Ight Vehicle Research Articles

Personal Perspective of Aeroelasticity During the Years 1953-1993

A brief review is given of the author’ s personal impressions of the development of aeroelasticity during the sixth to ninth decades of e ight. It is not intended to be all-inclusive, but rather a brief look at the e eld as the author saw it, from his own personal viewpoint as a worker in this area. It is limited to aircraft problems and does not embrace the parallel and large e eld of aeroelastic problems in helicopters. The e eld of aeroelasticity encompasses the interaction of aerodynamic, elastic, and inertia forces acting on a e ight vehicle. This was evident in the early days of e ight, even at low speeds, because aircraft structures were relatively e exible. As aircraft speeds increased and knowledge of aeroelasticity became more widespread, aircraft structures were made stiffer and were better designed to avoid these aeroelastic interaction problems. During World War II, with the advent of high speeds and slender-wing aircraft, a number of signie cant aeroelastic problems began to appear, ranging from loss of aileron power and control reversal, to wing and control surface e utter. The postwar period, with the emergence of highspeed jet transports with swept wings and engine nacelles, led to careful studies of the effects of aeroelasticity in the initial design of these aircraft. The basic approach to aeroelastic effects became centered around four general areas: 1) the development of effective analytic and numerical studies for e utter and aeroelasticity, 2) wind-tunnel testing of dynamically scaled wind tunnel models, 3) ground vibration tests to check the natural frequencies and stiffness properties of the actual aircraft, and 4) careful techniques for e ight testing the actual aircraft. By the mid-1950s, aeroelasticity had become recognized as an important part of the aircraft design process. An excellent historical review of developments in aeroelasticity up to this period, may be found by Garrick and Reed. 1 A seminal paper that marked the emergence of the e eld was given by Collar 2 in 1946. Another notable review emphasizing the British efforts in aeroelasticity up to this period was also given by Collar. 3

Development of Direct-Measuring Skin-Friction Gauges for Hypersonic Flight Tests

Anewtypeofdirect-measuring skin-frictiongaugewasdevelopedforthehigh-speed,high-temperatureenvironmentofturbulentboundary-layere owsinsupersoniccombustionramjet (scramjet)engines, with tests progressing from ground tests to an actual hypersonic scramjet-integrated e ight vehicle. The skin friction gauge was specie cally developed for installation on the Hyper-X e ight vehicle (X-43A). The design was nonnulling, with a sensing head that was e ush with the model wall and surrounded by a small gap. Unlike previous skin-friction sensor designs, this gauge eliminated water cooling and gap-e lling oil, in addition to a radically different e exure design. The design was verie ed by repeatable tests in a well-documented Mach 2.4 cold e ow, with results within 10 ‐15% of Cf estimates from simple theory. The sensor was qualie ed for e ight through a rigorous series of environmental tests, including pressure, temperature, vibration, shock, acceleration, and heat e ux tests. Finally, the skin-friction gauge was tested in the Hyper-X engine model (HXEM), a full-scale ‐partial-width wind-tunnel model of the e ight vehicle engine, at Mach 6.5 enthalpy in a freejet facility at NASA Langley Research Center. Successful testing in the HXEM provided the e nal verie cation of the gauge before installation in the X-43A e ight vehicle engine.

Stabilization of Optimum Trajectory Costate Differential Equations

Introduction I N trajectory optimizationand synthesis of initial-valueand twopoint boundary-valueguidance and control laws, most applications of the Euler–Lagrange Ž rst-variation necessary conditions of the calculusof variationshave led to asymptoticallyunstablecostate differentialequationsand, hence,unstablebehaviorof systemphysical states.This instabilityhas beenmost troublesomewhen solving for long-durationextremal arcs, particularlywhen signiŽ cant statedependent nonlinear forces act on the  ight vehicle. For example, the purposemight be to designminimum fuel consumptiontrajectories for cruising  ight of a high L=D aircraft; the optimumsolutions may involve sustainedphugoidal oscillations, but that behavior can be obscured by rapid costate divergence. ThisEngineeringNote introducesamethod for stabilizingcostate differential equations via analytical phase adjustment of the timevarying costates. The method employs a time-varying phaseadjustment parameter. For a given model of the vehicle dynamics, a necessary lower bound on the phase-adjustment parameter is derived, and a stronger sufŽ cient condition is suggested. For simplicity, the subject is here presented in the context of a Ž rst-order one-dimensionalsystem,but extensiontomultidimensionalsystems is straightforward. Consider Ž rst a one-dimensionalguided object (such as a sled or railcar) that moves horizontallyon a straight path and has the linear, time-invariant state equations:

Computer-Aided Assessment of Flight Simulator Fidelity

A technique for computer-aided assessment of e ight simulator e delity is presented. The assessment procedure utilizes a mathematical model of the human pilot that includes proprioceptive, visual, and vestibular cues. The quality ofthelattertwo cuescanbevariedtocapturetheeffectsofe ightsimulatorlimitationsassociatedwithvisual and motion cueing. A MATLAB ® -based tool that automatesthe selection of the majority of parameters in the pilot model and the generation of a e delity metric is described. In addition, a prediction of the task-dependent handling qualities of the nominal e ight vehicle can be obtained. The assessment procedure is exercised in a hypothetical example involving the six-degree-of-freedom control of a rotorcraft completing a vertical remask maneuver. It is demonstrated that by varying the quality of the visual and vestibular cues in the pilot/vehicle model the effects of simulator limitations upon e delity can be systematically addressed.

Proposed System-Level Multidisciplinary Analysis Technique Based on Exergy Methods

It is suggested that it may be time to consider whether we have reached a plateau in terms of the evolutionary nature of e ight vehicle design and optimization. For a time, progress was measured in terms of maximum speed, which is a straightforward metric when the next design is evolved from the preceding model. There are times, however, when we need to depart from the evolutionary process and create a breakthrough design. The question to be asked is whether there is any way to dee ne system-level analysis and optimization techniques to facilitate the vehicle design process with a more global measure of effectiveness. This paper presents such a methodology for the design of the complete integrated system of systems. Work that has been done in energy-based methods is briee y reviewed, sinceenergy is an implicit consideration in many aerospace disciplines. In addition, methods such as exergy and thermoeconomics have been applied in the design of ground power stations and they are currently being studied for application to aircraft subsystems. The objective of this paperis to expand exergy methods to the design of a complete e ight vehicle by dee ning mission requirements as an exergy/work problem cascading down to each component in the same framework. This paper also serves to introduce a special section of this journal devoted to the application of exergy methods to all levels of e ight vehicle design. Overall, the proposed technique provides a method to facilitate system-level optimization at all levels of the design process.

Angle-of-Attack Effect on Transonic/Supersonic Aeroelasticity of Wing-Box Model

N understanding of the aeroelastic behavior of e ight vehicles in the transonic and low-supersonic regimes is of great importance for e ight safety. The e utter boundary in this regime varies with changes in the initial angle of attack. Aeroelastic analyses and experiments on the effect of initial angle of attack have been performed previously. Early studies of the two-degree-of freedom airfoil system were performed using the HYTRAN2 (Ref. 1) and an Euler code. 2 For a three-dimensional wing at high angles of attack in incompressible e ow, there is the work by Strganac and Mook. 3 In their paper, using the unsteady vortex-lattice method, the equations of motion were integrated, considering the nonlinear effects of the separated vortex. Yates et al. 4 analyzed the effect of angle of attack on a large aspect ratio transport-type wing with a supercritical airfoil, using a modie ed strip analysis employing wind-tunnel steady aerodynamic data. Also, the CAP-TSD code 5 has been applied to the active e exible wing wind-tunnel model to investigate static and dynamic aeroelastic behaviors below Mach 0.95. These studiesprovideagoodfoundationforunderstandinginitialangle-ofattack effects, both theoretically and practically, and motivated by this,wewillexamineindetailtheeffectofbothpositiveandnegative angles of attack on a typical e ghter wing-box model with an asymmetric airfoil in the transonic and low-supersonic e ow regions. The critical effect of a negative angle of attack and unusual frequency changes due to the effect of normal shocks are presented. The computed steady aerodynamic results for rigid and deformed shapes of the model are presented and compared. Also, detailed dynamic aeroelastic responses are computed using a coupled time-marching method based on the effective computational structural dynamic (CSD) and computational e uid dynamic (CFD) techniques, 6 which are similar to those used in Ref. 5. CSD analyses for the wing-box model have beenperformed using MSC/NASTRAN.Thevariations

Unified Hypersonic/Supersonic Panel Method for Aeroelastic Applications to Arbitrary Bodies

A unie ed unsteady hypersonic/supersonic panel method has been developed and incorporated into ZAERO (a unie ed unsteady/steady aerodynamics/aeroelasticity methodology/software for aeroelastic applications ). The present unie ed panel method is a viable design/analysis tool forfuturehypersonic e ight vehicles and hypervelocity missiles. The method is established with a rotationality correction function to the general three-dimensional linear potential solution. An equivalent Mach number transformation is introduced to circumvent the superinclined panel limitation. Next, a local pulsating cone analogy is applied to each panel considered, thus yielding the inphase and out-of-phase forces and moments for an arbitrary body with proper rotationality correction based on steady/unsteady Euler solutions. The obtained results were verie ed and validated with CFL3D solutions, AP98 results, and measured data. For all cases studied, the unie ed panel method is found to be computationally efe cient (2‐3 min per case ), acceptably accurate, and very versatile in its applicability, including aerodynamic/stability derivative/aeroelastic/aeroservoelastic applications for complex wing ‐body or missile-e n cone gurations in the unie ed hypersonic/supersonic e ight regime.

Tail Buffet Alleviation Through the Use of Wing-Strake Fillet Shapes

Various e llet shapes at the strake ‐wing juncture of a 76/40 deg double-delta wing model were experimentally investigated to gauge the effect of the e llet in alleviating adverse vortex/tail interaction. The vertical tails were instrumented with 28 fast-response pressure transducers. Pressure time histories and frequency power spectral densities were analyzed for four wing ‐strake e llet cone gurations and at three spanwise tail locations. Angle of attack was varied from i 2 to 40 deg, tunnel dynamic pressurewas 26.74 psf (1280 pascals ), and Reynolds number was 1.3 ££ 10 6 . The results show that the e llet geometry can increase or decrease tail buffet in comparison to the baseline, no e llet case. Linear and parabolic e llet shapes were found to increase tail buffet above the baseline case. A diamond e llet shape was found to alleviate tail buffet and shift the tail buffet to a higher frequency. The use of such a shape on e ight vehicles may result in a longer fatigue life of the vertical tails. Nomenclature A = constant,b=V1, 0.0091/s AR = aspect ratio, 2.41 a1 = speed of sound, 1155 ft/s, 352 m/s b = span, 1.36 ft, 0.414 m C = coherence CP = pressure coefe cient, p i p1=q1 c = chord, ft, m cR = root chord, 1.33 ft, 0.405 m cRtail = tail root chord, 0.38 ft, 0.116 m ct = tip chord, 0.21 ft, 0.064 m cttail = tail tip chord, 0.13 ft, 0.040 m f = frequency, Hz G = amplie er gain Iraw = raw integer value (digital) M1 = freestream Mach number, V1=a1, 0.13 P = pressure, psf Pref = reference pressure, psf q1 = tunnel dynamic pressure, 26.74 psf, 1280 pascals Re = Reynolds number, (Ω1V1CR/=π1, 1.3E6 S = wing area, 0.768 ft 2 , 0.071 m 2 Se = transducer sensitivity, V/psf Sr = Strouhal number, (bf /=V1

Flutter Prediction Methods for Aeroelastic Design Optimization

The objective of the paper is to demonstrate the validity of e utter predictions obtained using eigenvector orientations (EVO) for realistic wings used in aircraft preliminary design. Comparisons of e utter predictions between the V-g method and the EVO method are presented for an intermediate complexity wing used in optimization studies. The results presented show that the EVO method can predict the onset of e utter for aircraft wings with reasonable accuracy and can be used in design optimization studies of aerospace vehicles/components. I. Introduction T HEdesign of modern e ight vehicles requires multidisciplinary designoptimization (MDO),theintegrationofstructures,aerodynamics,controlandpropulsiondisciplines,toachieveanoptimum design based on certain performance criteria and constraints. In the design of e ight vehicles, e utter is a critical parameter that must be considered in the early stage of the design cycle. Also, with aging aircraft,certainrestrictionsneedtobeimposedinordertoavoide utterinstabilityduring thee ightregimeoftheaircraft.Alargenumber of parameter changes must be investigated in the e utter design of military as well as commercial aircraft as a result of the complexity of modern composite/metallic structures. Accurate and efe cient methods of e utter prediction for aerospace vehicle components are needed for use in aeroelastic design optimization studies. There are several aeroelastic design optimization codes being used for aircraft and spacecraft structures, such as the Automated Structural Optimization System (ASTROS) 1 and MSCNASTRAN. 2 In aeroelastic design optimization the predominant dynamic e utter instability needs to be investigated for most e ight vehicles. Several methods are being used to predict the onset of e utter instability in aeroelastic design optimization. Most of these methods arebasedoncomplexeigenvalues (p-kmethod, K-method, orKE method). 2 Recently,Afolabietal. 3 developed amethodbased on eigenvector orientations (EVO) to predict the onset of e utter instability. In their work the method of EVO was used for limited examples of panel e utter. To gain a better understanding of the EVO method of e utter prediction, further investigations need to be carried out on realistic aircraft structures. This understanding will help to devise a prediction methodology that may be implemented in MDO codes for the preliminary design of e ight vehicles. The objectiveofthispaperisto further validate the EVOmethod 3 of e utter prediction for realistic examples used in aircraft preliminary design. Abriefreviewofthe EVOmethod for e utterprediction isdescribed inthe nextsection.Anexampleofanintermediatecomplexity wing used in design optimization studies is considered, and e utter prediction results are compared to the commonly used V-g method.

Scaling Nonequilibrium-Reacting Flows: The Legacy of Gerhard Damko-yen;hler

Damks personal background is e rst outlined, followed by an examination of his early pioneering work in the 1930s on reacting e ows, which led to the basic nondimensional similitude parameters that now bear his name. We then survey the role that these parameters have played in dealing with e nite reaction-rate effects in acoustics, combustion, and the aerothermodynamics of hypervelocity e ight vehicles and their propulsion systems. Emphasis is placed on the aerospace sciences and the value of the Damk ¨ ohler number not only in understanding and correlating experimental data, but also in conducting efe cient numerical studies and analyses of reacting-gas e owe elds.

Helical-Groove and Circular-Trip Effects on Side Force

When a slender body, such as a missile, is pitched at high angle of attack to an oncoming e ow, it may experience a large side force due to the asymmetric shedding of the tip vortices. The side force is well known to be highly detrimental to the performance of the e ight vehicle. We assess the effectiveness of two control devices, namely, the circular trips and the helical grooves, in alleviating the side force on a tangent ogive nose cylinder. Simultaneous side force and pressure measurements taken in a wind tunnel show that the circular trip is generally more effective in reducing the side force than the helical grooves over a wide range of angle of attack. Detailed e ndings of their performances are reported.

Optimal Vibration Control of Thin-Walled Anisotropic Cantilevers Exposed to Blast Loadings

Astudyofoptimalfeedbackcontrolofthedynamicresponseofcantileversexposed to blastloadingsispresented. The structure to be controlled consists of a thin-walled beam of closed cross section contour that encompasses a number of nonclassical features such as transverse shear, primary and secondary warping, and anisotropy of the constitutent material. The control is achieved via the use of actuating and sensing capabilities provided by piezoelectric devices that are bonded or embedded into the host structure. In addition, the directionality property of advanced e ber-reinforced composite materials is also used to tailor the host structure to obtain an enhanced dynamic response. The cases of piezoactuators spread over the entire span of the structure or in the form of a patch are considered, and issues related with the ine uence of patch location and size on the control efe ciency are discussed. Other issues related to the minimization of the required input power and implications of the limitation of control input voltage and of those generated by the inclusion/discard in the quadratic performance index of time-dependent external excitations are also addressed. I. Introduction T HE increasing need for high-speed, highly e exible, and lightweight structural e ight vehicles, capable of operating in severe environmental conditions, has prompted the development of a new concept in the modeling, analysis, and control of structures. This concept is based on the extensive use of advanced composite material structures 1 and integration in their construction of smart materials technology. 2 Such structures are often called intelligent/smart composite material structures.

Aeroservoelastic Interaction Between Aircraft Structural and Control Design Schemes

The behavior of the structural and control systems of  ight vehicles are highly coupled through aeroelastic effects. An interactionmodulewas formulated and software was developed to facilitate efŽ cient transfer of models, data and design requirements between the structural and control design schemes. The aeroelastic plant statespace equations are based on a minimum-state rational function approximation of the unsteady force coefŽ cient matrices. The control system is deŽ ned in a way that allows incorporationof most general linear control laws in the aeroservoelastic loop, and yet allows efŽ cient control margin and sensitivity computations by separating between changeablegainsto otherŽ xedcontrol elements andparameters.Analyticalexpressionsfor systemstabilitymargins and their sensitivities with respect to structural and control designparameters, which take advantageof the special controller representation form for incorporation in a multidisciplinarydesign scheme, are presented. A numerical example demonstrates a design task that involves both control synthesis and structural optimization.

Dynamic Response of Pressure Sensing Systems in Slip-Flow with Temperature Gradients

Introduction T HE accurate sensing of surface pressures on hypersonic  ight vehicles presents formidable measurement challenges. The hostility of the sensing environment precludes intrusion into the  ow, and measurements must be obtained via sizable lengths of small-diameter pneumatic tubing that connect the surface ports to the remotely located pressure transducers.An idealized pneumatic layout is depicted in Fig. 1. Further complicating the sensing problem is surface boundary layer heating,which causes sizable temperature gradientswithin the pneumatic tubing.Away from the stagnation region, local surface pressure levels,which must be sensed, are extremely low, often less than 0.002 atmospheres (0.2 KPa). The combination of very low pressure levels, small diameter pneumatic lines, and large temperature gradients makes molecular effects no longer negligible. The problem of predicting tube  ow dynamics has been studied extensively. For nonrareŽ ed conditions, Lamb,1 Iberall,2 Schuder and Binder, and Hougen et al., have developed closed-form frequencydomain solutions for simple tubing geometries and constant wall temperatures.Berg and Tijdeman and Tijdeman extended the analyses of Refs. 3 and 4 to develop a recursion formula for complexgeometriesthat consistof cascadesof tubesandvolumes.Parrot and Zorumski7 investigated the dynamic transmission of sound in a simple geometry tube subjected to very large temperature gradients. References 1–7 considered only continuum  ow conditions; rareŽ ed  ow effects were not investigated. Tube  ow for rareŽ ed conditionsat high temperatureswith large temperaturegradientshas been investigated by Maxwell,8 von Knudsen,9 and Tompkins and Wheeler. These rareŽ ed  ow investigations,however, considered only steady- ow conditions. The dynamic in uence of rareŽ ed  ow phenomenaon pneumatic pressure sensing systems has not been generally well understood; consequently, research was initiated at the NASA Dryden Flight Research Center with a primary objective to develop a dynamic response model for pneumatic pressure sensing systems that is applicable to both continuumand rareŽ ed  ows.Whitmore et al.11 presented a detailedanalyticaldevelopmentand empiricalvalidationof one such model. This Note summarizes the information presented in that text.

Uncertainty Modeling for Multivariable-Control Robustness Analysis of Elastic High-Speed Vehicles

High-speed (supersonic or hypersonic ) atmospheric e ight vehicles are typically characterized by a signie cant degree of interaction between the highly elastic airframe and the propulsion system. To achieve adequate stability and performance requirements, robust, integrated multivariable control laws will be required. But to apply robust-control analysis or synthesis techniques such as structured-singular-value techniques (π) or quantitative feedbacktheory,theuncertaintyintheplantdynamicsmustbecharacterized inspecialways.Furthermore,certain assumptions regarding the uncertainties present are frequently made in the application of these techniques. The focus of this research is the development of uncertainty models for this class of e ight vehicle that are derived from the physics of the system, yet are compatible with the cited control synthesis techniques. The potential sources of uncertainty for this class of vehicle are discussed, and three forms of uncertainty models are developed: real parameter,unstructured,and structured. Weareespecially interestedinhowtheusualsourcesofuncertaintymanifest themselves in this context. It will be shown that for this class of vehicle care is required in making the usual assumptions regarding the uncertainty. It is also shown that the e exible degrees of freedom must be considered in the e ight-control synthesis for this class of vehicle.

Autonomous Maneuver Tracking for Self-Piloted Vehicles

A fully autonomous  ight vehicle must be capable of accurately tracking a desired trajectory for the safe and effective completion of a given mission. An approach to maneuver autopilot design is presented based on a six-degree-of-freedom nonlinear trimming algorithm coupled with a robust tracking controller. Acceleration commandsrequired for closed-loop tracking of a desired trajectory in an Earth-Ž xed reference frame are provided bya simpleguidancestrategy.Thetrimmingalgorithmgenerates optimalstate commandsandfeedforward controls to accomplish themaneuver,while assuringthat these commandscan be tracked by an inner-loopcontrollerwithout violatingvehicle performanceoractuatorconstraints.Thisapproachis demonstratedin simulated ightof anactual Ž xed-wing uninhabitedaerial vehicle, namedAriel. Excellent trackingperformance is achieved for a representative three-dimensional trajectory using a high-Ž delity aircraft model with nonlinear dynamics and realistic wind gust disturbances. In addition, an analysis of system robustness to uncertainty in the vehicle model is performed.

Estimation of Wind from Airplane States in Coordinated Flight

A e lter is dee ned that extracts the wind aloft from the Earth velocity and the orientation of a e ight vehicle in coordinated e ight (data that are available in the context of distributed interactive simulation ). A convergence theorem is proved stating that the estimated wind monotonically approaches the true wind when the latter is horizontal and constant. The e lter builds the transverse component of the wind and converges to the full wind as the e ight vehicle changes heading. The fastest convergence occurs when the e lter is applied at heading intervals equal to a critical interval that depends on the bank.

Simulations of Wind Along Flight Trajectories in the Ocean Surface Layer

A simulation model for numerical realizations of turbulence along e ight trajectories in the atmospheric ocean surface layer is developed and tested. The model is based on a Monte Carlo simulation technique in which a digitally generated random signal is e ltered to represent a random realization of the target process. In addition to purely atmospheric/shear-driven turbulence, the model includes a surface perturbation effect on the wind e eld caused by the underlying wave surface. It is shown that the wind e uctuations may take varying forms, depending on factors such as e ight direction, e ight height, and wave state. The differences found in frequency contents may be of crucial importance for the design of control systems for e ight vehicles operating in this type of climate, such as sea-skimming missiles.

Evolution of Secondary Electron Emission Characteristics of Spacecraft Surfaces

Helical Motion In this motion, the center of gravity of the  ight vehicle traverses a helical  ight path.The longitudinalaxis of the vehicle is orientated in the same directionas the axis of the helix but displacedfrom it by a constantdistance.A Ž xed body axial spin rate equal to the angular rate of rotation of the helical  ight path is present. The resulting motion is sometimes referred to as lunar motion because the same surface of the body is always orientated toward the central axis of the maneuver; in aircraft maneuver terminology it is equivalent to a barrel roll. A three-dimensional view of the motion is given in Fig. 4. Again, with gravity neglected, we can write u D w D 0; P u D P v D P w D 0, and q D r D 0. It follows that XF D YF D 0 and that ZF D pv. Because is now 0 deg, d3⁄4=dt ; P ®, and R are all zero. This result is again consistent with that predicted by Eqs. (4–6), (10), and (11). Note that, because of the presence of axial spin, this is not the same case as that considered in Ref. 2.