structures, Vol. 17, 2008, 8pp.18Charles, J. A., Crane, F. A. A., and Furness, J. A. G., Selection and Use of Engineering Materials, Butterworth Heinemann,

3rd Edition, Oxford, 1997.19Callister, W. D. J., Materials Science and Engineering: An Introduction, 7th Edition, Wiley, 2006.20Kudva, J. N., “Overview of DARPA Smart Wing Project,” Journal of Intelligent Material Systems and Structures, Vol.15,

No. 4, 2004, pp. 261-267.21Strelec, J. K., Lagoudas, D.C., Khan, M. A., and Yen, J. "Design and Implementation of a Shape Memory Alloy Actuated

Reconfigurable Wing," Journal of Intelligent Material Systems and Structures, Vol. 14, 2003, pp. 257-273.22Reich, G., W., and Sanders, B., “Development of Skins for Morphing Aircraft Applications via Topology Optimization,”

48th AIAA Structures, Structural Dynamics and Materials Conference, Honolulu, Hawaii, 2007.23McCroskey, W. J., “A Critical Assessment of Wind Tunnel Results for the NACA0012 Airfoil,” NASA Technical

Memorandum 10001 9, USAAVSCOM Technical Report 87-A-5, 1987.24Drela, M., Xfoil 6.8 User Primer, MIT Aero & Astro, 1996.

Application of Smart Materials for Adaptive Airfoil Shape Control

Documents