System Noise Assessment of Hybrid Wing–BodyAircraft with Open-Rotor Propulsion

Yueping Guo∗

Boeing Research and Technology, Huntington Beach, California 92647and

Russell H. Thomas†

NASA Langley Research Center, Hampton, Virginia 23681

DOI: 10.2514/1.C033048

An aircraft system noise study is presented for the hybrid wing–body aircraft concept with open-rotor enginesmounted on the upper surface of the airframe. The aircraft chosen for the study is of a size comparable to the Boeing787 aircraft. It is shown that, for such a hybrid wing–body aircraft, the cumulative effective perceived noise level isabout 24 dB below the current aircraft noise regulations of stage 4. Although thismakes the design acoustically viablein meeting the regulatory requirements, even with the consideration of more stringent noise regulations in the nextdecade or so, the design will likely meet stiff competition from aircraft with turbofan engines. The noise levels of thehybridwing–bodydesign are heldupby the inherentlyhighnoise levels of the open-rotor engines and the limitation onthe shielding benefit due to the practical design constraint on the engine location. Furthermore, it is shown that thehybridwing–body design has high levels of noise from themain landing gear, due to their exposure to high-speed flowat the junction between the centerbody and outer wing. To identify approaches that may further reduce noise,parametric studies are also presented, including variations in engine location, vertical tail and elevon variations, andairframe surface acoustic liner treatment effect. These have the potential to further reduce noise, but some of thesetechnologies are only at the proof-of-concept stage.

NomenclatureBPF = blade-passing frequencyD = rotor diameter

I. Introduction

O PEN-ROTOR propulsion is believed to be more fuel-efficientin comparison with turbofan engines and thus has attracted

much attention, especially during time periods when the cost ofaviation fuel is high. The concept, however, facesmany technical andregulatory challenges, one of which is the noise. Open-rotor noisemostly consists of annoying tones that are not attenuated, due to thelack of engine nacelle casing, and thus freely propagate to the far field[1–3]. Before the latest generation of open-rotor designs, it wasknown that aircraft of conventional tube-and-wing design withopen-rotor propulsion would have difficulties in meeting the noiseregulations. Design technologies in recent years have improved theacoustic characteristics of open rotors so that they are now projectedto meet the noise regulations, but only with limited margins, and thusstill may have a competitive disadvantage to turbofan engines. Forexample, the most current open-rotor designs may have a cumulativeeffective perceived noise level (EPNL) of about 15 to 17 dBbelow thenoise regulations of stage 4 [4] for a tail-mounted tube-and-wingconfiguration, which is less attractive, for the noise metric, thanturbofan engines that can give a cumulative EPNL between 15 to20 dB below stage 4. With the concept of a blended wing–body orhybrid wing–body (HWB) aircraft, however, this noise disadvantagemay be overcome by the shielding of the engine noise by the airframe

body, due to the design of mounting the engines above the airframe.This design feature, of course, benefits both open-rotor and turbofanengine applications with the HWB [5–9], but considering that noisemay be a roadblock in the use of open-rotor engines, it is especially ofinterest to assess the acoustics benefit of such designs, and this is theobjective of the study reported here.In this study, a detailed assessment will be given on the system

noise of a HWB aircraft design with open-rotor propulsion. Both theairframe and the open-rotor engine designs, resulting from acomprehensive study [10] with detailed description of the designprocess, follow practical design principles to reflect feasibility forrealistic applications. The designs, however, also include selectedadvanced technologies that are not fully mature now but are in activedevelopment and are expected to mature and/or enter service in thenext decade or so. Thus, the design both satisfies practical feasibility,in aircraft configuration as well as in aircraft operational procedures,and incorporates emerging technologies. It is important for aircraftnoise assessment to be based on realistic configurations becauseachieving noise goals is never the only or the main objective ofaircraft design. Instead, many other important factors such aspropulsion efficiency must be prioritized in the design process tomeet various mission requirements. Because of time constraints inthe design studies reported in [10], therewas no acoustic optimizationin the design process. Instead, general guidelines [11] were followedin the early stages to achieve low-noise features, but the quantitativeassessment of the noise levelswere done after the design, which is thework reported here. Thus, the noise levels reported here canpotentially be improved by optimizing the design with acousticrequirements [12] or by including acoustic considerations at variousstages of the design process with quantitative feedback into thedesign cycle from the acoustic evaluation.For the configuration studied here, it will be shown that the aircraft,

of comparable size to the Boeing 787 aircraft, can achieve acumulative EPNL level about 24 dB below the current noiseregulations of stage 4, a comfortable margin to the current aircraftnoise regulations and to the potentially more stringent regulations inthe next decade or so, which makes the configuration acousticallyviable in meeting current and anticipated noise regulations. The lownoise levels largely result from the shielding effects of the HWBairframe. It is, however, not necessarily a significant competitiveadvantage over conventional designs with turbofan engines. The

Presented as Paper 2015-1215 at the 53rd AIAA Aerospace SciencesMeeting, Kissimmee, FL, 5–9 January 2015; received 26 June 2014; revisionreceived 1 August 2015; accepted for publication 2 August 2015; publishedonline 8 October 2015. This material is declared a work of the U.S.Government and is not subject to copyright protection in the United States.Copies of this paper may be made for personal or internal use, on conditionthat the copier pay the $10.00 per-copy fee to the Copyright Clearance Center,Inc., 222 RosewoodDrive, Danvers,MA 01923; include the code 1533-3868/15 and $10.00 in correspondence with the CCC.

*President; currently NEAT Consulting, 3830 Daisy Circle, Seal Beach,CA 90740. Technical Fellow AIAA.

†Senior Research Engineer, Aeroacoustics Branch, MS 461. SeniorMember AIAA.

1767

JOURNAL OF AIRCRAFTVol. 52, No. 6, November–December 2015

Dow

nloa

ded

by JO

HN

S H

OPK

INS

UN

IVER

SITY

on

Dec

embe

r 28,

201

5 | h

ttp://

arc.

aiaa

.org

| D

OI:

10.2

514/

1.C0

3304

8

latest generations of aircraft in service, such as theBoeing 787 and theAirbus 380, already have cumulative EPNLmargins to stage 4 on theorder of 15 to 20 dB. This margin can be expected to increase forfuture aircraft that are in active development, helped both by betterhigh-lift system design and by more advanced turbofan engines.To identify the challenges and technologies for further noise

reduction, parametric studies will be presented to demonstrate theimpact of various design changes on the total aircraft noise. Theseinclude the variations in the engine locations that have a direct impacton the noise shielding efficiency, the local design features such asvertical tails and trailing-edge elevons that may enhance the HWBnoise shielding, and the concept of acoustic liner treatment on theHWB airframe surface to absorb the acoustic wave impinging on thesurfaces. It will be shown that these concepts all have the potential toreduce noise, and the noise reduction will be quantified as a functionof the design parameters. However, it is important to emphasize thatsome of the concepts are only in their early stage of research, andsome may not be feasible or favorable in aircraft design even whenthe technologies aremature, because of their potential adverse impacton aircraft performance.

II. Baseline ConfigurationThe baseline configuration of the HWB design results from a

comprehensive design study [10], following the best practice inaircraft design as well as incorporating potential technologies that arelikely to mature in the next decade or so. This allows the design tomeet various mission requirements and to achieve a good balancebetween the aircraft system-level goals. The configuration is thebaseline HWB design with three open rotors mounted on the uppersurface of the airframe structure, as illustrated in Fig. 1. The designdetails that affect the acoustic characteristics of the aircraft will bedescribed when discussing the various noise components, whereasthe basic features of the design are summarized in Table 1.An important parameter in engine noise shielding by the HWB

airframe is the locations of the engines. For the baseline config-uration, the design puts the center of the two-stage rotors at 94% of arotor diameter D upstream of the HWB trailing edge (TE). Formaximum noise reduction from shielding, it is intuitive that theengines should be as far away from the edges of the airframe aspossible. This is, however, constrained by the design requirements ofaerodynamic performance. The upper side of the lifting body isdesigned to have high-speed flows at cruise conditions. The intrusionof engines into this high-speed flow region would destroy the flowpattern and severely degrade the aerodynamic performance of theaircraft. The high-speed flow into the engine rotors would alsodecrease the propulsion efficiency. For the baseline configuration, thedesign study reported in [10] has shown that the engines cannot beplaced more than 1D upstream of the trailing edge, to avoid severeinterference with the high-speed flows at cruise. In fact, an enginelocation of about 0.75D upstream of the trailing edge would posemuch less challenges for the aeropropulsion integration. The 0.94D

location has already pushed the aeropropulsion integration to thedesign envelope to maximize the acoustic benefit. It should also benoted that this design guideline applies to not only open-rotor enginesbut also turbofan engines, which is why the engine locations for thevarious aircraft concepts studied in [10] are all within 1D from thetrailing edge.The use of vertical tails in HWB aircraft design is an issue that has

not been satisfactorily resolved. They are thought to be helpful instability control but are often included in the design in the hope ofacoustic benefit; it is intuitive that the vertical tails can block some ofthe noise in the sideline direction. Although the limited data availablehave not been able to conclusively demonstrate the acoustic benefit ofthe vertical tails, their added weight and the cost of aerodynamicperformance has been a concern. It has been estimated that up to 3%of fuel consumption would be needed to have the verticals. Thus, itbecomes really questionable whether the fuel consumption is worththe potential acoustic benefit. For this reason, the vertical tailswill notbe included in the baseline configuration for the acoustic assessment(even though they are included in Fig. 1 for illustration purpose).Instead, their effects on the total aircraft noise will be discussed as anoptional design feature whose utilization will depend on the balancebetween the potential acoustic benefit and the benefit/penalty onother design parameters.Another noise reduction concept that is not included in the baseline

configuration is surface liner treatment. This is a proven technologyfor tone-dominated noise, as is the case for open-rotor engines; how-ever, acoustic liners are typically applied inside engine casings. Thelack of engine casing for the open rotors naturally leads to the idea ofusing the airframe surfaces in the vicinity of the engine installation todeploy liners to attenuate the noise from the rotors before itpropagates to the far field. The obstacle for this application, however,is the potential for increased drag; the airframe surfaces are allexposed to the external aerodynamic flows and hence all requireminimum drag. It is not clear at this time how much additional dragthe liner treatment will induce. If it turns out to be materiallynonnegligible, the critical step for developing this technology for thisapplication would be a dragless, or minimum-drag, liner design. Forthis reason, the surface liner treatment is considered as an early-stagetechnology for future development, and it is excluded from thebaseline configuration because it is now only at the proof-of-concept stage.Similar care must also be exercised in defining baseline tech-

nologies for airframe noise reduction. In the past two decades or so,there has been extensive research in developing noise reductiontechnologies for airframe noise components, including concepts suchas cove filler and sealed gap for slats [13,14], fences and continuousmold line for flap side edges [15–17], and fairings of various kinds forlanding gears [18–20]. None of the concepts, however, has made itsway to current production aircraft. Considering the long lead timeneeded to develop and mature technologies in aircraft industry, thisdoes not necessarilymean that none of the conceptswill eventually beviable for practical applications, but it is instructive and helpful toexamine the obstacles that have prolonged the transition of theseconcepts to reality. These obstacles include the added weight to theaircraft, the cost of implementation and maintenance, and the lack ofrobustness in noise reduction. Some of these may be resolved in thenext decade or so, but not all of them. For the baseline HWBconfiguration considered here, slat noise reduction is considered inFig. 1 HWB configuration with open-rotor engines.

Table 1 Characteristics of baseline HWB design

Parameter ValueMaximum takeoff weight 419,848 lbCruise Mach number 0.85Wing span 229.3 ftReference wing area 8048 ft2

Wing aspect ratio 5.62Reference thrust 138,000 lbNumber of engines 3Engine diameter D 182 n.Engine position to HWB trailing edge 0.94D

1768 GUO AND THOMAS

Dow

nloa

ded

by JO

HN

S H

OPK

INS

UN

IVER

SITY

on

Dec

embe

r 28,

201

5 | h

ttp://

arc.

aiaa

.org

| D

OI:

10.2

514/

1.C0

3304

8

the form of Krueger flaps including with a sealed slat gap. TheKrueger flaps are needed for the implementation of hybrid laminarflow control, to protect the control actuators. They are also likely tohave acoustic benefits because of the reduced cove region flowseparation. This has recently been demonstrated on a conventionalwing [21] where an optimized Krueger wing can lower the slat-related noise by asmuch as 5 dB.Other technologies are also feasible,for example, sealing the slat gaps at normal operation conditions butopening up the gap at extreme conditions when maximum lift isrequired, which has been an active research topic for many years[14,22–24]. The noise reduction potential of this technology has beenwell demonstrated, and the implementation issue is the actuation toopen the gap and the added weight for the actuation system.These projected technologies make the HWB slat noise very low,

in comparison with other noise components, as will be seen in latersections, because the Krueger flaps efficiently reduce the cove noiseand the sealed gaps effectively eliminate the gap noise. Technologiesfor reducing landing-gear noise are also expected to be available inthe next 10 to 15 years, and thus, some are included in the baselineconfiguration here. The technologies may be in the form of localfairings, redesign of the gear parts, and overall fairings, all of whichhave been actively researched in recent years.In aircraft system noise assessment, flight operational conditions

play an important role because the flight parameters such as the flightMach number and the aircraft angle of attack (AOA) determine thenoise source levels, and the flight path determines the distance of thenoise propagation and hence the amplitude of the noise received at themeasurement locations. Similar to aircraft design, flight proceduresmust also follow practical requirements, set by regulatory rules forsafety and/or by airport authorities for operation efficiency. Forexample, current airport practice requires an aircraft to approach forlanding at a 3 deg flight path angle. Following rules such as this, theflight profile for the baseline configuration is designed [10] andillustrated in Fig. 2 for all three conditions of aircraft noisecertification, with the upper diagram plotting the flight altitude as afunction of the distance from brake release and the lower diagramplotting the flight velocity. The three noise certification conditionsare all shown in the figures as conventionally done; the first segmentrepresents the approach condition with decreasing altitude andvelocity, the short second segment is for ground operation, the thirdsegment with increasing altitude and velocity represents normal

takeoff when the sideline noise ismeasured, and the fourth segment isfor engine power cutback operations with reduced climb rate.Although it is not obvious from the flight altitude shown in the

upper diagram of Fig. 2 how the flight profile differs from that ofconventional aircraft operations, the velocity profile in the lowerdiagram clearly illustrates one of the features of the HWB design; ithas better lift characteristics so that it can take off and land at lowervelocities. For example, the approach velocity of 146 kt for the HWBis probably about 10 to 15 kt lower than the velocity of a comparableconventional aircraft. This feature will also clearly manifest itself inthe aircraft angle of attack and its total engine thrust, respectivelyshown in the upper and lower diagram of Fig. 3, for the three noisecertification conditions. For conventional aircraft, the angle of attackat takeoff and landing operations is usually in the range between 4 to8 deg. For the HWB aircraft, it can be as high as 12 deg. Because ofthe lower flight velocity and higher angle of attack, the HWB aircraftcan supply a large amount of lift, reducing the requirement for enginepower, especially at approach conditions. For example, the totalengine thrust of about 3000 lb for the HWB to land is about onequarter of what is required for a conventional aircraft of comparablesize. These have direct acoustic effects; the low flight velocity leadsto low airframe noise, and the low engine thrust corresponds to lessengine noise, which will all be included in the noise assessmentpresented in later sections. The flight conditions are also summarizedin Table 2.It should be pointed out that the aircraft take different

configurations in the three conditions for noise assessment, namelyat approach, cutback, and sideline conditions. These include landinggears (deployed at approach but retracted at takeoff), slats (open gapat approach but closed gap at takeoff), and the various operationparameters listed in Table 2. These differences in configurations areall considered in the noise assessment.

Distance (ft)

Alti

tude

(ft)

×103

-20,000 0 20,0000

1

2

3

ApproachTakeoff

Cutback

Distance (ft)

Vel

ocity

(kt)

-20,000 0 20,000120

140

160

180

Fig. 2 Flight altitude and velocity for acoustic analysis, as a function ofthe distance from brake-release.

Distance (ft)

AO

A(d

eg)

-20,000 0 20,0008

10

12

14

Distance (ft)

Thr

ust(

lb)×

104

-20,000 0 20,0000

2

4

6

8

10

Fig. 3 Aircraft angle of attack (upper) and engine thrust (lower) for theHWB aircraft acoustic analysis.

Table 2 Flight parameters at noise certificationconditions

Parameter Approach Cutback SidelineAltitude, ft 400 2099 1000Speed, kt 146.1 161.9 159.4Angle of attack, deg 11.06 12.65 12.26Thrust, lb 3087 46590 95064

GUO AND THOMAS 1769

Dow

nloa

ded

by JO

HN

S H

OPK

INS

UN

IVER

SITY

on

Dec

embe

r 28,

201

5 | h

ttp://

arc.

aiaa

.org

| D

OI:

10.2

514/

1.C0

3304

8

III. Analysis MethodologyAlthough there are commonly used and validated tools for system

noise assessment of conventional aircraft configurations, such as theNASA tool package Aircraft Noise Prediction Program (ANOPP)and the Boeing in-house tool Modular Component Prediction,acoustic tools for advanced aircraft configurations such as HWB arestill in the development stage. Predictions of open-rotor engine noisecurrently rely heavily on very limited data with strong empiricalnature, and the empirical engine noise tools are mostly standaloneand are not incorporated in system noise assessment tools. For enginenoise shielding by the airframe structures, though much effort hasbeen made to develop prediction tools in recent years, these tools arelargely in development, some of which are still to be validated andothers are limited by their heavy computation resource require-ments [5–8]. Thus, noise shielding for full-configuration aircraft canonly be dealt with on the empirical basis by using wind-tunnel testdata, which is also a standalone process outside any system noiseprediction tool package.Thus, it is of interest to describe the process used in the study

presented here for system noise analysis, which is a combination ofempirical prediction, component modeling, and local featurenumerical computation. The process is illustrated in Fig. 4, consistingof the following elements. The acoustic analysis process starts withthe design specifications, including the design of the airframestructure, engine type and power setting, and flight profiles. Thedesign specifications are the basis for the noise component sourcelevel analysis for both the propulsion system and the airframestructure, and the flight profiles determine the operation conditions atthe noise certification points. From the aircraft design, propulsionsystem definitions are used to establish the engine noise source levelsfor all major engine noise components. For open-rotor engines, theycontain tones from the front rotor, the rear rotor, and the interactionsbetween the two, as well as the broadband noise component of therotor system. The component source levels also include those of theairframe structures, namely the landing gears, the leading-edge slats,and the trailing-edge elevons. The component noise source levels arefor individual, isolated components, which need to be assembledunder the constraints of the particular aircraft configuration to takeinto account the intercomponent interactions and shielding. This is amajor feature of the HWB configuration where the engine noise maybe significantly shielded by the airframe structure, effectivelyreducing the source levels on the far-field radiation. Far-field noisecan also be significantly affected by local flow changes, which areconfiguration-dependent, such as the landing-gear locations. The far-field noise from all the components is assembled into the total aircraftnoise on an energy basis, meaning that the acoustic energies from allthe components are summed together incoherently without consid-ering the potential acoustic interactions between the components inthe far field. The far-field total noise levels are then used to compute

the standard noisemetrics, such as the tone-corrected perceived noiselevel (PNLT) and the effective perceived noise level (EPNL).In the acoustic analysis procedure described previously, the

predictions of the airframe noise components, namely the leading-edge slats, the landing gears, and the trailing-edge elevons, are basedon methodologies developed for airframe noise components ofconventional aircraft [21–28]. These prediction models can beapplied here for the HWB design because the basic elements of theprediction methodologies, such as the spectral features, the Mach-number dependence, and the far-field directivity of the radiated noise,are based on the fundamental theory of aerodynamic sound gener-ation [11] that captures the flow physics. The predictions are allcomponent-based, with the noise prediction as a function of the localfeatures of the individual components, rather than the overall designof the aircraft. Thevalidation and calibration of the predictionmodelsare also done for the individual components and are not anchored toany particular aircraft type. This allows the models to have wide androbust applications. The HWB aircraft configuration, however, doeshave features different from conventional designs, an example beingthe large angle of attack at both takeoff and landing configurations.Slat noise database for conventional aircraft usually does not extendto angles of attack much higher than 8 deg, and the prediction ofHWB slat noise needs to be validated and calibrated for large anglesof attack. This is done by using the database reported in [29], where amodel-scale HWB was tested in the Boeing Low Speed Aero-acoustics Facility (LSAF) with variations in the aircraft angle ofattack covering the entire range of operations, from zero to 15 deg.The original prediction model is then extended to cover all the anglesof attack for the slat noise.Another feature of the HWB design that is different from

conventional aircraft is the locations of themain landing gears, whichaffect the noise because the local geometry determines the local flowvelocity that in turn sets the flow-dependent part of the noise ampli-tude. For conventional aircraft, the main landing gears are located inlow-speed flows under the lifting wings so that the locations areacoustically advantageous [30,31]. In comparison, the HWB mainlanding gears are in relatively high-speed flows without takingadvantage of the circulatory flow around the wings because they arelocated at the junctions between the HWB centerbody and its outerwings where the flows accelerate. For landing-gear noise prediction,the sound generation mechanisms are all modeled with the incomingflow velocity as an input parameter. Thus, the landing-gear noisemodel is readily applicable, provided that the local flow velocity issupplied. This is done by using computational-fluid-dynamicsmethod to calculate the mean flow for the HWB aircraft config-uration, from which the local flow velocities are extracted for thelanding-gear noise prediction.The prediction of engine noise is another critical element in the

acoustic analysis methodology. In addition to the prediction of thenoise source levels of the open-rotor engines, the effects of the HWBairframe on the engine noise, namely the shielding effects, must beaccurately and realistically accounted for because the engine noisereduction due to the HWB shielding is the main factor that maypotentially make the HWB with open-rotor propulsion acousticallyviable as a candidate for future commercial aircraft. The engine noisesource levels, including the tones and the broadband components, arepredicted by empirical methods, calibrated by the data reported in[10]. Because of the lack of full-scale open-rotor noise tests anddatabase, especially for more-advanced designs developed in recentyears, the predictions can only be calibrated with limited small-scalewind-tunnel test data. The projection of the predictions to full scale,however, has shown good consistency with other independentmethods and database [4], which all indicate that the more recentdesigns of open-rotor engines may meet the aircraft noise regulationof stage 4 with about 15 to 17 dB cumulative EPNL margin. Thisnumber is of course only a reference, which may somehow vary fordifferent designs and different studies. It is quoted here only to showthe consistency of the open-rotor noise levels used here with others.The effects of the HWB shielding on the open-rotor noise are

obtained from a database resulting from a model-scale wind-tunnelexperiment in the Boeing LSAF reported in [32]. The method for

Component Source Definition

Geometry and Operation Conditions

Design Specification

Incoherent Summation

Noise Metrics

Component Noise Prediction

Y N Comparison Study

Inter-Component Interaction/Shielding

N Y

Test Data and Empirical Modeling

Isolated Engine Noise

Fig. 4 Illustration of acoustic analysis process.

1770 GUO AND THOMAS

Dow

nloa

ded

by JO

HN

S H

OPK

INS

UN

IVER

SITY

on

Dec

embe

r 28,

201

5 | h

ttp://

arc.

aiaa

.org

| D

OI:

10.2

514/

1.C0

3304

8

extracting and applying the shielding effects in this system noiseassessment is described in [33]. This is a huge database coveringparametric variations in flow conditions, HWB airframe designfeatures, engine power settings, rotor operational conditions, engine–airframe integration configurations, and noise reduction concepts,withmeasurements on the aircraft, in the near-field flow, and in the farfield. The processing and analysis of the database is beyond the scopeof the work discussed here, but some aspects should be discussed,which are related to the system noise prediction and may havesignificant impact on the assessment methodology and results.Because of the tonal nature of the open-rotor noise, the application

of the noise shielding effects from a wind-tunnel test to full-scaleengines needs to be on the individual tones, which is different fromthe case of turbofan engines dominated by broadband jet noise,wherethe noise shielding is usually applied on the 1/3-octave band spectra,once the power settings are matched between the full-scale enginesand the scaled-up wind-tunnel test engines. This is because turbofanengines dominated by jet noise follow similar acoustic behavior thatis in turnmostly dominated by the engine power settings.On the otherhand, the acoustic characteristics of open-rotor engines, their tonalfrequencies, the tone amplitude distributions, and the directivities ofthe tones, are critically determined by the detailed designs andoperating conditions of the open rotors. Thus, there is no guaranteefor matching acoustic characteristics between two different designs,even if they can be operated to have the same power outputs. Ofcourse, if the engines in the wind-tunnel test have the same design asthe full-scale engine, the acoustic features of the two would bescalable, and the shielding effects can be applied on the 1/3 octaveband spectra. This is, however, not likely to be the case in manysituations, especially when the technologies of open-rotor design arestill in development. Instead, costly wind-tunnel tests may be doneusing a generic rotor design, and the results should be suitablyprocessed and applied to various full-configuration designs. In thisapproach, the open rotors in the wind-tunnel test serve, among otherreasons, as a realistic database of tones to gather shielding data at theindividual tone frequencies. The shielding effects on the individualtones are relevant to other engines as long as the directivities of thetones individually match those of the tones in the full-scale engines,even if the overall acoustic characteristics of the two are different, asanalyzed in detail in [33].

IV. Engine Noise ShieldingFor the case studied here, the shielding effects are derived from a

scale-model test in the Boeing LSAF wind tunnel. The matchingbetween the full-scale HWB aircraft and the LSAF scaled model isillustrated in Table 3, where various geometric and fan designparameters are listed for both configurations. The first parameterlisted in Table 3 is the fan diameters, the ratio of which gives thescaling factor of 0.066 that is used to scale the frequencies so that theStrouhal numbers are matched between the two cases. The next twoparameters are the fan blade counts, which determine the tonalfrequencies of the rotors. Although the aft fan blade numbers for thetwo cases are close, the forward fans differ significantly. Preferably,the blade counts of the two cases should be equal so that the blade-passing frequency (BPF) and its high-order harmonics would bescaled by the fan speeds. Because this is not the case here, the tonal

frequencies are scaled by the Strouhal number, and the shieldingeffects are applied to the tones regardless of their origin andengine order.The next two parameters in the table are the distance from the

engine center to the HWB airframe trailing edge and that to itsupper surface, respectively. These two parameters are critical indetermining the efficiency of the shielding because, together with thefan diameters, they determine the shielding, transitional, andinsonified regions, as illustrated in Fig. 5. Once the frequencies arescaled by the Strouhal number, these two geometric parametersshould be scaled in terms of the fan diameter. As shown in Table 3, thetwo parameters arematchedwell between the full-scaleHWBaircraftand the LSAF test setup. For engine locations that are not in thewind-tunnel test database, a comprehensive data analysis and empiricalmodeling should be able to make the data useable, by interpolation,for example. In the study presented here, Table 3 shows that theengine location of the HWB design matches one of the scale-modeltest configuration so that interpolation is not needed.The last two parameters in the table are for the vertical tails, which

will be discussed in a later section as a potential noise reductiondevice. The comparisons are given here for convenience. For both thefull configuration and the model-scale test, the heights of the verticaltail are about one fan diameter, giving a closematch between the two.The vertical cant angles, however, are different between the two; thevertical tails of the full configuration are canted 15 degmore outwardfrom the engines. Although it is believed that large cant angles aremore efficient in reflecting the noise upward, the impact of the angledifference between 45 and 30 deg has not been quantified and willneed to be studied in the future. For the work reported here, thisdifference will simply be accepted with the understanding of theuncertainty and potential impact.Once the full-configuration design is matched with a wind-tunnel

test configuration, the data from thewind-tunnel tests are extracted toderive the shielding effects, respectively for the three operationconditions. This is done by scaling themodel-scale test engine powerto full configuration and matching the individual conditions. At eachoperating condition, the shielding effects are further decomposed intotones from the front rotor, the rear rotor, and the interactions of thetwo. This decomposition is necessary because each group of toneshas its own distinctive patterns and far-field directivity, which cansignificantly affect the shielding results. Some examples areillustrated in Fig. 6, which plots the noise reduction due to shieldingfor the tones of the full-scale engine, located off the centerline of theHWB airframe for the baseline configuration. The data are derivedfrom the model-scale LSAF test with the shielding effectsextrapolated to full scale. The two diagrams are respectively for theoverhead and the sideline direction. The figure plots the difference insound pressure level (SPL) between isolated and installed engines asa function of the emission angle and frequency. For the engine locatedon the HWB airframe centerline, the results are similar butquantitatively different.From the results shown in Fig. 6, some representative trends can be

derived. For both the overhead and the sideline direction, theeffectiveness of noise shielding increases with frequency, indicated

Table 3 Parameters between the full-scale HWB aircraftand the LSAF model

Parameter Full-scale HWB LSAF modelFan diameter, in. 182 12Forward fan blade count 12 8Aft fan blade count 9 8Aft rotor cropping Yes NoEngine distance to HWB TE 0.94D 1DClearance (rotor tip to airframe) 0.2D 0.25DVertical tail height 1D 1DVertical tail angle, deg 45 30

InsonifiedRegion

AOA

Source Distribution

Flight Direction

Partial Shielding

Shielded Region

Fig. 5 Definition and acoustic features of shielding angles.

GUO AND THOMAS 1771

Dow

nloa

ded

by JO

HN

S H

OPK

INS

UN

IVER

SITY

on

Dec

embe

r 28,

201

5 | h

ttp://

arc.

aiaa

.org

| D

OI:

10.2

514/

1.C0

3304

8

by the angular widening of the high shielding area with frequency.This is the well-known feature of better shielding of shorter soundwaves associated with higher frequency for fixed source position.The heavily shielded areawith the high shielding area is mostly in theforward quadrant because of the engine location near the trailing edgeof the HWB airframe, which also leads to the result of basically noshielding at large emission angles in the aft quadrant, namely, theinsonified region after about 120 deg for the overhead case and afterabout 105 deg for the sideline case. This also states that the overheaddirection sees more shielding than the sideline direction, a featurereadily explainable by the shape of the HWB airframe. The enginesare located near the airframe centerline where the chord length islargest, which causes a large angular domain with blocked soundpropagation if the propagation is downward, namely, in the overheaddirection. The waves propagating in the sideline direction, however,encounter a shielding surface with smaller chord length because ofthe tapering of the wing design. Furthermore, the trailing edge of theHWB curves in the upstream direction along the span, which leads toan unblocked area for the sound waves to propagate to the sidelinewithout interference with the airframe and is the reason for theforward shift of the shielding angle in the results shown in the twodiagrams in Fig. 6.When using the wind-tunnel test data to extract the shielding

effects of the tones, the remaining spectra after the tone extractionalso gives the shielding effects for the broadband noise of the rotorengines,which are scaled to full scale by the engine size ratio and thenapplied to the full-scale configuration for 1/3 octave bands from 50 to10,000Hz. Examples of the broadband shielding effects are shown inFig. 7, with the difference in the 1/3 octave band SPL plotted as afunction of the emission angle and the logarithmic of frequency forengines located off the HWB centerline. Clearly, the amount ofbroadband shielding is different from tonal shielding, but thefunctional trends in frequency and emission angle hold for both tonaland broadband shielding, as evidenced by the similar featuresrevealed in Figs. 6 and 7.

V. System Noise of Baseline ConfigurationThe system noise study starts with the baseline configurationswith

both the propulsion systems and the airframe configurations asdesigned, details of which are described in [10]. The designs followsome general principles to minimize noise, but there is no acousticoptimization and parametric studies. Instead, all the configurationsare designed to meet propulsion requirements. Thus, the acousticanalysis serves only as assessments of these as-designed config-urations, implying the potential of further noise reduction withoptimization studies within the requirements of other design criteria.This is also the case for noise reduction technologies; the baselinedesigns do not include various noise reduction concepts that arecurrently in development. An example is landing-gear fairings,which, although still facing significant hurdles in practical imple-mentations, have been demonstrated to have noise reduction potential[18–20]. The fairings can be implemented in various forms, fromsmall local fairings (to cover some particular gear parts) to overallfairings (to completely shield the gear from the incoming flows).These come with varying degrees of difficulty in practicalimplementation and will of course result in varying degrees ofsuccess in noise reduction. The baseline HWB considered here takessome credit for gear noise reduction, corresponding to local fairings.One reason for focusing first on the acoustic assessment of the

baseline configuration is to highlight the design that is considered tobe most practically feasible within the 2025 timeframe. Anotherreason for the baseline acoustic analysis is to demonstrate themultiple paths to achieve further noise reduction goals. With thebaseline configuration as the starting point, further noise reductioncan be projected from various combinations of approaches rangingfrom configuration optimization, to component noise reduction, tolow-noise operation procedures. The selection of the combinationwill most likely depend on other practical constraints in the aircraftdesign and the maturity of the individual technologies. It alsoillustrates the necessity of this multipath approach in achieving noisereduction goals such as the NASA N+2 (NASA terminology for

Frequency (Hz)

Em

issi

onA

ngle

(Deg

)

500 1000 1500

30

60

90

120

150dB

0-2-4-6-8-10-12-14-16-18-20

Approach and Cutback

Frequency (Hz)

Em

issi

onA

ngle

(Deg

)

500 1000 1500

30

60

90

120

150dB

0-2-4-6-8-10-12-14-16-18-20

Sideline

Fig. 6 Tonal noise shielding for engines located off the HWB center forthe baseline configurationwith emission angles from 0 to 90 deg being theforward quadrant.

Log(Frequency) (Hz)

Em

issi

onA

ngle

(Deg

)

2 2.5 3 3.5 4

30

60

90

120

150dB

0-2-4-6-8-10-12-14-16-18-20

Approach and Cutback

Log(Frequency) (Hz)

Em

issi

onA

ngle

(Deg

)

2 2.5 3 3.5 4

30

60

90

120

150dB

0-2-4-6-8-10-12-14-16-18-20

Sideline

Fig. 7 Broadband shielding for engines located off the HWB centerlinefor the baseline configuration.

1772 GUO AND THOMAS

Dow

nloa

ded

by JO

HN

S H

OPK

INS

UN

IVER

SITY

on

Dec

embe

r 28,

201

5 | h

ttp://

arc.

aiaa

.org

| D

OI:

10.2

514/

1.C0

3304

8

aircraft technology entering service in 2025) noise goal; an individualtechnology, even if optimized, may not be able to achieve such a goal,due to the system nature of the aircraft design.With the methodologies and data described in the previous

sections, the noise metrics for the baseline HWB are calculated andsummarized inTable 4,which lists the limits of the noise regulation ofstage 3 at the three certification conditions and their cumulative valueas the first row of data in the table. The EPNL for the baseline HWB isthen given in the second row, which leads to the margins given in thenext two rows, respectively in reference to the regulation limits ofstages 3 and 4, with the latter being the current regulation for aircraftcertification. Clearly, this is a quiet aircraft, with about 24 dBcumulative EPNL margin to stage 4. As a reference comparison, thelatest generation of commercial aircraft currently in service, theBoeing 787, for example, which has the conventional design withturbofan engines and has comparable takeoff weight to the HWBconfiguration studied here, has cumulative EPNL about 16 to 20 dBbelow stage 4. The low noise levels of the HWB design become evenmore impressive when considering the HWB design uses open-rotorengines; for the latest advanced rotor designs installed on aconventional tail-mounted tube-and-wing aircraft, Khalid et al. [4]have reported a cumulative EPNL margin to stage 4 of about 15 to17 dB, much less than the 24 dB for the HWB design.The low noise levels of the HWB aircraft are achieved mostly by

the shielding of engine noise by theHWBairframe structure. This canbe demonstrated by a reference comparison between theHWBdesignand a hypothetical aircraft that uses the same open-rotor engines butdoes not provide any engine noise shielding. To be relevant, thehypothetical aircraft is assumed to have the same levels of airframenoise components as the HWB aircraft. It is also assumed to operatewith the same flight profiles. The noise levels of this hypotheticalaircraft are compared with the HWB aircraft in Table 5, showing thesignificant benefit of the HWB noise shielding. The shieldingbenefits all three certification conditions with noise reduction of 2.5,3.8, and 4.3 dB, respectively for the approach, the cutback, and thesideline condition. It should be pointed out that this hypotheticalaircraft is used here only for the purpose of demonstrating the effectsof noise shielding by theHWBairframe, on the basis of EPNL. It is byno means an indication of the noise levels for conventional aircraftdesigns with open-rotor engines, which have drastically differentairframes and, thus, very different airframe noise levels. Because ofthe differences in airframe design, conventional aircraft also operatewith flight profiles different from those of the HWB aircraft, such asthe flight Mach number and aircraft angle of attack. Furthermore, forengines installed under thewings of conventional aircraft designs, theinstallation effects can also have significant effects on the totalaircraft noise.From Table 5, it can be seen that, though the HWB shielding is

largely responsible for the comfortable EPNL margin of the baselineconfiguration, the advanced rotor design itself also has good acousticcharacteristics. Even without the HWB shielding, the hypotheticaircraft would still meet the stage 4 regulatory noise requirements,

with a noisemargin of EPNL about 13.4 dB, consistent with the resultreported in [4]. The combination of advanced rotor design and HWBshielding is so efficient in reducing the engine noise at approachconditions that the engine noise is no longer the dominantcomponent. This can be seen from the component decompositionshown in the top diagram of Fig. 8, which plots the tone-correctedperceived noise level (PNLT) as a function of the observer time forboth the airframe and the engine noise component as well as the total.In this case, the engine noise, indicated by the dashed line, still hasnoticeable contributions in the aft quadrant, namely at large observertimes after the peak noise point in the diagram, but the dominantcomponent elsewhere is the airframe, shown by the triangle curve inthe figure. The other two diagrams, themiddle and the bottom one, inFig. 8 plot the noise decomposition in PNLT respectively for thecutback and the sideline conditions. It can be seen that, for all threeconditions, the engine noise components are significantly reduced inthe forward quadrant, namely at observer times before the peak noisetime in the figure when the observer is ahead of the aircraft. Theengine noise levels are much lower on the left side of the peak noisepoints, compared with the levels on the right side. This is expectedbecause the engines are located close to the trailing edges of theHWBairframe so that most shielding occurs in the forward directions. Thefigure also shows that the engine noise from the open rotors, thoughbenefitting from theHWB shielding, is still a dominant contributor tothe total aircraft noise, especially in the aft quadrant where shieldingis limited. This is true for all three conditions, but especially atsideline and cutback.

Table 4 Acoustic results for the baseline configuration

Parameter Approach Cutback Sideline CumulativeStage 3 EPNL limits, dB 103.7 99.9 100.3 303.9Baseline EPNL, dB 94.5 86.2 89.2 269.9Margin to stage 3, dB 9.2 13.7 11.0 34.0Margin to stage 4, dB — — — — — — 24.0

Table 5 Comparison between HWB and a hypothetical aircraftwithout noise shielding

Parameter Approach Cutback Sideline CumulativeStage 3 EPNL limits, dB 103.7 99.9 100.3 303.9HWB EPNL, dB 94.5 86.2 89.2 269.9Hypothetical aircraftEPNL, dB

97.0 90.0 93.5 280.5

Shielding effect, ΔdB 2.5 3.8 4.3 10.6

Observer Time (s)

PNL

T(d

B)

185 190 195 200 205 210 21560

70

80

90

100 TotalAirframeEngine

Approach

Observer Time (s)

PNL

T(d

B)

15 20 25 30 35 40 4560

70

80

90

100TotalAirframeEngine

Cutback

Observer Time (s)

PNL

T(d

B)

15 20 25 30 35 40 4560

70

80

90

100 TotalAirframeEngine

Sideline

Fig. 8 PNLT, as a function of flight time, for the baseline HWBconfiguration at each of the three noise certification conditions.

GUO AND THOMAS 1773

Dow

nloa

ded

by JO

HN

S H

OPK

INS

UN

IVER

SITY

on

Dec

embe

r 28,

201

5 | h

ttp://

arc.

aiaa

.org

| D

OI:

10.2

514/

1.C0

3304

8

For the approach conditions when the airframe is the majorcontributor to the aircraft total noise, the noise decomposition can becarried out further for all the airframe noise components, as shown inFig. 9, where the main landing gear is seen to be the dominantcomponent. This relatively high noise level is the result of high localvelocities at the main landing-gear location. As designed, the maingears are positioned at the junction between theHWBcenterbody andits outerwings,where the sectional lift is very small, corresponding tosmall circulation at this spanwise location, which in turn makes theflow velocity under the airframe close to the freestream velocity. Incomparison with conventional aircraft with the main landing gearslocated under wing sections with large sectional lift, the local flowvelocity under the wing is usually reduced by as much as 20% by thecirculatory flow that is in the opposite direction as the freestream.Because the landing-gear noise is proportional to the sixth-power lawin the flow Mach number, a 20% reduction in flow velocitycorresponds to about 6 dBnoise reduction. TheHWBdesign does nottake advantage of this velocity reduction and thus suffers fromrelatively high landing-gear noise.The dominance of the main gear noise is in part a function of the

relatively low amplitudes of the other airframe components,especially the slat noise component that is usually somewhatcomparable to landing-gear noise for conventional aircraft. Therelatively low slat noise results from the use of Krueger flaps,minimizing large-scale cove region flow separation, thus reducing itsnoise. The effect is equivalent to the slat noise reduction technique ofa cove filler. The slat noise is also low because sealed slats areassumed for theHWBdesign,whichmaybe implemented in the formof hinged Krueger flaps. The elimination of the slat gap significantlyreduces the slat noise, which has been experimentally demonstratedin the past (e.g., [13,14]). It is interesting to point out that the HWBslat noise would be much higher, probably comparable to or higherthan that of a conventional aircraft design of comparable size, had thedesign used conventional slotted slats, as analyzed in [29]. Thehinged Krueger flaps used here were originally designed to protectlaminar flow control devices, but they could turn out to be of someacoustic advantage as well. It should also be pointed out that, for bothconventional slotted slats and Krueger flaps, the noise modeling andprediction are for the slats themselves without considering the effectsof the slat brackets. Currently, the noise from the slats themselves isconsidered to be much higher than that from the supporting brackets,so that the bracket noise is ignored, similarly assumed for otherairframe noise components. For Krueger flaps, the number ofbrackets and the complexity of the brackets may be very differentfrom those for conventional slotted slats, which would then representa new noise component. This has not been studied so far becauseKrueger flaps have not been widely used in the current generation ofcommercial aircraft, and there are no data available for even an orderof magnitude assessment, a situation that certainly needs to berectified if Krueger flaps become the choice of future aircraft design.From the results discussed previously, it is clear that the noise levels

of the open-rotor-powered HWB aircraft benefit significantly from thenoise reduction provided by the noise shielding of the HWB airframe,which gives the aircraft a comfortable EPNL margin of about 24 dBrelative to stage 4. Even with the potential of a more-stringent

regulation in the next decade or so, thiswould still be sufficient tomeetthese regulatory requirements. Thus, purely from the acoustics point ofa view, theHWB aircraft with open-rotor propulsionwould be a viablecandidate as a commercial aircraft. This, however, does not necessarilyimply the commercial competiveness of such aircraft design. In fact, itis likely to meet stiff competition from turbofan-powered aircraft,either conventional or HWB design, because of their acousticadvantages.For example, the latest generations of conventional aircraftin service already have cumulative EPNL margins to stage 4 in therange of 15 to 20 dB, which is not as much as the 24 dBmargin for theHWB aircraft, but the noise levels of the conventional aircraft areachieved with current designs and technologies, whereas the HWBaircraft is projected to mature in a decade or so. There is no doubt thatconventional aircraft design will also advance in the timeframe, andnew technologies will make its noise levels lower. This will definitelyput competitive pressure on the HWB aircraft; a noise advantage ofabout 4 dB in cumulative EPNL for a future design can be definitelyconsidered as a competitive risk.It is also obvious that the noise levels do not meet the NASA N+2

noise goal of 42 dB below stage 4. There are various reasons for thisrelative status including the high source noise levels of the open rotorengines compared with turbofan engines, the limited shieldingefficiency of low frequency tones with the engines mounted close tothe HWB trailing edges, and the high airframe noise components ofthe HWB design. These reasons also point to directions to improvethe design and further reduce the noise levels, whichwill be discussedin the following sections.

VI. Effects of Engine LocationsThe most influential parameter for the efficiency of HWB engine

noise shielding is probably the engine location. It is also known thatthe engine locations are constrained by factors such as aerodynamic-propulsion integration, aircraft weight balance, and stability control.The baseline design discussed in previous sections with the enginesinstalled at 0.94D upstreamof the trailing edges of theHWBairframeis probably pushing the design envelope; this location is approachingthe edges of the high-speed flow zone on the upper side of the aircraft.This is a situation to be avoided in aircraft design because theintrusion of the engines into the high-speed flows would not onlydestroy the flow pattern, severely degrading the aerodynamicperformance of the design, but also expose the engine to the high-speed inflow, decreasing the engine propulsion efficiency. It ispossible that, to ensure feasible aerodynamic propulsion integration,the engines may have to be located within 0.75D upstream of thetrailing edge. It should be realized that this is the design status withcurrent and near-term technologies of aeropropulsion integration;future development of advanced integration technologiesmay be ableto extend this design envelope to allow engine/airframe integrationswith better acoustic benefits. For this reason, it is instructive to assessthe effects of engine location on the total aircraft noise. Thisquantitative trend between the total noise and the engine location canbe used to perform trade studies to achieve a best balance betweennoise and other design parameters.To this end, Table 6 shows the results of at various engine locations,

measured by the rotor diameter, with zero indicating the engines

PNL

T (d

B)

Observer Time (s)185 190 195 200 205 210 215

60

70

80

90

100 NoseGearMainGearSlatTE

Fig. 9 PNLT for airframe components for the baseline HWB design.

Table 6 EPNL (in decibels) for various engine locationsin the flow direction

Engine positionfrom trailingedge Approach Cutback Sideline Cumulative

Margin tostage 4

−0.5D 97.2 91.1 94.2 282.4 11.5−0.25D 97.1 91.1 94.0 282.2 11.70D 96.3 90.3 93.8 280.3 13.60.25D 96.5 89.7 93.2 279.3 14.60.5D 95.2 87.9 91.8 274.9 19.00.75D 95.7 87.2 90.4 273.3 20.71D (baseline) 94.5 86.2 89.2 269.9 24.01.5D 93.6 82.6 87.7 263.9 30.02D 93.2 79.2 85.9 258.3 35.6

1774 GUO AND THOMAS

Dow

nloa

ded

by JO

HN

S H

OPK

INS

UN

IVER

SITY

on

Dec

embe

r 28,

201

5 | h

ttp://

arc.

aiaa

.org

| D

OI:

10.2

514/

1.C0

3304

8

being at the trailing edge, positive numbers for positions upstream ofthe trailing edge, and negative numbers for positions downstream ofthe trailing edge. For all the cases in this table, the aircraftconfigurations and flight conditions are the same as those for thebaseline configuration discussed in the previous section. The mostobvious trend shown in this table is the increase in noise shielding asthe engines move upstream of the trailing edge, as can be intuitivelyexpected because of the increase in shielding surface area. The resultsalso show that the variations in shielding effects are not linear.Instead, the amount of shielding varies very gradually when theengines are behind the trailing edge but increases significantly as theengines move farther upstream of the trailing edge, beyond one rotordiameter. The increase in shielding as a function of the enginepositions is not uniform for all three certification conditions either.The increased shielding tends to benefit the cutback condition themost. These features are also illustrated in Fig. 10, where the EPNLvariations for the three certification conditions are plotted in the topdiagram and the cumulative EPNL in the bottom diagram.Though a HWB aircraft design with the engines located 2D

upstream of the trailing edge would have to overcome somesignificant technical hurdles to become a viable aircraft product, it isinstructive for acoustics purposes to examine the noise decom-position for this configuration to reveal the effects of shielding and toidentify the dominant noise sources. The PNLT components areshown in Fig. 11 for this configuration, as a function of the observertime for the three noise certification conditions. Clearly, in compar-ison with the baseline case of engine locations 1D upstream of thetrailing edge, shown in Figs. 8 and 9, the engine noise from the openrotors is significantly reduced, as expected by the increased effectiveshielding surface area. The amounts of noise reduction for the threeconditions, however, are not uniform, with more than 10 dB at thepeak PNLT levels for the approach and cutback conditions, but onlyabout half of that amount at the sideline conditions. This is alsoexpected because the sideline angles are blocked by the airframe lessthan the overhead angles.It can be seen from the top and middle diagram of Fig. 11 that the

engine noise reduction at the approach and cutback conditions are sosignificant that, for the former, themain landing-gear noise is now thedominant component, given the ranking of airframe components inFig. 11, and for the latter, the slat noise and the trailing-edge noise

components are nowonly slightly lower than the engine noise. On theother hand, the bottom diagram of this figure shows that the sidelinenoise is still dominantly from the rotors. From these observations, itcan be suggested that the efficiency of engine noise shielding by theHWB airframe is probably approaching its maximum when theengines are located about 2D upstream of the trailing edge; furtherreduction of the total aircraft noisewould have to come from airframenoise reduction and/or engine noise source level reduction.Engine locations can also vary in the direction normal to the

airframe surface, within the range of practical constraints to avoidthe boundary-layer ingestion into the engine on the lower side and theissues of extralong supporting structure and stability control on thehigher side of the range. This effect is shown by the results in Table 7for the variations of aircraft EPNL as a function of the engine heightfrom the HWB surface, respectively at the streamwise position of0.5D and 1D upstream of the trailing edge. The engine height in thesetables is measured in rotor diameter from the engine axis line to theHWB upper surface, and the acoustic results given both in absolutelevels and in the margins to stage 4. In both cases, the variations arewithin a maximum of 1.7 dB for the individual conditions and about1.5 dB for the cumulative levels. It can be noted that the variations innoise levels are not monotonic with the engine installation heightbecause the installation effects include both the shielding and thesource level changes. The latter is due to possible potential fieldinteraction, inflow distortion into the rotor, entrainment of theairframe boundary layer, and/or interaction with the rotor wake andHWB trailing edge because the rotors are very close to the HWB

Engine Position From Trailing Edge

EPN

L(d

B)

-0.5 0 0.5 1 1.5 270

75

80

85

90

95

100

ApproachCutbackSideline

Engine Position From Trailing Edge

CU

ME

PNL

(dB

)

-0.5 0 0.5 1 1.5 2250

260

270

280

290

300

Stage 4 Limit: 293.9

Fig. 10 PNLT variations as a function of engine position in the flowdirection, with the engine position measured in fan diameter and 1Dbeing the baseline position.

Observer Time (s)

PNL

T(d

B)

185 190 195 200 205 210 21560

70

80

90

100 TotalAirframeEngine

Approach

Observer Time (s)

PNL

T(d

B)

15 20 25 30 35 40 4560

70

80

90

100TotalAirframeEngine

Cutback

Observer Time (s)

PNL

T(d

B)

15 20 25 30 35 40 4560

70

80

90

100 TotalAirframeEngine

Sideline

Fig. 11 PNLT (in decibels) for the configuration with the engines at 2Dupstream of the HWB trailing edge.

GUO AND THOMAS 1775

Dow

nloa

ded

by JO

HN

S H

OPK

INS

UN

IVER

SITY

on

Dec

embe

r 28,

201

5 | h

ttp://

arc.

aiaa

.org

| D

OI:

10.2

514/

1.C0

3304

8

surface. Obviously, the effects of shielding and source level changescompete with each other; the former increasingly lowers the far-fieldnoise by the increased shielding region as the rotors move closer tothe airframe, as illustrated in Fig. 5, whereas the latter causes morenoise radiation with decreasing engine height because of theincreased flow effects. Thus, an optimal engine installation heightmay exist at which the aggregate effects of all the features yieldminimum far-field noise. This can be seen from the limited data setsshown in Table 7. For example, the cumulative EPNL margins tostage 4, shown in the last columns of the tables, seem to achievemaximum around the engine centerline height of about 0.83D andaround 0.75Dwhen the rotors are moved further downstream, wherethe boundary-layer growth requires the engines farther away from theHWB surface to avoid the flow interaction.

VII. Effects of Vertical TailsThe use of inboard vertical tails in HWB design is one of the

unresolved issues; it is needed for stability control because of the lackof flaps in theHWBdesign, but this functionmay also be achieved byverticals at the wing tips. Thus, the benefit of vertical tails to theoverall design of aircraft performance is not conclusive. For theacoustic analysis, the analysis starts with the baseline configurationdiscussed earlier, with the engines located one rotor diameterupstream of the trailing edge and three quarters of a rotor diameterabove the HWB surface, and compares the baseline results with fouradditional configurations that involve the use of vertical tails. Thefour configurations consist of two sets of vertical tails respectivelydeployed at two cant angles. The two sets of vertical tails are namedV1 andV3, where the former differs from the latter in its larger heightby approximately 50%. The two installation angles are respectively102 and 120 deg, with the straight vertical position defined as 90 deg.An example is illustrated by the photos in Fig. 12, which is the largermodel deployed at 120 deg.For the configurations with vertical tails, the results of EPNL are

shown in Table 8, together with the results of the baselineconfiguration (no verticals), both for the absolute noise levels and forthe margins to stage 4 limits. Clearly, the vertical tails do not seem to

provide a large benefit, with the V1 configuration providing noisereduction of less than 1 dB for the cumulative EPNL and the V3configuration increasing the noise. This may seem surprising at firstglance, considering the additional shielding surfaces provided by thevertical tails, but can be understood from the process of noiseradiation and shielding from the open rotor sources to the far-fieldmicrophones. For the sideline conditions, the far-field measurementlocations are not in two-dimensional geometry with the microphonesreceiving the noise radiated sideways from the source. The sidelineemission angles are fully three-dimensional. Thus, while the verticaltails may block the 90 deg emission angle that is truly the sidewaydirection for the sideline certification conditions, the shielding effectsin this direction are already significant due to the HWB baselineairframe. This is also the case for the angles in the forward quadrantwhere the shielding due to the large baseline airframe is alreadysignificant and the extra shielding surfaces due to the vertical tails areall enclosed in the shielding zone of the baseline airframe. For theangles in the aft quadrant that experience little shielding from thebaseline airframe, because the vertical tails are located upstream ofthe trailing edge, they do not add any shielding surface for the anglesin the aft quadrant, and thus no significant acoustic benefits areexpected.From the results in Table 8, the noise increments from the baseline

configuration, due to the vertical tails, can be derived, shown inTable 9, for the four cases with vertical tails, at the three certificationconditions as well as for the cumulative levels. In this table, positivenumbers mean noise increase, and negative numbers indicate noisereduction. For the cumulative levels, it is clear that the vertical tailshave, at best, a noise reduction of less than 1 dB, and the largest noiseimpact is actually an increase. The effects of the vertical tails can befurther understood by the results shown in this table for the individualcertification conditions. At sideline conditions, the effects are allnoise reduction, which, though only by a small amount, is consistentfor all four configurations. These effects, however, are largely offsetby the noise increases in the other two conditions, namely at approachand cutback. This is because the approach and cutback conditions areboth for noise in the flyover path. The vertical tails obviously do nothelp the shielding along the flight path because they are located on thesides of the engines. Furthermore, the noise reflected back by thevertical tails from the sideway directionswill propagate away in otherdirections, including the flyover directions, leading to the potential ofincreased noise along the flight path. This is similar to the effects ofmegaphones that enhance the sound in the directions which themegaphones are pointing at, as clearly seen by the results shown inthe table; at approach and cutback conditions, the changes in noiselevels due to the vertical tails are mostly noise increase. This noiseincrease of the megaphone effects is more prominent at the cutbackcondition than the approach condition, probably due to the former’slarger climb angle and larger angle of attack.

VIII. Effects of Surface Liner TreatmentAcoustic liner is a proven noise reduction technology widely used

inside turbofan engines, usually deployed on the engine casingwalls.It is known to be very efficient for tone-dominated noise field. For theopen-rotor engines that are tone-dominated, there is no engine casingfor the treatment, and the amendment for this disadvantage is to usethe airframe surfaces in the vicinity of the engines for the deploymentof the liner treatment. It should be acknowledged up front that this can

Table 7 EPNL (in decibels) for various engine heights

Engine location Engine height from HWB Approach Cutback Sideline Cumulative Margin to stage 40.5D 0.67D 95.6 87.9 92.8 276.3 17.6

0.75D 95.2 87.9 91.8 274.9 19.00.83D 95.4 87.4 92.6 275.4 18.51D 95.3 87.6 93.1 276.0 17.9

1D 0.67D 95.1 86.9 88.3 270.3 23.60.75D (baseline) 94.5 86.2 89.2 269.9 24.0

0.83D 93.4 85.9 89.4 268.7 25.21D 93.5 86.0 89.7 269.1 24.8

Fig. 12 HWB configuration with vertical tails.

1776 GUO AND THOMAS

Dow

nloa

ded

by JO

HN

S H

OPK

INS

UN

IVER

SITY

on

Dec

embe

r 28,

201

5 | h

ttp://

arc.

aiaa

.org

| D

OI:

10.2

514/

1.C0

3304

8

be a radical and aggressive idea that is against the principle ofaerodynamic design; the airframe surfaces are all exposed to externalaerodynamic flows and thus are to be designed with minimum dragfor aerodynamic efficiency. Currently available liner treatments allinduce drag, equivalent to converting smooth surfaces to roughsurfaces, making this concept unfeasible for current applications.Thus, to make liner treatment practical for external airframe surfaces,new technologies will have to be developed for low-drag or draglessliners. In this section, the acoustic effects of surface liners arepresented, both to demonstrate the acoustic potential of this conceptand to quantify the benefits. The latter can hopefully be used to guidethe decision making in developing this technology in the future.For the HWB configurations studied here, the engines are located

near theHWB trailing edge. Thus, the liner treatment can be deployedin the vicinity of the engines on both the main wings and the elevons.This is illustrated in Fig. 13, where the surface areas with gridlikestructures are the locations for the liner treatments. To test the noisereduction efficiencies of various liner designs, three types of linerswere tested; they are straight liner, hook liner, and bulk broadbandliner, all of which were designed by the NASA Langley LinerTechnology team specifically for the LSAF open-rotor experiment.The liners were designed for peak attenuation of frequencies from1BPF to 2BPF. The straight and hook liner designs were differentapproaches at configuring the liner chambers within the limitedthickness available. Three engine locations were tested for theacoustic liner, respectively at 1D, 1.5D, and 2D upstream of theHWB trailing edge, with the first being the baseline configuration.These relatively large distances from the engines to the HWB trailingedge allow sufficient space for the interactions between the soundwaves and the treated surfaces; obviously, if the engines are very

close to the trailing edge, the sound waves can mostly propagatedirectly from the rotors to the far field without interacting with thesurfaces, rendering the liner treatment ineffective. For all the cases,the V1 vertical tails are installed at 120 deg for the liner tests. This is anecessary design choice when considering surface treatmentbecause, without the verticals, the waves impinging on the airframesurface would be reflected upward, becoming irrelevant to the noiseon the ground. In other words, the potential effects of the surface linercan be expected to be confined only to the sound waves that also hitthe verticals.The results for the EPNL for the configurations with the surface

liner treatment are shown in Table 10, for the three noise certificationconditions, as well as the cumulative levels and their margins to thestage 4 limit. For comparison, a reference configuration is alsoincluded in the table for each engine location. These are config-urations without the verticals and the liner treatment. Because thecases with liner and verticals were tested only for rotors located offthe centerline of the HWB airframe, the EPNL calculations for theconfigurations in the table, including the reference configurations,only use the data for off-center engine locations, even though theHWB design has one of its three engines on the centerline, for whichcase the shielding results from the off-center engine were used as anapproximation. This is also why the reference case for the 1D enginelocation is different from the baseline configuration defined inprevious sections; the two have the same aircraft and engine config-uration but use different shielding data for the center engine. Toclearly reveal the effects of surface liners, it would need referenceswith exactly the same design except for the liner. Unfortunately, theLSAF test database does not have such references, except for the caseof 1D engine location, and hence, the liner effects will have to bestudied together with those due to the verticals.From the results shown in Table 10, the combined effects of the

surface liner and the vertical tails can be more clearly revealed bysubtracting the results of the reference cases from the EPNL results,as shown in Table 11, in which negative numbers indicate noisereduction from the reference case, whereas positive numbers are fornoise increase. It can be seen from the table that the overall trendswith various liner types are consistent with only small variationsbetween the liner types for a fixed configuration. This consistencymanifests itself not only in the cumulative values but also in the resultat each certification conditions, which can also be considered as arepeatability test to ensure that the results are reliable, especiallyimportant for the cases discussed here because the effects are sosmall. The variations with the engine location, however, reveal somemore complex trends. For the cumulative effects shown in the lastcolumn in the table, the 1D engine location seems to have the largestnoise reduction, an averagevalue of−0.9 dB. This reduction is cut byone third when the engines move to the 1.5D location and changes tobecome a noise increase of 2.6 dB when the engines move fartherupstream to the 2D location. By examining the individual conditionslisted in the table, it can be seen that this trend is followed by theresults for approach and cutback conditions, with different amountsin the variations. This trend is probably dictated by the megaphonephenomenon discussed earlier with the vertical tails; as the enginesmove farther upstream, the equivalent megaphone length increases,leading to more focused radiation in the flyover plane. The results forthe sideline conditions showvery different trends; the noise reductionincreases initially as the engines move upstream, but the reductionbecomes noise increase at the most upstream location. It can be notedthat both the vertical tails and the surface liner are supposed to workmainly for the sideline conditions. The success of the combination of

Table 8 Effects of vertical tails on aircraft EPNL (in decibels)

Vertical tail configuration Approach Cutback Sideline Cumulative Margin to stage 4V1 at 102 deg 94.2 86.6 88.3 269.1 24.8V3 at 102 deg 94.6 87.2 89.0 270.8 23.1V1 at 120 deg 94.3 86.8 88.3 269.3 24.6V3 at 120 deg 95.5 88.0 88.3 271.7 22.2Baseline 94.5 86.2 89.2 269.9 24.0

Table 9 ΔdB of vertical tail effects relative to the baselineconfiguration

Vertical tail configuration Approach Cutback Sideline CumulativeV1 at 102 deg −0.4 0.3 −0.9 −0.9V3 at 102 deg 0.1 1.0 −0.2 0.8V1 at 120 deg −0.2 0.5 −0.9 −0.6V3 at 120 deg 1.0 1.7 −0.9 1.8

Fig. 13 HWB model with local liner treatments.

GUO AND THOMAS 1777

Dow

nloa

ded

by JO

HN

S H

OPK

INS

UN

IVER

SITY

on

Dec

embe

r 28,

201

5 | h

ttp://

arc.

aiaa

.org

| D

OI:

10.2

514/

1.C0

3304

8

the two, however, depends on the relative positions between theengines, the verticals, and the treated surface areas, which not only setthe shielding angles in the far field but also determine the angles atwhich the sound waves impinge on the liner surfaces. The latter isknown to be an important parameter for the effectiveness of the noiseabsorption by the liner. The complex behavior of the noise reductioneffects for the sideline conditions shown in Table 11 may imply theexistence of an optimal combination of the engine location, thevertical tail deployment, and the surface liner treatment formaximumnoise reduction. Clearly, additional data and prediction methodswould be helpful.

IX. ConclusionsThis paper has presented a detailed assessment of the noise levels

for an hybrid wing–body (HWB) aircraft, comparable in size to theBoeing 787 aircraft. The HWB design is powered by open-rotorengines, including the design characteristics of the baselineconfiguration with considerations of practical feasibility andemerging technologies, the component noise levels, the method-ologies for the acoustic assessment, and the analysis and discussionsof the results. It has been shown that, though the benefits of the HWBnoise shielding lead to a comfortable margin in total aircraft noiselevels to current and potential future aircraft noise regulations, about24 dB below the current regulation of stage 4 for the baselineconfiguration, the design will likely face stiff competition fromconventional aircraft designs, which already have about 16 to 20 dBmargin to stage 4 for aircraft currently in service withcomparable size.By detailed component analysis, it has been shown that the noise

levels of the baseline HWB design are held up by the main landing-gear noise and the inherently high levels of open-rotor noise. Theformer is due to the exposure of the main landing gears to high-speedflows at the junction between the HWB centerbody and its outerwings, which significantly increases the gear noise and represents amajor disadvantage of the HWB design, regardless of the enginetype. For the latter, the open-rotor source noise is significantlyreduced with airframe shielding, but the shielding effects are limitedby the engine locationwithin about one rotor diameter from theHWBtrailing edge, a practical and critical design constraint that ensures theaerodynamic-propulsion integration.

To identify the potential for further noise reduction, variousparametric studies have also been presented, including enginelocations that may enhance the shielding effects, vertical tails thatmay be optimized for sideline noise reduction, and surface linertreatment for possible sound attenuation. It has been shown that someof these concepts have the potential of further noise reduction, anddiscussions have been presented for the practical feasibility of theconcepts and the significant challenges in bringing these conceptsinto reality.

AcknowledgmentsThe authors thank the NASA Environmentally Responsible

Aviation Project (Fay Collier, Project Manager) for funding thisresearch.

References[1] Ffowcs Williams, J. E., and Guo, Y. P., “Sound Generated from the

Interaction of a Steady Flow by a Supersonically Moving Aerofoil,”Journal of Fluid Mechanics, Vol. 195, No. 1, 1988, pp. 113–135.doi:10.1017/S0022112088002344

[2] Guo, Y. P., “ANote on Sound from the Interaction of a Cylindrical Flowby a Semi-Infinite Aerofoil of Subsonic Speed,” Journal of SoundVibration, Vol. 128, No. 2, 1989, pp. 275–286.doi:10.1016/0022-460X(89)90771-2

[3] Guo, Y. P., “Sound Generation by a Supersonic Aerofoil CuttingThrough a Steady Jet Flow,” Journal of Fluid Mechanics, Vol. 216,No. 1, 1990, pp. 193–212.doi:10.1017/S0022112090000398

[4] Khalid, S. A., Wojno, J. P., Breeze-Stringfellow, A., Lurie, D. P., Wood,T. H., Ramakrishnan, K., and Paliath, U., “Open Rotor Designs for LowNoise and High Efficiency,” Proceedings of the ASME Turbo Expo2013, American Soc. of Mechanical Engineers Paper GT2013-94736,New York, 2013.

[5] Clark, L.R., andGerhold,C.H., “InletNoiseReduction byShielding forthe Blended Wing Body Airplane,” 5th AIAA/CEAS AeroacousticsConference and Exhibit, AIAA Paper 1999-1937, June 1999.

[6] Hill, G. A., Brown, S. A., Geiselhart, K. A., and Burg, C., M.,“Integration of Propulsion Airframe Aeroacoustic Technologies andDesign Concepts for a Quiet BlendedWing Body Transport,” AIAA 4thAviation Technology, Integration and Operations (ATIO) Forum,AIAA Paper 2004-6306, Sept. 2004.

[7] Reimann,C.A., Tinetti, A. F., andDunn,M.H., “NoiseScattering by theBlended Wing Body Airplane: Measurements and Prediction,” 12th

Table 10 EPNL (in decibels) for configurations with acoustic liner on HWB surface

Engine location Liner configuration Approach Cutback Sideline Cumulative Margin to stage 41D Straight liner 94.1 85.2 88.6 267.8 26.1

Hook liner 94.0 85.3 88.6 267.9 26.1Bulk liner 94.0 85.4 88.1 267.5 26.4Reference 94.4 85.5 88.7 268.6 25.3

1.5D Straight liner 93.5 82.3 86.1 261.9 32.0Hook liner 93.5 82.3 86.2 261.9 32.0Bulk liner 93.4 82.0 86.4 261.8 32.1Reference 93.4 81.9 87.2 262.5 31.4

2D Straight liner 93.2 80.5 86.9 260.7 33.2Hook liner 93.2 80.5 87.3 261.1 32.9Bulk liner 93.3 80.9 87.0 261.2 32.7Reference 93.2 79.3 85.9 258.3 35.6

Table 11 Effects of acoustic liner andverticals inΔdB from the reference configuration

Engine location Liner configuration Approach Cutback Sideline Cumulative1D Straight liner −0.3 −0.3 −0.1 −0.8

Hook liner −0.4 −0.2 −0.1 −0.8Bulk liner −0.4 −0.1 −0.6 −1.1

1.5D Straight liner 0.0 0.4 −1.1 −0.6Hook liner 0.0 0.4 −1.0 −0.5Bulk liner −0.1 0.2 −0.7 −0.7

2D Straight liner 0.1 1.3 1.0 2.3Hook liner 0.1 1.3 1.4 2.7Bulk liner 0.1 1.6 1.1 2.8

1778 GUO AND THOMAS

Dow

nloa

ded

by JO

HN

S H

OPK

INS

UN

IVER

SITY

on

Dec

embe

r 28,

201

5 | h

ttp://

arc.

aiaa

.org

| D

OI:

10.2

514/

1.C0

3304

8

AIAA/CEAS Aeroacoustics Conference, AIAA Paper 2006-2474,May 2006.

[8] Thomas, R. H., Burley, C. L., and Olson, E. D., “Hybrid Wing BodyAircraft System Noise Assessment with Propulsion AirframeAeroacoustic Experiments,” International Journal of Aeroacoustics,Vol. 11, Nos. 3–4, 2012, pp. 369–410.doi:10.1260/1475-472X.11.3-4.369

[9] Czech, M. J., Thomas, R. H., and Elkoby, R., “Propulsion AirframeAeroacoustic Integration Effects of a Hybrid Wing Body AircraftConfiguration,” International Journal of Aeroacoustics, Vol. 11,Nos. 3–4, 2012, pp. 335–368.doi:10.1260/1475-472X.11.3-4.335

[10] Bonet, J. T., Schellenger, H. G., Rawdon, B. K., Elmer, K. R.,Wakayama, S. R., Brown, D., and Guo, Y. P., “EnvironmentallyResponsible Aviation (ERA) Project—N+2 Advanced VehicleConcepts Study and Conceptual Design of Subscale Test Vehicle(STV),” NASA CR-NND11AG03C, Dec. 2011.

[11] Lighthill, M. J., “On Sound Generated Aerodynamically 1. GeneralTheory,”Proceedings of theRoyal Society of London, Series A, Vol. 211,No. 1107, pp. 564–587, 1952.doi:10.1098/rspa.1952.0060

[12] Thomas, R. H., Burley, C. L., Lopes, L. V., Bahr, C. J., Gern, F. H., andVan Zante, D. E., “System Noise Assessment and the Potential for LowNoise Hybrid Wing Body Aircraft with Open Rotor Propulsion,” 52ndAerospace Sciences Meeting, AIAA Paper 2014-0258, 2014.

[13] Imamura, T., Ura, H., Yokokawa, Y., Enomoto, S., Yamamoto, K., andHirai, T., “Designing of Slat CoveFiller as aNoiseReductionDevice forLeading-Edge Slat,” 13th AIAA/CEAS Aeroacoustics Conference,AIAA Paper 2007-3473, 2007.

[14] Guo, Y. P., “A Discrete Vortex Model for Slat Noise Prediction,” 7thAIAA/CEASAeroacoustics Conference andExhibit, AIAAPaper 2001-2157, May 2001.

[15] Guo, Y. P., and Joshi,M. C., “Noise Characteristics of Aircraft High LiftSystems,” AIAA Journal, Vol. 41, No. 7, 2003, pp. 1247–1256.doi:10.2514/2.2093

[16] Guo, Y. P., “OnNoise Reduction by Flap Side Edge Fences,” Journal ofSound and Vibration, Vol. 277, No. 2, 2004, pp. 369–390.doi:10.1016/j.jsv.2003.09.029

[17] Hutcheson, F. V., Brooks, T. F., and Humphreys, W. M., “NoiseRadiation from aContinuousMold-LineLinkFlapConfiguration,” 14thAIAA/CEAS Aeroacoustics Conference, AIAA Paper 2008-2966,May 2008.

[18] Ravetta, P. A., Burdisso, R. A., Ng,W. F., Khorrami, M. R., and Stoker,R. W., “Screening of Potential Noise Control Devices at Virginia Techfor QTD II Flight Test,” 13th AIAA/CEAS Aeroacoustics Conference,AIAA Paper 2007-3455, 2007.

[19] Dobrzynski, W., Chow, L. C., Smith, M., Boillot, A., Dereure, O., andMolin, N., “Experimental Assessment of Low Noise Landing Gear

Component Design,” 15th AIAA/CEAS Aeroacoustics Conference,AIAA Paper 2009-3276, May 2009.

[20] Dobrzynski, W., Chow, L. C., Guion, P., and Shiells, D., “Research intoLanding Gear Airframe Noise Reduction,” 8th AIAA/CEAS Aeroacous-tics Conference & Exhibit, AIAA Paper 2002-2409, June 2002.

[21] Pott-Pollenske, M., Almoneit, D., and Wild, J., “On the NoiseGeneration of Krueger Leading Edge Devices” 21st AIAA/CEASAeroacoustics Conference, AIAA Paper 2015-3142, 2015.

[22] Mendoza, J. M., Brooks, T. F., and Humphreys, W. M., “AeroacousticMeasurements of a Wing/Slat Model,” 8th AIAA/CEAS AeroacousticsConference & Exhibit, AIAA Paper 2002-2604, June 2002.

[23] Pott-Pollenske, M., Alvarez-Gonzalez, J., and Dobrzynski, W., “Effectof Slat Gap on Farfield Radiated Noise and Correlation with Local FlowCharacteristics,” 9th AIAA/CEAS Aeroacoustics Conference andExhibit, AIAA Paper 2003-3228, May 2003.

[24] Herr, M., Pott-Pollenske, M., Ewert, R., Dirk, B., Siebert, J., Delfs, J.,Rudenko,A., Büscher, A., and Irene, H. F., “Large-Scale Studies on SlatNoise Reduction,” 21st AIAA/CEAS Aeroacoustics Conference, AIAAPaper 2015-3140, 2015.

[25] Guo, Y. P., “A Statistical Model for Landing Gear Noise Prediction,”Journal of Sound and Vibration, Vol. 282, No. 2, 2004, pp. 61–87.doi:10.1016/j.jsv.2004.02.021

[26] Guo, Y. P., “A Component-Based Model for Aircraft Landing GearNoise Prediction,” Journal of Sound Vibration, Vol. 312, No. 5, 2008,pp. 801–820.doi:10.1016/j.jsv.2007.11.013

[27] Guo, Y. P., “On Trailing Edge Noise Modeling and Prediction forAircraft High Lift Wings,” NASA CR-NNL07AA03A, 2010.

[28] Guo, Y. P., “Slat Noise Modeling and Prediction,” Journal of SoundVibration, Vol. 331, No. 15, 2012, pp. 3567–3586.doi:10.1016/j.jsv.2012.03.016

[29] Guo, Y. P., Brusniak, L., Czech, M., and Thomas, R. H., “Hybrid WingBody (HWB) Slat Noise Analysis,” 51st AIAA Aerospace SciencesMeeting, AIAA Paper 2013-0462, Jan. 2013.

[30] Guo, Y. P., “A Study on Local Flow Variations for Landing Gear NoiseResearch,” 14th AIAA/CEAS Aeroacoustics Conference, AIAA Paper2008-2915, May 2008.

[31] Guo, Y. P., “Effects of Local Flow Variations on Landing Gear NoisePrediction and Analysis,” Journal of Aircraft, Vol. 47, No. 2, 2010,pp. 383–391.doi:10.2514/1.43615

[32] Czech, M., and Thomas, R. H., “Open Rotor Aeroacoustic InstallationEffects for Conventional and Unconventional Airframes,” 19th AIAA/CEAS Aeroacoustics Conference, AIAA Paper 2013-2185, 2013.

[33] Guo, Y. P., and Thomas, R. H., “An Experimental Study on Open RotorNoise Shielding by Hybrid-Wing–Body Aircraft,” AIAA Journal, (to bepublished).doi:10.2514/1.J054243

GUO AND THOMAS 1779

Dow

nloa

ded

by JO

HN

S H

OPK

INS

UN

IVER

SITY

on

Dec

embe

r 28,

201

5 | h

ttp://

arc.

aiaa

.org

| D

OI:

10.2

514/

1.C0

3304

8