Aeroelasticity - NASA Good

8/3/2019 Aeroelasticity - NASA Good

1/8

11-l

ACTIVE AEROELASTIC AIRCRAFT AND ITS IMPACT ON STRUCTURE ANDFLIGHT CONTROL SYSTEMS DESIGN

Johannes Schweiger, Johann KrammerDaimlerChrysler Aerospace AG

Military AircraftP. 0. Box 801160D-81663 Muenchen

Germany

AbstractActive aeroelastic concepts have been proposed for severalyears now. Their common incentive are improvements ofaircraft performance and stability by the intentional use ofaeroelastic effects. This means that the basic flexibilitycharacteristics of a new aircraft project must be included in theearly conceptual design process, and the structural and flightcontrol system design must be coupled very closely.The knowledge about the magnitude of aeroelastic impacts onaerodynamic forces and aircraft stability is still very limitedwithin the community of people involved in aeronauticalengineering - even among the specialists in aeroelasticity. Fora successful application of active aeroelastic concepts, theirproper identification is therefore the first step. It will be shownfor some selected examples, which static aeroelastic effectsare usually very important for conventional designs, and howthey can be made even more effective in a positive sense forfuture designs.The accuracy and proper use of aeroelastic prediction methodsand analysis models is addressed briefly in the context ofinteractions with other disciplines, and ideas are developed forthe multi-disciplinary design process of active aeroelasticaircraft concepts.Whereas static aeroelastic effects usually only becomeimportant with increasing airspeed, a concept will bedemonstrated for aeroelastic improvements, which also worksat low speeds.

1. IntroductionTraditionally. aircraft structure and flight control design couldbe handled as quite independent processes. The flight controlconcept was defined as a part of the conceptual designprocess. After this, the structural design concept w as defined.taking into account that the structure had to be strong enoughto bear the loads for all desired maneuvers, including theforces from the predefined control surfaces. The detaileddimensions of the individual structural components could thenbe determined by a refined assessment and distribution ofaerodynamic and inertia loads. For each of the followingdesign loops, these loads were considered invariable fromchanges of the local mass distributions, from resulting changesof the control forces, or from aeroelastic effects on theaerodynamic loads. These changes could only be analyzedafter each major design loop, and then be used as an update forthe next loop.

This hierarchical approach allowed no feed-back from thestructural design to the flight control system design. In thepast, flexibility or structural dynamic effects could only beidentified and quantified very late in the design process, Thisresulted in additional weight. degraded performance, or costlyredesigns. On the flight control side. adjustments to optimizethe handling qualities could be made quite easily during theflight test program, as long as the flight control system wasmanually actuated. This was also possible, if servo actuatorswere used. Even an analog electrical flight control system withfeed-back loops allowed quick fixes or adjustments by trialand error methods.For a modern airplane the development of the digital flightcontrol system is a time consuming and costly process.Therefore. it is today much more important, to know theaeroelastic characteristics of the airplane as good and as earlyas possible during the design process.This fact becomes even more important, when activeaeroelastic concepts are considered for a design. They willeither have their own control system, which may create stronginteractions with the aircrafts main flight control system, orthey are directly controlled by this one. In this case, it meansdirect impacts on the flight control systems authority andstability.

2. Historic developmentsAlthough aeroelastici ty was still completely unknown to thepioneers of aviation. the success of the first powered flightmay be contributed to a great extend to a sophisticated activeaeroelastic concept for directional control. In his book Howwe invented the airplane, Orville Wright describes thissystem of cables, connecting the sliding cradle, which couldbe moved by the pilot. to the wing tips, which were twisted bythe pilot motion in opposite directions. This provided rollcontrol without ailerons.On the other hand. the Wright brothers main competitor,Samuel P. Langley, was very likely less fortunate with hisAerodrome designs because of insufficient aeroelasticstabilityAn other example demonstrates how strongly interdisciplinarythe design of airplanes already was in those early years.During the First World War. designers and governmentauthorities began to fear the loss of structural integrity frombattle damage and looked for redundancy of major load-carrying parts, In the case of the monoplane Fokker D.VIII,shown in Figure 1. a superior design with cantilevered wings.where the box structure provided excellent redundancy. therigid certification rules caused a series of fatal accidents.

Paper presented at the RTO AVT Specialists Meeting on Structural Aspects of Flexible Aircrafi Control >held in Ottawa, Canada, 18-20 October 1999, and published in RTO MP-36.


2/8

11-2

which were caused by aeroelastic divergence. As AnthonyFokker describes in his book3, sufficient strength of the designhad already been demonstrated by proof load and flight tests,when regulations called for a reinforced rear spar withproportional strength capacity to the front spar. Thisredistribution of stiffness caused torsional divergence underflight loads.

Figure 1: Fokker monoplane D-SIn the following years, designers began to fear the flexibilityof the structure, as quoted from a review paper on AeroelasticTailoring by T. A. Weisshaar4 As a result, aeroelasticityhelped the phrase stiffness penalty to enter into the designengineer S language. Aeroelasticity became, in a manner ofspeaking, a four-letter-word. . .it deserves substantial creditfor the widespread belief that the only good structure is a rigidstructure. Only recently the authors could listen to this demand again,when a colleague from flight control systems design asked tobuild future airplanes as rigid as possible.Today, the high performance of fighter aircraft, and theincreasing size of transport aircraft, together with modern lightweight structures, have enlarged the flexibility effects on theaerodynamic characteristics of airplanes. Modern airplanes arealso operating more often near and closer to the high speededge of the flight envelope. In the past, it was for examplesufficient to ensure the avoidance of aileron reversal at limitspeed, a speed where the aircraft would usually not operate.In the past, the structural dynamic characteristics of anairplane had only rarely to do with the flight control system.Flutter as the classical aeroelastic instability could be treatedindependently from the flight control system by properadjustments to the stiffness and mass distributions. Thefrequency band of the aircrafts rigid body Eigenmodes andthe dynamic characteristics of the control surface actuatorswere usually well below those of the flexible aircraft. Todaysactuators however as well as the speed of the flight controlcomputers are causing overlaps which require carefu laeroservoelastic analysis. Stability deficiencies are usuallytreated by implementing notch filters for the structuraldynamic Eigenfrequencies into the flight control laws. For afighter airplane which is usually flying in many differentconfigurations and at different flight conditions, a multitude ofthese filters may be required to provide sufficient stability.This usually results in a considerable degradation of theaircrafts agility.This implies that a different approach will be required in thefuture for the treatment of flight control system and structure

in the design process, with or without active aeroelasticconcepts.

3. Active aeroelastic concepts3.1. Definitions for active aeroelastic conceptsIn a more traditional sense, active aeroelastic concepts can bedefined as active control concepts for the cure of static ordynamic aeroelastic deficiencies with respect to stability,maneuverability, loads. or aerodynamic performance. In thiscase, the aeroelastic impacts are considered to be bad ingeneral.Examples are gust load alleviation or active flutter suppressionconcepts, where control surfaces are actively deflected tocounteract loads or create unsteady aerodynamic dampingforces. One reason why these systems did not becomecommon practice, is the insufficient static aeroelasticeffectiveness of typical control surfaces like ailerons.Since several years, the expression active aeroelastic is moreused for concepts, where aeroelastic effec ts are exploited in abeneficial way to improve aircraft performance, handling, ordirectional stability compared to a rigid aircraft. Using thisdefinition, only static aeroelastic concepts are addressed.Possible benefits from aeroelasticity were already addressed inthe sevent ies and eighties, when advanced composite materialstogether with formal structural optimization methods offeredthe possibility of Aeroelastic Tailoring an aircraft structure.One of the first demonstrations was the Active Flexible Wingwind tunnel test program6. On this model, Figure 2, twoleading and two trailing edge surfaces are adaptively deflectedat different aerodynamic conditions to achieve optimum rollcontrol power. This multiple surface control concept allows touse the trailing edge surfaces beyond their reversal speed.

To demonstrate this concept in flight, an F-18 is currentlymodified with a more flexible wing, other control surfaces,and the appropriate flight control laws .A more flexible wingmeans in this case, that an original F-18 wing torque box willbe used. This structure had to be reinforced after initial testsbecause it did not provide the desired roll control power.The expression Active Flexible Wing may have mislead tothe believe that an airframe structure must now be made asflexible as possible to improve an airplanes performance.


3/8

11-3

This is as wrong as the above quoted as rigid as possible.because excessive weight penalties would be created in bothcases. The objective must be a minimum weight design, wherethe external geometry, the arrangement and shape of controlsurfaces, are optimized together with the flight control law toachieve aeroelastic tuning or amplification of aerodynamicforces at all flight conditions.Additional interest in active aeroelastic concepts arose inrecent years with the development of active materials, wherethe stiffness can actively be adjusted. Active structuresconcepts, where the stiffness of individual components isactively modified, also belong to this category.The overview paper from McGowan et al.8 gives an excellentoverv iew on recent activities for static and dynamicaeroelastic applications.3.2. Classification of active aeroelastic concepts3.2.1. Classification by aeroelastic phenomenaIf the definition of active aeroelastic concepts is expanded toall active concepts, where structural dynamics or aeroelasticityare involved, the elastic mode control system ILAF(Identically Located Acceleration and Force) o f the XB-70must be considered as one of the first applications. A similarsystem is installed today on the B-IB to counter turbulence.In the seventies, active flutter suppression systems by meansof activated control surfaces were developed and flighttested. Besides the criticality aspect of a potential systemfailure, an other reason, why they are not yet in use, may betheir limited static aeroelastic effectiveness, as alreadymentioned above.In recent years, active concepts for the alleviation of dynamicloads from buffeting conditions of vertical tails were designedand tested in wind tunnels and on full scale ground tests withsimulated loads,.The first demonstration of active materials concepts for thereduction of dynamic loads in a flight test was a smallerstructural component, Piezo-active elements we re used toreduce the vibration loads on a skin panel of the B-IB rearfuselage section3.The group of concepts, where aeroelastic phenomena can beused in a beneficial way , can be subdivided for the followingapplications.Improvements of directional control forces by using classicalaerodynamic control surfaces as tabs to initiate the maincontrol force by an aeroelastic deformation of the fixedsurface. Theoretically, the same effec t could be achieved by anactive deformation of the fixed surface directly. Severalstudies, ~5~6~7~8 demonstrate the principle of these conceptsand explain the required design approach. Mainly roll controlis addressed by these concepts, because the outboard aileronson wings usually show the highest aeroelastic sensitivity.A second application of static aeroelastic concepts is thereduction of gust or maneuver loads. Active load alleviationconcepts in the past suffered from a lack of aeroelasticeffectiveness of the control surfaces - usually the outboardailerons.

Mainly for transport aircraft wings . active aeroelastic conceptsoffer an attractive opportunity to adjust the shape of theflexible wing for minimum drag under varying flight andinternal loading conditions.So far, mainly wings have been addressed by activeaeroelastic concepts. But horizontal and vertical tails could beat least as attractive. Whereas wings have to meet a multitudeof design objectives and requirements, which somehow limitthe design space for active aeroelastic tailoring, the onlypurpose of empannage surfaces is the provision of directionalstability and directional control about the pitch and yaw axis.This means, that they offer a larger design space. One suchconcept for a vertical tail is described below.3.2.2. Classification by active devicesThere are two major groups of active dev ices: aerodynamiccontrol surfaces and structural devices. The first one createsexternal aerodynamic forces to stimulate deformations of theflexible fixed surface, while the other one is based oninteractions between the active elements and the passivestructure by internal forces .The effectiveness of aerodynamic actuators relies upon theaerodynamic flow conditions. Their power increases linearwith the dynamic pressure at smooth fl ow conditions. Forturbulent flow, for example at high angles of attack, or forlarge deflections, they can completely lose their effectiveness.This makes them very efficient for applications at high speeds,where only small deflections are required. But this alsorequires, that their natural static aeroelastic effectiveness ishigh. Natural means that it already comes from the wing orstabilizer planform geometry and location of the controlsurfaces, with no additional investment of stiffness andweight.Aerodynamic control surfaces are limited for dynamicapplications by the frequency range of their hydraulic actuator.The active control system has to be integrated into the mainflight control system. and depending on the required controlsurface authority. they will limit the basic aircraftperformance.The effectiveness of the achievable actuation from activestructures and materials concepts is independent from theexternal flow conditions. The achievable stimulation ofaeroelastic servo-effec ts however also here depends on thebasic geometry and structural arrangement.

Their effectiveness relies upon the optimum placement withinthe passive structure to achieve the best possible deformationof the flexible structure. Active materials can be embeddedwithin the passive structure or attached to, distributed overlarger areas, or concentrated active elements are actingbetween a few selected points for high authority.It is sometimes said that these concepts could completelyreplace conventional control surfaces. But this looks veryunrealistic at the moment. The major difficulties for asuccessful application are here the limited deformationcapacity of active materials, as well as their strain allowables,which are usually below those of the passive structure.However, this can be resolved by a proper design of theinterface between passive and active structure. But theessential difficulties are the stiffness and strain limitations ofthe passive structure itself. It can not be expected that thematerial of the passive structure just needs to be replaced bymore flexible materials without an excessive weight penalty. It


4/8

11-4

is also not correct to believe that an active aeroelastic conceptwill become more effective, if the flexibility of the structure isincreased. The aeroelastic effectiveness depends on properaeroelastic design, which needs a certain rigidity of thestructure to produce the desired loads. A very flexiblestructure would also not be desirable from the standpoints ofaerodynamic shape, stability of the flight control system, andtransmission of static loads.Because large control surface deflections are required at lowspeeds, where aeroelastic effec ts on a fixed surface are small.it is more realistic to use conventional control surfaces for thispart of the flight envelope, and make use of active aeroelasticdeformations only at higher speeds. This would still saveweight on the control surfaces and their actuation system dueto the reduced loads and actuation power requirements.

4. Optimizat ion methods in aircraft designAny improvement of a technical system is often referred to asan optimization. In structural design, this expression is todaymainly used for formal analytical and numerical methods.Some years after the introduction of finite element methods(FEM) for the analysis of aircraft structures. the first attemptswere made to use these tools in an automated design process.Although the structural weight is usually used as the objectivefunction for the optimization. the major advantage of thesetools is not the weight saving. but the fulfillment of aeroelasticconstraints. Other than static strength requirements, which canbe met by af.justing the individual finite elements dimensions,the sensitivities for the elements with respect to aeroelasticconstraints can not be expressed so easily.In the world of aerodynamics. the design of the required twis tand camber distribution for a desired lift at minimum drag isalso an optimization task. Assuming that minimum drag isachieved by an elliptical lift distribution along the wing span.this task can be solved by a closed formal solution andpotential f low theory. More sophisticated numerical methodsare required for the 2D-airfoil design or for Euler and Navier-Stokes CFD methods, which are now maturing for practicaluse in aircraft design.For the conceptual aircraft design, formal optimizationmethods are used since many years. Here, quantities like directoperating costs (DOC) can be expressed by rather simpleequations, and the structural weight can be derived fromempirical data. Formal methods like optimum control theoryare also available for the design of the flight control system.

So one might think that these individual optimization taskscould easi ly be coupled for one global aircraft optimizationprocess. The reasons why this task is not so simple are thedifferent nature of the individual disciplines design variables,and their cross sensitivities with other disciplines. Theexpression Multi-Disciplinary Optimization (MD01summarizes all activities in this area. which have beenintensified in recent years. It must be admitted that today mostexisting tools and methods in this area are still singlediscipline optimization tasks with multi-disciplinaryconstraints.In order to design and analyze active aeroelastic aircraftconcepts, especially when they are based on active materialsor other active structural members, new quantities are requiredto describe their interaction with the structure, the flightcontrol system. and the resulting aeroelastic effects.

Besides formal methods, an efficient MD0 approach alsorequires experienced engineers w ith a broad knowledge in allinvolves disciplines and their interactions. It is also essential,that the proper levels of single-discipline analysis models andmethods are used for the integrated design process.5. Approach for the aeroelastic analysis of an active

structureFor the static aeroelastic analysis of the achievable rollingmoment, induced by active structural elements, theoptimization program LAGRANGE9 was modified. For thestatic aeroelastic analysis of a conventional passive structure,the analysis process is initialized by defining the aerodynamicdeflections of control surfaces. The resulting rigidaerodynamic load is then used as a starting point to obtain theaeroelast ically balanced equilibrium condition, as depicted inFigure 3.

Aerodynamic fcfi

modfied angles d attackmadfied loads

Figure 3: Process for static the aeroelastic analysis withcontrol surfacesFor an active structure, the chosen new approach firstsimulates the deformation of the structure under the loads ofthe activated elements. This static solution delivers the initialangle-of-attack distribution fatshown in Figure 4.

Apply internal forces to generate Calculate aerodynamic angle of attackdeformations in struch~ral model dishibutmn from inital deformations m

Calculate ae*oelastc equddmum frominital deformatmns

the aeroelastic analysis, as

*-Ophmlz e locahon and duxhon ofachve components

Figure 4: Static aeroelastic analysis steps for an activestructureTo verify the approach, three different cases were analyzed,using the wing model from the example below: one for aconventional control surface deflection by specifying its initialaerodynamic deflection. one with the equivalent deflectionfrom static loads applied to the element that represents theactuator, and one. where in-plane loads are applied to the skinsof the control surface. The results are shown in Figure 5.


5/8

,I, I 1Jili ITEfiB-active actuator

~TE/lB~n ctive~strrtrh III skins

Figure 5: Results for test cases6. Examples for active aeroelastic concepts

6.1. Active aeroelastic wingTo demonstrate the principle of active aeroelastic concepts forimproved roll performance, a low aspect ratio tighter aircraftwing was chosen., because here one would not expectconsiderable aeroelastic improvements, The Finite Elementmodel for this generic wing is shown in Figure 6.

3lements 1329$odes 513;~zine Design Varables:onPtramtekgrees of Freedom

Figure 6: Finite Element model for low aspect ratio fighterwingThe original model, which had only two trailing edgeflaperons for roll control, was modified by two additionleading edge surfaces for roll control. The had to be rathersmall to avoid modifications of the torque box structure.Figure 7 shows their representation in the aerodynamicanalysis model. The basic design conditions for the

ltimization are summarized in table 1.

T/O

.. T/l

LFigure 7: Aerodynamic model of the wing

11-5

Static load casesRoll rate at Ma 1.2, S/LMax. hinge moment per surfaceMax. control surface deflection

+ 9g 3 -4g120 O/s15 kNm15O

Table 1: Basic wing design conditionsFigure 8 depicts the achievable rolling moment withincreasing airspeed for rigid conditions. These two graphsclarify, why usually only trailing edge surfaces are consideredfor roll control. Besides their different size, main reasons forbetter aerodynamic performance are the different sweepangles. and mainly the camber effect, which supports thetrailing edge surfaces and counteracts at trailing edgedeflections

igure 8: Achievable rolling moment for a lo deflection ofleading and trailing edge surfaces without structuralflexibilityBut things look quite different, as soon as the aeroelasticeffects on the flexible structure are taken into account. For theinitial structure, which had already been optimized forincreased rolling moment effectiveness of the trailing edgesurfaces, the achiev/able rolling moments are compared inFigure 9.

Fi gure 9: Rolling moments for leading and trailing edgesurfaces with flexibilities of the initial design

Dynamic Pressure (Wa)

Traditionally, if only the trailing edge surfaces are used forroll control, the achievable roll rates for a design with fixedplanform and control surface geometry can only be improvedby additional structural weight. For this example, thestructural optimization of the wing box skins with differentconstraints for the rolling moment effectiveness results in thegraph of Figure IO for skin weight and effect iveness. If theleading edge surfaces are used in addition, the results in table2 can be achieved for the basic static design. If buckling


6/8

11-6stability is not considered for the wing skin design, whichcould be possible by additional spars, the rolling moments vs.dynamic pressure for the individual control surfaces in Figure11 are obtained. In this case, the leading edge surfaces getmore effective, and the trailing edge surface could be usedbeyond the reversal speed. More details about this study arereported in refs.6.7.

0 50 100 150 200 250 300 350Weight (kg)

gure 10: Optimization results with different rollingmoment effectiveness constraints

Table 2: Optimization results for the wing

Figure 11: Rolling moments for the static design withoutbuckling stability6.2. Active aeroelastic vertical tailThe structural design of the vertical tail for a tighter aircraftusually requires additional stiffness fo r the static aeroelasticeffectiveness of the lateral stability and for the rudder yawingmoment. A typical effectiveness-vs.-weight trend is depictedin Figure 12. To improve this situation, the concept of aDiverging Tail was developed by Sensburg et al.20.1n a firststep, the effectiveness is increased by means of aeroelasticallytailored skins.

I I I200 30.0 40.0 500

Skin WeIghi [kg]Figure 12: Static aeroelastic effectiveness vs. structuralweight for a typical vertical tailAlthough this means a weight increase for the cover skins, thetotal structural weight can be decreased by reducing the size ofthe tail proportional to the increase in effectiveness,Additional increases are then obtained by relaxing the stiffnessof the forw ard root attachment and modifying the rearattachment in such a way that the aerodynamic surface can bedeformed in a more favorable way .For further improvements, the concept of an Active VerticalTail was developed. The principle is depicted in Figure 13. Itis an all-movable tail, where the attachment is positioned insuch a way , that the aeroelastic effectiveness is above 1.0 forall aerodynamic flow conditions. The amount o f effectivenesscan be adjusted by a variable torsional stiffness element. Thiscan for example be achieved by a mechanical, hydraulic,electric, or active materials system. The actuation of the tailfor yaw control can be integrated into the same system, or itcan be designed as a separate system. A separate actuationsystem by a conventional actuator wi th constant stiffnesswould be less complex for the flight control system.

All movable vertical tailwith active attachment

onventlonal hydraulic

Figure 13: Concept of an all-movable Active Vertical TailThis vertical tail needs no additional weight for aeroelasticeffectiveness. Its size can be reduced to the value, wheresufficient directional stability and yaw control are provided bythe proper amount of effectiveness from the variableattachment stiffness. The lower boundaries for the stiffness aredefined by sufficient flutter stability. This means, that flutterstability and aeroelastic effectiveness have the same stiffnessdemands: low at low speeds, and high at high speeds.This concept reverses the traditional design approach fo rimproved aeroelastic e ffectiveness, where an increase is


7/8

11-7achieved by additional stif fness. Whereas the minimum size 01the passive design for the diverging tail is limited by stabilityand control requirements at low speeds, where no aeroelasticeffectiveness improvements are possible. the Active VerticalTail also provides increased effectiveness at low speeds.A different concept fo r a variable stiffness actuation system isalready used for the new F/A-ISEIF. Here the hydraulicpressure switches from 207 bar to 345 bar (3000 to 5000 PSI)at high dynamic pressures to compensate aeroelastic losses

7. Impacts from active aeroelastic concepts on theFCS designThe design of the flight control system should not becomemore complex because of an active aeroelastic concept. Butthe impacts from this active concept on the aircraftsparameters, which are implemented in the flight control laws.must be known and respected.The performance of the flight control system should not bedegraded in the presence of an active aeroelastic system. Ifdesigned properly, there should even be improvements, likereduced power and stiffness demands for the flight controlactuation system.As an example, the simplified schedule in the flight controllaws for the leading and trailing edge surfaces for roll controlof the wing above could look like in Figure 14.

GAIN o

IDynamic Pressure AxFigure 14: Leading anh trailing edge control surfaceauthority for roll control

The variable sti ffness for the attachment of the vertical tailmight look like in Figure 15. Here, the optimum stif fness foraeroelastic effectiveness must be tuned together with the flightcontrol laws for handling and stability requirements.

8. Impacts from active aeroelastic concepts on thestructural designIn order to incorporate active aeroelastic concepts into thestructural design, it is no longer suff icient to speci fyaeroelastic constraints like for flutter or control surfaceeffect iveness, and apply it to the structural optimizationprocess for a predefined structural concept.The design space for aeroelastic effects must be as wide aspossible in the beginning. That means. the sensitivities ofbasic geometry parameters for wings and control surfaces, thepositions of control surfaces, and their functions must beconsidered as design variables in the beginning.

Attachmentstiffness

Required for optimal effectiveness

I Dynamic pressureFigure 15: Attachment stiffness requirements for ActiveVertical TailThe design space for aeroelastic effects must be as wide aspossible in the beginning. That means, the sensitivities ofbasic geometry parameters for wings and control surfaces. thepositions of control surfaces, and their functions must beconsidered as design variables in the beginning.The analytical description of active aeroelastic concepts mustdirectly be included in the structural analysis model because ofthe impacts from the passive structures design constraints onthe effectiveness of active aeroelastic systems. In order tomake them efficient. it is required to understand, design andsimulate the interfaces between components and the passive,load-carrying structure.

9. Needs for the integrated design of airplanes withactive aeroelastic conceptsIt is obvious that integrated design and multi-disciplinaryoptimization processes are an absolute must for activeaeroelastic concepts.MD0 does not mean to combine single discipline analysistools by formal computing processes. It means first a goodunderstanding of what is going on. This is already essential fora conventional design. Only after this understanding thecreative design of an active concept can start.It is then very important to choose the proper analysis methodsfor the individual disciplines. Usually, not the highest level ofaccuracy is suitable for the simulation of important effects forother disciplines. This also refe rs to refinement of the analysismodels, where local details are not interesting for interactions.It is more important to keep the models a versatile as possiblefor changes of the design concepts to allow the simulation asmany variants as possible. This also means an effic ientprocess for the generation of models, including the knowledgeof the user for this process. Fully automated model generatorscan create terrible results, if the user can not interpret orunderstand the modeling process.Also the quality and completeness of analysis models isessential, as far as impacts on neighbor disciplines areconcerned. Especially for formal optimization processes,model errors will create foolish results. To achieve goodresults, a careful selection and combination of the designvariables and the completeness of the design requirements areimportant.

10. ConclusionsThe qualities and quantities of impacts from aeroelasticity onstructural loads, aerodynamic performance, maneuverability.


8/8

11-8stability and agility of the flight control system of an airplanebecame more and more important in recent years. This fact isnow more and more often also recognized outside theaeroelastic community.Especially the complexity of a modern digital flight controlsystem requires a careful identification of aeroelastic impactsto avoid degradations or costly redesigns. If an efficient MD0process can be set up early enough for a new design.aeroelastic impacts can be minimized or can even be used in apositive sense.While this is slow ly being accepted for a conventional design.concerns arc already expressed, that active aeroelasticconcepts may not be desirable because of possible negativeinterferences with the flight control system. \vhich is alread!,complex enough.The development of active aeroelastic concepts shouldthcrcfore not merely be seen as a task in aeroelasticity. It mustbe a creative part of the overall flight control system design.together w ith the aerodynamic and structural design. Thisprocess must include experts from all involved disciplines(flight control laws, actuation systems. including those foractive structures, aerodynamics, structure. and aeroelasticity)with a good understanding of the other disciplines.If this is possible. great achievements from active aeroelasticconcepts can be expected for future designs o f airplanes andall kinds of flying vehicles.

[]IPI

131

141

t51

161

References

Wright, 0.: How we Invented the Airplane. Dove1Publications, Mineola. NY. 1988Flomenhoft. H. I.: The Revolution in StructuralDynamics. Dynaflow Press, Palm Beach Gardens, FL.U.S.A.. 1997Fokker, A. H. G.; Gould. B.: Flying Dutchman. The Lifeof Anthony Fokker. Henry Holt and Company, NewYork. U.S:A., year unknownWeisshaar. T. A.: Aeroelastic Tailoring ~ Creative Useof Unusual Materials. AIAA-87-0976-CPShirk. M. H.; Hertz, T. J.; Weisshaar, T. A.: A Survey ofAeroelastic Tailoring - Theory. Practice, Promise.AIAA-84-0982-CPJournal o f Aircra ft, Vol. 32. No. I. Special Section:Active Flexible Wing. 1995Pendlcton. E.; Griffin, K.E.: Kehoe. M.W.; Perry. B.: AFlight Research Program for Active AeroelasticWing Technology. Conference Proceedings AIAA-96-I574-CPMcGowan. A.-M. et al.: Aeroservoclastic and structuraldynamic research on smart structures conducted atNASA Langley Research Center. SPIE Vol. 3326. 1998.Pace. S.: Valkyrie- North American XB-70.4. AeroSeries Vol. 30; Tab Books, Blue Ridge Summit, PA.U.S.A.. 1984Nell. T.; Hfinlinger, Il.: Sensburg. 0.: Schmidt. K.:Active Flutter Suppression Design and Test. a Joint

[111

1121

[I31

[I41

u51

[I61

[I71

PSI

[I91

PO1

1211

U:S.-F.R.G. Program. ICAS- CP 80-5.5. Munich,Germany, 1980.Hopkins, M.A.: Henderson. D.A.: Moses. R.W.: ActiveVibration Suppression Systems Applied to Twin TailBuffeting. SPIEs 5th Annual International Symposium,,Smart Structures and Materials San Diego, CA.1998.Simpson. J.; Schweiger. J.: Industrial Approach toPiezo-electric Damping of Large Fighter Aircraf tComponents. SPIES 5th Annual InternationalSymposium Smart Structures and Materials. SanDiego. CA, 1998.Larson. C.R.: Falanges. I.: l)obbs. S.K.: PiezoceramicActive Vibration Suppression Control SystemDevelopment for the B-l B Airc raft . SPIEs 5th AnnualInterntional Symposium Smart Structures andMaterials. San Diego. CA. 1998.Crowe, C.R.; Sater, J.M.: Smart Aircraft Structures.AGARD Symposium on Future Aerospace Technologyin the Services of the Alliance. CP-600, Vol. I. Paris.1997Andersen. G.; Forster, E.: Kolonay, R.; Eastep. F.: AStudy of Control Surface Blending for ActiveAeroelastic Wing Technolog). 37th AIAA-SDMConference CP-96- I443Khot. N.: Eastep. F.: Kolonay. R.: A Method forEnhancement of the Rolling Maneuver of a FlexibleWing. AIAA-CP-96-1361.Schweigcr, J.: Simpson. J.: Weiss. F.; Coetzce. E.;Boiler. Ch.: Needs for the analysis and intedrated designoptimization of active and passive structure for activeaeroelastic wings. SPlEs 6th Annual InternationalSymposium on Smart Structures and Materials.Newport Beach. CA, 1999.Schweiger. J.: Krammcr. J.; Coetzee, E.: MD0Application for Active Flexible Aircraft Design. 7thAIAA/NASA/ISSMO S) mposium on Multidisciplinar>Analysis and Optimization. CP-4169. St. Louis. MO.1998.Schweiger. J.: Krammer. J.; I liirnlein. H.: Developmentand Application of the Integrated Structural Design ToolLAGRANGE. 6th AIAA/NASA/ISSMO Symposiumon Multidisciplinary Analysis and Optimization. CP-4169, Seattle, WA, 1996Sensburg, 0.: Schneider. G.: Tischler. V.; Venkayya, V.:A Unique Design for a Diverging Flexible Vertica l Tail.Specialists Meeting on Structural Aspects of FlexibleAircraft Control. RTA Meeting on Design Issues.Ottawa. Canada: October 1999.Balanced lipgrade. Flight lntcrnational, January 20-261999.

Aeroelasticity - NASA Good

Documents

Transcript of Aeroelasticity - NASA Good