Rotor blade–vortex interaction noise

JPAS=76=Elizabeth Mathew=Venkatachala=BG

Progress in Aerospace Sciences 36 (2000) 97}115

Rotor blade}vortex interaction noise Yung H. Yu* Aeroyightdynamics Directorate (AFDD), NASA Ames Research Center, Mowett Field, CA, 94035, USA

Abstract Blade}vortex interaction noise-generated by helicopter main rotor blades is one of the most severe noise problems and is very important both in military applications and community acceptance of rotorcraft. Research over the decades has substantially improved physical understanding of noise-generating mechanisms, and various design concepts have been investigated to control noise radiation using advanced blade planform shapes and active blade control techniques. The important parameters to control rotor blade}vortex interaction noise and vibration have been identi"ed: blade tip vortex structures and its trajectory, blade aeroelastic deformation, and airloads. Several blade tip design concepts have been investigated for di!using tip vortices and also for reducing noise. However, these tip shapes have not been able to substantially reduce blade}vortex interaction noise without degradation of rotor performance. Meanwhile, blade root control techniques, such as higher-harmonic pitch control (HHC) and individual blade control (IBC) concepts, have been extensively investigated for noise and vibration reduction. The HHC technique has proved the substantial blade}vortex interaction noise reduction, up to 6 dB, while vibration and low-frequency noise have been increased. Tests with IBC techniques have shown the simultaneous reduction of rotor noise and vibratory loads with 2/rev pitch control inputs. Recently, active blade control concepts with smart structures have been investigated with the emphasis on active blade twist and trailing edge #ap. Smart structures technologies are very promising, but further advancements are needed to meet all the requirements of rotorcraft applications in frequency, force, and displacement.  2000 Published by Elsevier Science Ltd. All rights reserved.

Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Blade}vortex interaction (BVI) noise characteristics . . . . 3. Aerodynamics and dynamics of blade}vortex interactions 4. Analytical prediction capability . . . . . . . . . . . . . . . . 5. Noise reduction concepts . . . . . . . . . . . . . . . . . . . 6. Active blade control concepts . . . . . . . . . . . . . . . . . 7. Concluding remarks . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction Helicopter impulsive noise is generated from two distinct aerodynamic events. One is from compressible #ow

* Tel.: 001-650-604-5834. E-mail address: [email protected] (Y.H. Yu).

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

97 99 101 103 106 109 112 113

"eld due to high tip Mach numbers on the rotor's advancing side, called high-speed impulsive noise, and the second type is from unsteady pressure #uctuations on a rotor blade due to interactions with vortices generated by previous blades, called blade}vortex interaction noise. These types of noise are loud and impulsive in nature and have signi"cant e!ects on both military detection and community annoyance.

0376-0421/00/$ - see front matter  2000 Published by Elsevier Science Ltd. All rights reserved. PII: S 0 3 7 6 - 0 4 2 1 ( 9 9 ) 0 0 0 1 2 - 3

Jpas=76=EM=VVC=BG 98

Y.H. Yu / Progress in Aerospace Sciences 36 (2000) 97}115

Nomenclature a  c c , dB dS d< M P p GH p Pa r r R R/D SP SPL ¹ GH v x z d GH o t k

speed of sound in ambient medium at a standard condition blade chord blade sectional normal force coe$cient decibel blade surface element source volume element source convection speed divided by the speed of sound in the direction to an observer pd GH air pressure Pascal distance between source and observer (in acoustics equations) radial position on a blade (in rotor blade coordinates) rotor blade radius rate of descent sound pressure in Pascal sound pressure level in dB Lighthill stress tensor (ov v #p !a od ) G H GH  GH forward velocity observer position vertical coordinate Dirac delta function air density blade azimuth angle (deg.) advance ratio

Understanding of noise-generating mechanisms has been substantially advanced in recent years with the help of rapid advancement of computational methods and comprehensive experiments. The generating mechanisms of high-speed impulsive rotor noise have been reasonably well understood. Basically, the blade tip volume and shape are the major parameters to noise generation. Particularly, blade airfoil and tip shapes are the important design parameters. The e!ect of blade tip and airfoil shapes on noise radiation can be explained with a delocalization phenomenon. At and above the certain Mach number for a given blade tip and airfoil shape, noise pulse shapes are suddenly changed from smooth triangular shapes to sharp saw-tooth shapes. At this Mach number, local shock waves are no longer contained in the near "eld and suddenly propagate to the acoustic far "eld. This Mach number is called the delocalization Mach number [1,2]. The sudden connection, and also propagation, of the local shock waves to the acoustic far "eld has been explained in terms of the characteristics of hyperbolic and elliptic equations in a blade-"xed rotating system. Many blade design concepts have been investigated in terms of the delocalization Mach number, resulting in new advanced blade tip and airfoil shapes. These advanced blade tip shapes have, in general, smaller blade tip volumes (thin airfoils and tapered tips) and sweeps. In terms of directivity, high-speed impulsive noise propa-

gates directly forward in the rotor plane. Due to this directivity pattern, pilots are in most cases unaware of the noise radiation, while its presence may be recognized in a forward region. Blade}vortex interaction (BVI) noise is generated mainly from unsteady pressure #uctuations on a blade due to interactions with previously generated tip vortices during descent or maneuvering #ight. There are two salient features of the rotor BVI noise characteristics. First, unsteady blade surface pressure #uctuations responsible for intense noise are concentrated mainly near the leading edge of a blade. Second, the noise has a strong radiation directivity pattern, mostly forward and under the rotor plane. Due to this directivity pattern, helicopter pilots are keenly aware of this noise propagation and also that BVI noise causes considerable community annoyance. Rotor wakes have been recognized as the most important parameter to generate BVI noise. Several blade tip shapes were investigated to di!use blade tip vortex structures, but without much success in reducing blade}vortex interaction noise. Recently, higher-harmonic pitch control (HHC) and individual blade control (IBC) techniques have been investigated to reduce BVI noise and vibration. These active blade root control techniques have shown substantial noise reduction, but the physical mechanisms for noise and vibration reduction are not fully understood. From experimental and

Jpas=76=EM=VVC=BG Y.H. Yu / Progress in Aerospace Sciences 36 (2000) 97}115

99

analytical results, a blade}vortex miss distance has been considered as an important parameter to BVI noise generation and reduction. The miss distance is strongly in#uenced by tip vortex trajectory, blade elastic deformation, and induced velocity distribution. In this paper, current understanding of blade}vortex interaction noise generating mechanisms, analytical prediction capabilities, and noise reduction concepts will be examined. There are several review articles on helicopter noise in general [3}5]. The readers are referred to these review articles for broader aspects of noise.

2. Blade}vortex interaction (BVI) noise characteristics Rotor blade}vortex interaction noise is generated when a rotor blade passes close to trailing tip vortices previously generated. This type of noise is generated during descending or maneuvering #ight. The noise signatures are impulsive and have large amplitudes. Fig. 1 shows typical interaction patterns on a rotor disk in a descending #ight of a rotorcraft, in which several interactions on both the advancing and retreating sides occurred [6]. During these interactions, a blade experiences impulsive pressure #uctuations during a very short period of time. However, these interactions do not equally contribute to far-"eld noise. Pressure #uctuations on the retreating side are, for example, more intense than those on the advancing side, but contribute much less to far-"eld noise. And among several interactions on the advancing side, only parallel or near-parallel interactions turned out to be major contributors to noise. Typical time signatures and frequency spectra of blade}vortex interaction noise in a far "eld are shown in Fig. 2, in which distinctive pulse shapes are shown. The noise signatures show several interactions and each consists of positive and negative amplitudes, with positive amplitudes dominant on the advancing side and negative amplitudes on the retreating side. This sign change of noise signatures is due to the fact that rotational directions of vortices on the advancing and retreating sides are opposite during blade}vortex interactions. The noise pulses are rich in high harmonics of the rotor fundamental frequency and contain many distinctive harmonics in a high-frequency range. Recently, several wind tunnel tests with model-scale rotors were carried out to measure noise footprints in a horizontal plane under the rotor for examining residential noise exposures. Extensive acoustic measurements of noise footprints with a model-scale BO-105 rotor system show that high noise levels are concentrated on a pattern extended slightly forward and towards the advancing side of the rotor disk, with another lobe on the retreating side, as shown in Fig. 3 [7]. This noise directivity pattern was also con"rmed with a #ight test with the same rotor system.

Fig. 1. Blade}vortex interactions during descending #ight (number 1}7 indicates blade}vortex interactions) [6].

A simpler approach has been investigated on noise directivity patterns using blade}vortex interaction geometry. Since a blade}vortex interaction occurs at an angle between a blade and a vortex "lament, the velocity of the blade}vortex interaction point along a blade span can easily become supersonic when the interaction angle is relatively small. The velocity of the blade}vortex interaction point along a blade span is called a trace Mach number [8,9] and the interactions with a supersonic trace Mach number can be acoustically very e$cient in a certain direction [10]. It was suggested that the noise along the direction of the peaks would decay as an inverse "rst power rather than as an inverse square, indicating that very strong noise can be propagated to a particular direction in the far "eld [10]. This idea has been expanded to give useful physical explanations of sound radiation strengths and directivities [10,11]. Recently, simple wave tracing methods are used to understand the e$ciency and directivity of the noise radiation process. Due to curved blade}vortex interaction trajectories in space, sound waves can be easily focused to a smaller region and then defocused in the far "eld. From this

Jpas=76=EM=VVC=BG 100

Y.H. Yu / Progress in Aerospace Sciences 36 (2000) 97}115

Fig. 2. Typical BVI noise characteristics [7].

Fig. 3. Mid-frequency BVI noise directivity pattern [7].

analysis, supersonic decelerating BVI interactions radiate most of its acoustic energy to the front of the rotor near the rotor plane, while parallel interactions are more signi"cant out of the rotor plane [12]. It is common practice to obtain noise data with a model-scale rotor system in an acoustic wind tunnel. In wind tunnel acoustic tests in order to simulate full-scale #ight tests, two important issues have to be carefully addressed: scalability and rotor trim. The scalability issue was investigated using a two-bladed rotor system in wind tunnel and #ight tests [13]. It was reported that the comparison between model- and full-scale acoustic data was quite good for the low advance ratio of 0.167, but the comparison begins to deteriorate for the advance ratio of 0.224 or higher. This investigation shows that in general it is possible to duplicate full-scale BVI noise signatures in model-scale tests, keeping the following four nondimensional parameters constant: hover tip Mach number, advance ratio, thrust coe$cient, and tip-path-plane angle, besides geometric and structural dynamic similarity. However, the pulse widths of full-scale data are in general smaller than those of the model-scale data, potentially due to the fact that the trailing vortex core sizes of a full-scale rotor are smaller than the corresponding model-scale vortex cores. The trim issue is more complicated. In general, rotor is trimmed in a slightly di!erent manner between wind tunnel tests and #ight tests. Flight

Jpas=76=EM=VVC=BG Y.H. Yu / Progress in Aerospace Sciences 36 (2000) 97}115

101

tests trim a rotorcraft with thrust, hub moments and X-force, but most of wind tunnel tests trim a rotor with rotor thrust coe$cient (C /p) and zero #apping. 2 Naturally, di!erent trims produce di!erent rotor tippath-plane angles, which play an important role in determining blade airloads, rotor induced velocity "eld, blade}vortex miss distances, and rotor noise. This trim issue should be carefully investigated further. In addition, other important factors are the e!ects of rotorcraft fuselage and wind tunnel walls. These e!ects on rotor in#ow have not been fully re#ected in wind tunnel tests and also were not properly modeled in prediction analyses.

Fig. 5. Blade surface pressure distribution at 3% of a chord [7].

3. Aerodynamics and dynamics of blade}vortex interactions For better understanding of blade}vortex interaction noise generation, it is important to understand blade aerodynamics, wakes, and structural dynamics during descent #ight. A typical example of blade surface pressure #uctuations in a descent #ight is shown in Fig. 4 at several chordwise stations [14]. The leading edge pressure #uctuations identify blade}vortex interaction phenomena quite clearly. Strong pressure #uctuations from 45 to 903 in azimuth are seen corresponding to BVI locations on the advancing side of a rotor disk. Relatively strong pressure #uctuations are also seen on the retreating side (around 2903 in azimuth). Analytical studies show that the interaction at about 553 is with a 1.5-

Fig. 4. Blade surface pressures at di!erent chord positions from a full-scale OLS #ight test [14].

revolution-old tip vortex, whereas the second interaction at about 703 is with a 1-revolution-old tip vortex. On the retreating side, the interaction with a 1-revolution-old tip vortex is observed at around 2903, which starts from inboard of the span to outboard. One important "nding is that BVI phenomena are concentrated near the leading edge of a blade. In fact, the dominant BVI pressure #uctuations are con"ned to the "rst 10% of the blade chord. The same pattern also exists for lower-surface pressure transducers and can also be found in modelscale tests with BO-105 and UH-60 rotor systems at the DNW [15,7], indicating that the rotor blade}vortex interaction phenomenon is a leading-edge phenomenon. Contours of blade surface pressures at 3% of a chord are shown in Fig. 5 [7]. The strong pressure #uctuations are seen in the "rst and fourth quadrants of a rotor disk, which show several blade}vortex interactions on both the advancing and retreating sides. Near-parallel blade}vortex interactions, which generate strong BVI noise, appear on the advancing side between 40 and 603 in azimuth. A number of interactions appear also in the fourth quadrant between 305 and 3153. With these nearparallel interactions on the advancing and retreating sides, a typical blade}vortex interaction noise footprint shows two distinct radiation lobes, as shown in Fig. 3. The geometry of vortex "laments of two tip vortices (most signi"cant for BVI noise) on the advancing and retreating sides is shown in Fig. 6. This was obtained from measuring the vortex core centers in space of at least four discrete sections along the vortex of interest, using a laser}light sheet technique [16]. These data were measured for the blade azimuth positions of 35 and 2953. The actual shapes of the deformed blades are plotted in the side views, so that the vertical distances between the blade and the vortices provide the blade}vortex miss distances. This miss distance depends on three factors: blade airload distribution at the time of vortex

Jpas=76=EM=VVC=BG 102

Y.H. Yu / Progress in Aerospace Sciences 36 (2000) 97}115

Fig. 6. Tip vortex geometry segments in top and side views [7].

formation, induced downwash velocity "eld, and blade tip #apping de#ections at both the vortex formation and the blade}vortex interaction. Since blade tip de#ection plays an important role in determining the blade}vortex miss distance, further study on blade structural dynamics is needed. A tip vortex velocity "eld was also measured by a laser doppler velocimetry (LDV) system [7] and tight vortices appear on the retreating side, with bigger vortices on the advancing side. It was stipulated that the advancing side vortex be generated by a rotor blade near 1353 in azimuth and be approximately 4603 of rotation old when it produces a signi"cant blade}vortex interaction at near 553 in azimuth. For the retreating side, the vortex is generated near 2353 in azimuth and its age becomes about 4403 for a signi"cant interaction at near 3003. This suggests that the advancing side vortex may have been signi"cantly a!ected by the multiple (5 or 6) interactions during traveling, while the retreating side vortex had only two or three interactions. From these measurements, the vortex core radius was estimated to be

about 38% of the chord length on the advancing side and about 41% on the retreating side. The circulation value was estimated to be about 1.5 m/s on the advancing side and 2.6 m/s on the retreating side [7]. In order to understand BVI phenomena and acoustics characteristics in a simple environment, two types of methods were investigated: one for interactions of a rotor blade with an independent vortex generated upstream with a vortex generator [17,18] and the other for twodimensional vortex}airfoil interactions [19,20]. A BVI rotor test with an upstream vortex generator produced interesting observations on normal force, pitching moment and tangential force during interaction [17]. The leading edge suction peak grows just before the vortex passes the leading edge and then collapses toward the trailing edge. The normal force rises to a maximum and then reverses as the leading edge suction peak grows and collapses. The pitching moment achieves its peak near the leading edge and then a pitch-down moment is produced. The tangential force is initially dominated by the large forward force and then reverses its direction [17].

Jpas=76=EM=VVC=BG Y.H. Yu / Progress in Aerospace Sciences 36 (2000) 97}115

103

For further investigation on parallel blade}vortex interactions, a test was performed in the 80;120-ft wind tunnel [18]. The test results show that when a vortex impacts the leading edge of a blade, a wave emanates from the leading edge, travels downstream, and is re#ected back upstream from the trailing edge.

4. Analytical prediction capability The Ffowcs Williams and Hawkings formulation, Eq. (1), has been successfully used for predicting rotor noise. For low subsonic cases, the linear terms of monopoles and dipoles (the last two terms) are successfully used for noise prediction with an assumption that acoustic sources are con"ned in a small region. For high-speed cases, the nonlinear term of quadrupoles (the "rst term) is used along with the linear terms for noise prediction. But numerical calculations of this nonlinear volume term require extreme care due to the fact that rotor aerodynamics and acoustics are intrinsically coupled and rotor acoustic sources are not con"ned in a small region. The Ffowcs Williams and Hawkings formulation is expressed as follows [21]:

 

 

* ¹ GH 4nao(x, t)" d< M *x *x r"1!M " G H P * P n * o v GH H M L ! dS# dS. (1) *x r"1!M " r"1!M " *t G P P The terms inside a bracket are evaluated at a retarded time. The dipole term (the second term) can be expressed in a computationally convenient form developed by Farassat as shown in Eq. (2) [22].



 





1 l r( G G 4pP (x, t)" dS * r(1!M ) a M D P l !l M P G G dS # r(1!M ) D P 1 l (rM r( #a M !a M) P G G M P M # dS. (2) a r(1!M ) M D P For normal descent #ight, blade}vortex interaction noise becomes dominant and can be predicted with Eq. (2) with a known surface pressure distribution over a blade. In earlier days, the transient lift distribution on a blade was obtained by using a two-dimensional linear unsteady aerodynamic theory for an oblique sinusoidal gust model [9]. Furthermore, the concept of convection speed (trace Mach number) in a three-dimensional rotating blade was introduced to represent the movement of the lift distribution (as acoustic sources) at interactions. Recently, prediction capability has been substantially advanced with surface pressure #uctuations by computa-

   





Fig. 7. Measured and predicted acoustic time histories for k"0.152 [23].

tional #uid dynamics (CFD) solvers. For example, acoustic signatures of various rotor blades were predicted using the surface pressures obtained from CFD solvers [23}28]. A typical comparison of measured and predicted acoustic time histories is shown in Fig. 7. In this "gure, the measured acoustic time history shows several blade}vortex interactions that consist of one strong impulse followed by several moderate pulses. The predicted acoustic time history shows the strong pulse well, but the moderate impulses far less well. From many validation e!orts over the years, it can be concluded that the FWH formulation, Eq. (2), predicts reasonably accurate acoustic signatures, as long as accurate airloads are provided experimentally or computationally. This implies that blade airloads are the most important element for predicting blade}vortex interaction noise. Several rotor tests were recently performed with model-scale rotor systems in the DNW to obtain comprehensive test data of blade airloads, wakes, blade deformation, and acoustics. For example, a joint international team from the US, Germany, France, and the Netherlands carried out a rotor test with a BO-105 model rotor, called the higher-harmonic control aeroacoustic rotor test (HART) program [7,29]. A test with a rotor system was performed with several blade tip shapes by researchers in Japan [30]. NASA researchers carried out a test with a model-scale tiltrotor called a Tilt Rotor Aeroacoustic Model (TRAM) program [31]. These new data are extremely important to understand basic physical phenomena of blade}vortex interactions and also to validate (and enhance) prediction capabilities. Predicted airload distributions are compared with the HART test data for three spanwise locations in Fig. 8 [32]. The measured sectional blade airloads (c M) were , obtained by integrating blade pressures along a chord. The normal coe$cient c is de"ned as aerodynamic force , per unit area, normal to a blade chord, normalized with chord and dynamic pressure. The test data show multiple blade}vortex interactions in the "rst and fourth

Jpas=76=EM=VVC=BG 104

Y.H. Yu / Progress in Aerospace Sciences 36 (2000) 97}115

Fig. 8. Comparison of airloads at three spanwise locations [32].

quadrants of the rotor disk. The data also show the large amplitude 2/rev harmonics, while the predicted results do not. Several structural and aerodynamic parameters, such as elastic torsion, aerodynamic center, and center of gravity, were carefully examined to investigate the 2/rev discrepancy [32]. No de"nite physical explanation of the 2/rev harmonics is given yet, but the study suggests that the positions of an aerodynamic center and a center of gravity play an important role. Signi"cant underprediction in the analytical results is observed on the advancing side, while the correlation appears fairly good on the

retreating side. In summary, rotor wakes and blade structural dynamics are the two most important parameters in predicting blade airloads. Now, analytical results of rotor wakes and blade deformation will be discussed with experimental data. First, vortex trajectories were measured with a laser-light-sheet visualization technique during the HART program [7] and are compared with analytical results with a free wake model as shown in Fig. 9. The predicted vortex segments in the top view at 353 in azimuth are consistently located upstream by roughly 2-chord length (0.24 m), relative to the test data. The predicted miss distances at this azimuth angle are larger than the measured data by about onechord length (0.12 m). At the azimuth angle of 2953, the predictions in the top view are signi"cantly improved and the miss distances are reasonably predicted. Good correlation of wakes and miss distances in the retreating side may explain excellent airload predictions in the retreating side. Due to lack of experimental data and physical understandings of vortex core sizes and vortex structures, a wide range of vortex core sizes is being used in current prediction modeling activities. Basically, vortex core sizes have been arbitrarily adjusted to match experimental data. The experimental data shows that the core radius of the 1.5-revolution-old vortex on the advancing side is about 38% of the chord and the core radius of the 1.25-revolution-old vortex on the retreating side is about 41% of the chord, as mentioned above. But most of the prediction analyses uses the core sizes from about 5 to 200% of the chord. The e!ect of core sizes on blade airloads was investigated computationally along with limited experimental data [33]. This computational study found that smaller core sizes (4.6% of the chord) generate a phase shift and extraneous pulses, and bigger core sizes (20% of the chord) produce a better correlation with test data. In summary, modeling of rotor wakes is not mature. Due to lack of comprehensive database on wakes up to 2-revolutions old in blade}vortex interactions, progress of modeling of rotor wakes has been slow over the years. Comprehensive test data on wake formation and development process, particularly vortex core sizes/strength and vortex trajectory, are needed. However, one interesting phenomenon observed from the experiment with higher-harmonic controls is the appearance of double vortices at the blade tip for the low vibration case as shown in Fig. 10. Separate vortex rollups (clockwise inboard and counterclockwise at the tip) are formed due to a negative lift distribution at the blade tip, particularly in the second quadrant of the rotor disk [7]. Blade aeroelastic de#ection is another area of importance [34]. Predicted de#ections of blade lead-lag, #ap, and elastic torsion at the blade tip are compared with the HART data in Fig. 11 [32]. The waveforms of leadlag de#ections are well predicted, but the mean value is

Jpas=76=EM=VVC=BG Y.H. Yu / Progress in Aerospace Sciences 36 (2000) 97}115

105

Fig. 9. Comparison of measured and predicted vortex segments at t"35 and 2953 [32].

shifted ahead by roughly a quarter-chord length. The peak-to-peak amplitudes of #ap de#ections are underpredicted in general, but the phase is reasonably correlated. The elastic torsion de#ections are poorly predicted in amplitude and phase, and sometimes the sign of amplitudes is reversed. There are two items to consider in blade aeroelastic de#ections: blade elastic torsion and #ap. As mentioned earlier, measured blade airload distributions show a clear 2/rev variation, while the predicted results do not. This discrepancy may be attributed to the poor correlation of blade elastic torsion deformation as shown earlier. One obvious problem is poor capabilities of current CFD codes on airfoil pitching moment predictions. The other important information is the history of #ap de#ections along azimuth angles, particularly at two speci"c azimuth angles of the vortex generation and of the

blade}vortex interaction. The #ap de#ections of the blade tip at these two azimuth angles along with induced downwash velocity distributions will play an important role in determining blade}vortex miss distances shown in Fig. 9. Another reason for poor correlation of blade aeroelastic deformation is lack of good comprehensive aeromechanics analytical capability, in which CFD and Computational Structural Dynamics (CSD) codes are close-coupled. Most of rotor aerodynamics}structural dynamics coupling e!orts have used a loose-coupling scheme to exchange the information, mainly lift only, between aerodynamic and structural dynamic analyses in one revolution of a blade at a time. This loose-coupling scheme may be suitable for performance calculations, but not be suitable for blade}vortex interactions, due to the fact that the BVI phenomenon is a very short time event.

Jpas=76=EM=VVC=BG 106

Y.H. Yu / Progress in Aerospace Sciences 36 (2000) 97}115

Fig. 10. Tip vortex geometry segments for baseline (a), low noise (b), and low vibration (c) cases [7].

Currently, a tight coupling scheme is being developed with exchanging the information of blade lift and pitching moments between aerodynamic and structural dynamic analyses, and with a "ne resolution in azimuth angles [35].

5. Noise reduction concepts Rotor blades are the main source of noise, lift force, propulsive force, and controls. Any quiet blade concept should be considered with the e!ect of rotorcraft performance and controls. For example, a simple way to reduce rotor noise is to reduce a blade tip speed, but the penalty on performance and structural weight may outweigh the noise bene"t. The low rotor noise issue has attracted much attention over the years, and several focused programs have been developed with useful technical results. In the US, a NASA-AHS National Rotorcraft Noise Reduction program was developed as a joint project of helicopter manufacturers and NASA and has produced

useful results [36]. In Europe, a cooperative research program, HELISHAPE, was initiated with 16 European countries to improve aerodynamic and aeroacoustic prediction capability for advanced blades and to investigate the e!ect of planform shapes on noise [37]. Another cooperative HELINOISE program, among European helicopter manufacturers, universities, and research institutes, was developed and interesting results have been reported [38]. Recently in Germany, an advanced technology rotor (ATR) program was developed for performance, noise, and operational cost [39]. The rotor parameters such as blade planform, twist distribution, blade tip shapes and blade tip speed have been carefully investigated through analysis, wind tunnel tests, and #ight tests. Another program, called rotor active control technology (RACT), has been developed in Germany to fully explore the potential of the IBC technology with #ight tests and also the full-scale piezoelectric #ap control technology with wind tunnel tests [40]. In France, a research program was initiated to develop a demonstrator helicopter whose overall noise level is 8 EPNdB below the ICAO

Jpas=76=EM=VVC=BG Y.H. Yu / Progress in Aerospace Sciences 36 (2000) 97}115

Fig. 11. Comparison of blade tip de#ections [32].

standards [41]. Blade tip shapes, airfoil shapes, and a variable RPM system are investigated in this research program. In Japan, an Advanced Technology Institute of Commuter-Helicopter (ATIC) program was initiated to investigate rotor noise reduction concepts with helicopter manufacturers, academia, and research institutes, and generates a comprehensive test dataset from the German}Dutch acoustic wind tunnel (DNW) [30]. Blade tip shapes play an important role in blade airloads, which determine tip vortex formations and gener-

107

ate far-"eld noise. Earlier e!orts have been concentrated on modi"cation of blade tip shapes to redistribute blade airloads in order to di!use vortex structures or to induce vortex instability. Several blade tip shapes have been investigated, such as swept-tapered, sub-wing, end plate, #ow spoiler, and tip air mass injection (TAMI) blade as shown in Fig. 12. Particularly, an Ogee tip shape was intended to prevent the formation of a concentrated vortex. Experimental results have shown that these tip shapes produced some vortex di!usion in a near wake, but did not show signi"cant BVI peak noise reductions [42,43]. Recently, advanced blade planform shapes have been generated with swept, tapered and anhedral tips to achieve both high-speed impulsive and blade}vortex interaction noise reduction, and performance enhancement. Blade sweeps can be e!ective in reducing BVI noise by avoiding or delaying parallel interactions. Also, swept, tapered, and thin blade tips can be very e!ective in high-speed impulsive noise reduction by reducing compressibility e!ects and increasing the delocalization Mach number. Anhedral tips also have positive e!ects on reducing blade}vortex interaction noise and improving hover performance. Now, several advanced blade tip shapes from various organizations and national laboratories will be described. Variations from the basic UH-60 and S-76 planform shapes have been investigated to both reduce noise levels and improve hover performance. These tip shapes have been reported to have considerable impacts on BVI noise reduction, up to 5 dB [44]. A swept parabolic tip shape (PF2) and an optimized planform shape were developed and tested at ONERA as shown in Fig. 13. The measured acoustic results show that the optimized tapered tip produces less BVI noise than the swept parabolic tip shape [45]. The swept parabolic tip shape generates high blade control loads, and the nonlinear twist tapered blade encountered power penalties in high forward #ight. A quiet helicopter demonstrator program was developed in France with new main and tail rotors [41]. A 5-blade main rotor was optimized from new analytical methods and experience, and the blade tip has a tapered, parabolic shape with thin airfoils as shown in Fig. 14. Flight test and analytical results show a 3.3 dB reduction in approach and about 4.5 dB below the ICAO certi"cation limit. The German advanced technology rotor (ATR) program generates several advanced blade tip shapes with a sweep-back leading edge, a sweep-forward trailing edge, and a vane tip for noise reduction as shown in Fig. 15 [39]. Test results clearly show the bene"ts of the parabolic blade tip on noise reduction for all #ight conditions. Furthermore, a large rotor RPM range was also investigated. With reduced RPM, noise level reductions of about 1 dB can be achieved for approach and #yover conditions [39]. An Agusta A109C blade has a tapered

Jpas=76=EM=VVC=BG 108

Y.H. Yu / Progress in Aerospace Sciences 36 (2000) 97}115

Fig. 12. Rotor blade tip shapes proposed for BVI noise reduction [42].

Fig. 13. ONERA blade planform shapes [45].

tip, but with only trailing edge. It was reported that this blade planform shape provides a 3 EPNdB reduction in approach and a 2 EPNdB improvement in #yover noise as well as a 10 kt. increase in forward speed. An innovative BERP tip was developed by British researchers to achieve the world speed record by reducing shock waves and delaying blade stalls through a design combination of sweep, thin airfoils, and a leading edge notch as shown in Fig. 16. Not only the world speed record, but also substantial noise reduction was achieved [5]. A reduction of 13.5 dB in the negative impulsive peak pressure and up to 5 dB in low-speed descent #ights was reported compared to a conventional rectangular blade. This tip design concept is very di!erent from those concepts discussed earlier and is innovative in combining the e!ects of compressibility, retreating blade stall, and noise. With many aerodynamic, dynamic, and acoustic

Fig. 14. French quiet blade shape [41].

design constraints (or requirements), the BERP blade made a remarkable achievement. The e!ect of anhedral, dihedral, swept-forward, and swept-back blade shapes were investigated on blade} vortex interaction noise in Japan [25,46]. This activity was mainly concentrated on the e!ect of blade}vortex miss distances, interaction angles between blade and tip vortex trajectory, and tip vortex strength on blade} vortex interaction noise. First, in order to increase the blade}vortex miss distance, anhedral or dihedral tip shapes were used. It was concluded that the miss distance is increased as the anhedral or dihedral angle is increased. Second, swept-back or swept-forward tip shapes are used

Jpas=76=EM=VVC=BG Y.H. Yu / Progress in Aerospace Sciences 36 (2000) 97}115

109

Fig. 18. Planform shape of the vane tip [47].

Fig. 15. ATR blade tip shapes [39].

Fig. 19. BVI noise reduction concepts under development at NASA Langley [42].

Fig. 16. BERP blade features [5].

Fig. 17. BERP-like, AT1 planform shapes tested in the DNW [25].

to avoid parallel interactions. A BERP planform shape with AK080A/AK100D airfoils and an AK1 tip shape, as shown in Fig. 17, were tested in the DNW [25]. Noise levels were reduced with the AT1 planform shape compared to a baseline blade planform (rectangular shape) and BVI noise reduction is attributed to avoiding

parallel interactions due to its large sweep angle of the AT1 tip. Vane tip shapes were investigated to produce multiple vortices of equal strength spaced as far apart as possible to reduce BVI noise as shown in Fig. 18. This vane tip shape employs many aerodynamic design parameters such as sweep, taper, twist, and dihedral together with a discontinuity at the trailing edge. This tip shape was reported to have no adverse e!ect on control loads or performance and a maximum reduction of BVI noise of 5.6 dBA [47]. Noise reduction concepts with a porous leading edge and a large forward sweep blade have been suggested as shown in Fig. 19 [42]. The use of a porous leading edge is to mitigate high BVI-induced pressure #uctuations and is intended to reduce the amplitude of unsteady surface pressure #uctuations at the leading edge. A very large forward sweep blade is to reduce transonic e!ects and the occurrence of parallel interactions.

6. Active blade control concepts Blade}vortex interaction noise is dominated by wake and blade aeroelastic de#ections, such as tip vortex

Jpas=76=EM=VVC=BG 110

Y.H. Yu / Progress in Aerospace Sciences 36 (2000) 97}115

strength, blade airloads, induced velocity, blade #apping de#ection and miss distance. To control these parameters for noise and vibration reduction, active blade control concepts have been extensively investigated in recent years [48,49]. Active blade control inputs can modify the airload distribution along a blade, which generates a tip vortex with weak strength and an additional induced velocity "eld. Also, active blade control inputs can modify blade #apping and torsional de#ections, which increase blade}vortex miss distances during interactions. Therefore, rotor noise and vibration can be reduced by the combined e!ect of blade airloads and blade de#ections with proper control inputs. Earlier, a higher-harmonic control (HHC) concept was investigated to reduce rotor vibratory loads. The HHC technique excites a blade pitch angle at a "xed frequency through a rotor swashplate. The actuators are placed in the "xed frame of a rotor swashplate. Since the actuators are placed under the swashplate, all blades are excited simultaneously at a given frequency, amplitude and phase, while the frequency variation is limited to N!1, N, and N#1 per revolution (N being a rotor rotational frequency). The frequency variation is added to a trimmed blade pitch schedule and produce di!erent blade #apping and torsional deformation, which will change blade airloads. The blade airloads will modify wake structures, wake trajectories, and induced velocity "eld. Recently, this HHC technique was investigated to reduce rotor blade}vortex interaction (BVI) noise. Several wind tunnel and full-scale #ight tests with HHC inputs have demonstrated that rotor BVI noise can be reduced by about 6 dB with open-loop controls [50}57,7]. Experimental and analytical activities under a higherharmonic aeroacoustic rotor test (HART) program have investigated noise reduction mechanisms with HHC inputs in terms of tip vortex strength, tip vortex trajectory, induced velocity "eld, and blade #apping de#ection [7,23,27,32]. For the HHC pitch inputs for low noise, an enlarged, di!used tip vortex is generated by smooth airload distributions and is moved further down vertically with an additional increased induced downwash velocity "eld. At the same time, the blade tip, with the control inputs for low noise, #aps down at the time of vortex formation, then #aps up at the time of blade}vortex interaction. With the combined e!ects of increased induced downwash velocity and blade tip de#ections, the blade}vortex miss distance is substantially increased as shown in Fig. 10 and reduces blade}vortex interaction noise. However, this particular HHC input for low noise generates intense blade}vortex interactions at around 903 in azimuth and increases vibration levels [54]. For the HHC inputs of low vibration, nearly the opposite trends compared to the low-noise case were observed. Due to the decreased negative lift on a blade in the second quadrant (time of vortex formation), the in-

duced upwash velocity is increased and the tip vortex is pushed up, thereby signi"cantly reducing the miss distance as shown in Fig. 10. This explains the increased BVI noise measured for the low vibration case. Furthermore, a double vortex system, of a clockwise vortex near the rotor plane and a counterclockwise vortex, was also observed in this case [7,16]. To increase the control authority of input variables, an individual blade control (IBC) technique has been developed by independently actuating individual blades in a rotating frame [58]. The pitch actuators are placed above a swashplate, thus allowing arbitrary pitch motion for each individual blade. Flight tests [59}63] and fullscale wind tunnel tests [64}66] were carried out with a BO-105 helicopter with an IBC system. The test data have shown that simultaneous reduction in BVI noise and hub vibratory loads could be obtained using multiharmonic IBC inputs. 2/rev IBC inputs were found to generate large reductions in noise. The comparison of the baseline case with the minimum noise case indicates that the blade}parallel blade}vortex interactions on the advancing side in the azimuth range between 30 and 603 have been largely suppressed by IBC, which is the case on the retreating side between 300 and 3303 as well. A 2/rev #5/rev multi-harmonic combination in the wind tunnel test generated an average of 12 dB reduction along with signi"cant 4P rotor vibration reduction [64]. The data also showed that performance improvements of up to 7% could be obtained using 2/rev inputs at high-speed forward #ight conditions. From the #ight test, it was shown that 2/rev IBC inputs reveal the most promising potential and generate simultaneous noise and vibration reduction, up to 5 dBA as shown in Fig. 20 [63]. A BO-105 helicopter #ight test was carried out with upgraded IBC actuators, which have maximum amplitudes of $1.13 at a bandwidth of approximately 60 Hz (corresponding to 8/rev) [59,62]. The test results showed a more than 6 dB noise reduction with 2/rev harmonic inputs and also showed that at the optimum control phase angle for minimum noise the vibratory loads remain almost unchanged. Furthermore, the noise reduction with 2/rev IBC inputs in a wind tunnel was con"rmed with the #ight test, but this comparison was rather qualitative due to di!erent trim conditions and weather condition e!ects [62]. The #ight test results showed that the noise reduction with 2/rev IBC inputs was uniform over the wide range of lateral positions on ground, indicating that the noise reduction is more a global e!ect. Fig. 21 shows leading edge blade pressure distributions along azimuth angles at three radial stations for a BO-105 #ight test with and without 2/rev IBC inputs for a descent #ight condition [62]. With 2/rev IBC inputs, the blade pressures due to BVIs on the advancing side are reduced and also smoothed, resulting in noise reduction. Main disadvantages of current IBC systems are weight, mechanical complexity of hydraulic sliprings, and

Jpas=76=EM=VVC=BG Y.H. Yu / Progress in Aerospace Sciences 36 (2000) 97}115

111

Fig. 20. Sound pressure histories for a baseline case (no IBC) and for minimum noise IBC settings [63].

power requirements. In order to overcome the disadvantages of rotor root controls, smart structures are being developed as on-blade actuators in recent years. Since smart actuators can be installed on a rotor blade individually, blades can be controlled individually over a larger bandwidth than with current blade-root-control technologies such as HHC and IBC. In recent years, smart structures for rotor on-blade control applications have been used in blade trailing edge #aps, active blade twist, or leading edge #ow controls. Even though it has extremely high potentials, current smart structures technology needs more advancement for rotorcraft applications due to stringent requirements of frequency, force, displacement, weight, reliability, and cost. In earlier years for concept demonstrations, a wind tunnel rotor test with a mechanical non-harmonic trail-

Fig. 21. Leading edge (3% of the chord) blade pressure distributions for a BO-105 #ight test for a 63 approach and 65 KIAS [62].

ing edge #ap was carried out and showed a potential of at least 4 dB noise reduction [67,68]. Distinctive impulsive noise characteristics have disappeared with varying #ap motions in this wind tunnel test. This trailing edge #ap concept has been analytically investigated to show the local in#ow changes and reduce the induced pressure pulse at the leading edge during blade}vortex interactions. Recently with the same idea, a one-blade model rotor was tested with a mechanically active trailing #ap to evaluate its e!ect on BVI noise reduction with the

Jpas=76=EM=VVC=BG 112

Y.H. Yu / Progress in Aerospace Sciences 36 (2000) 97}115

measurements of blade surface pressures, blade}vortex miss distances, noise, and tip vortex structures [51,52]. It was concluded from this test that the blade}vortex miss distance is a dominant controlling parameter for noise reduction. Active trailing edge #aps and active blade twist are the two common on-blade actuation methods under development with smart structures in recent years. Active trailing edge #ap concepts with piezo-bimorphs, piezo stacks, and magnetostrictive actuators are being investigated to induce changes in blade airload distribution and blade #apping characteristics [69}74]. Since these smart actuators generate extremely small displacements, mechanical ampli"cation mechanisms are needed to achieve desired #ap de#ections. Frictional forces, dynamic pressures, and centrifugal forces are the major technical barriers for mechanical ampli"cation schemes. The major advantages of active trailing edge #aps are that active #aps can be placed at desirable positions on a blade, only a small portion of a blade can be actuated to modify blade airloads, lower actuation power is required, and hydraulic sliprings are eliminated. But the drawbacks are the requirement of mechanical ampli"cations of displacements and the mass-balancing weight penalty from adding #aps and associated hardware. Froude or Mach scaled rotor models with piezo-stack actuators were tested in hover and in a wind tunnel and successfully demonstrated the e!ectiveness of smart actuators in rotor. For example, #ap de#ections of $63 at 4/rev were achieved in hover at the University of Maryland [71]. An on-blade elevon concept with piezo-bender actuators was successfully demonstrated in reducing individual 3, 4, and 5/rev harmonic blade vibratory #ap bending moments [75]. A Mach scaled model rotor was developed with a double-x stroke ampli"cation mechanism for piezo-stack actuation of a trailing edge #ap [76]. Mach scaled rotor blades with piezobender actuation was tested in a vacuum chamber to achieve the trailing-edge de#ections of $83 [77]. For simple cases, a two-dimensional wing model with an active trailing edge #ap using piezoelectric actuators was tested in several Mach numbers, as part of the RACT project [72]. Another approach is an active blade twist concept, in which smart structures are imbedded in a blade to induce blade torsional deformation or blade twist. Active blade twist can be achieved through piezo-ceramic materials embedded along a blade span at 453 relative to the blade span axis [69}71]. The main advantages of this approach are that smart structures are an integral part of loadbearing structures and there are no moving mechanical parts. But the drawbacks are that the full blade has to be deformed and high power is required. Since a large number of distributed piezo-ceramic elements may be required to generate a su$cient blade twist distribution, blade weight and sti!ness can be signi"cantly increased.

A Froude scaled rotor model with specially shaped piezo-electric elements at $453 was developed, and the tested results showed the blade twist of $0.43 at the tip [78]. A concept for controllable twist blades with an active ply of PZT "bers oriented at $453 was used for a Machscaled model to obtain $23 of blade tip twist [79]. Other concepts have also been investigated in recent years. A moving blade tip concept was developed with a piezo-induced bending-torsion-coupled composite beam, and tip de#ections of $23 were achieved in a model rotor test [80]. An active blade tip concept was introduced to actively pitch a movable blade tip by a piezo-induced bending torsion coupled actuator beam [71]. A #ip-tip rotor blade concept was developed to control blade}vortex miss distances by rotating an outboard portion of the rotor tip either up or down [81]. Altering aerodynamic responses at the leading edge during blade}vortex interactions is another application of smart structures [69,82]. For example, a surface pressure response on a leading edge can be controlled by changing blade surface boundary conditions using mezo-scales smart actuators [83], drooping a leading edge, or changing leading edge curvatures with smart structures. However, these concepts need more advancement for rotor applications.

7. Concluding remarks Rotor blade}vortex interaction noise is the most severe problem in community acceptance of rotorcraft operations. Recently, physical understanding of the noise generation mechanism has been substantially improved with advanced analyses and comprehensive rotor tests. Various blade design concepts and active blade control concepts have been investigated, analytically and experimentally, in recent years to reduce noise radiation. With analytical and experimental results, rotor wakes and blade aeroelastic de#ections are identi"ed as the important parameters for noise generation and reduction. However, comprehensive test data and advanced numerical simulations of rotor wakes (up to two revolutions) are needed for better understanding of wake behavior and its e!ects on blade airloads. The current analytical capability for predicting blade airloads is not mature mainly due to lack of good modeling of rotor wakes, such as vortex core variations over two rotor revolutions and vortex trajectories. For better prediction of blade aeroelastic de#ections, a comprehensive analysis capability is needed with coupled aerodynamic and structural dynamic codes. Higher harmonic control (HHC) or individual blade control (IBC) techniques have been investigated for noise and vibration reduction. These blade root control techniques can modify blade airloads, wake trajectory/structure and blade #ap/torsion de#ections along azimuth

Jpas=76=EM=VVC=BG Y.H. Yu / Progress in Aerospace Sciences 36 (2000) 97}115

angles through proper blade pitch inputs in frequency and phase. This will control the blade}vortex miss distance during interactions and eventually control noise radiation. However, there are drawbacks in these techniques, such as blade control limits, power consumption, and blade hub complexity. In order to overcome the disadvantages of the blade root control techniques, smart structures/actuators are being developed for smart rotor technologies such as active trailing edge #aps or active blade twist, and the smart structures technology have an excellent potential to substantially advance rotor capabilities.

References [1] Schmitz FH, Yu YH. Helicopter impulsive noise: theoretical and experimental status. J Sound Vibration 1986;109(3):361}422. [2] Yu YH, Caradonna FX, Schmitz FH. The in#uence of the transonic #ow "eld on high-speed helicopter impulsive noise. 4th European Rotorcraft and Powered Lift Aircraft Forum, Italy, 1978. [3] George R. Helicopter noise * state of the art. J Aircraft 1978;15(11):707}15. [4] JanakiRam RD. Aeroacoustics of rotorcraft. AGARD Report 781, 1990. [5] Lowson MV. Progress towards quieter civil helicopters. Aeronaut J 1992:209}23. [6] Tangler JL. Schlieren and noise studies of rotors in forward #ight. 33rd American Helicopter Society forum, Washington, DC, 1977. [7] Splettstoesser WR, Kube R, Wagner W, Seelhorst U, Boutier A, Micheli F, Mercker E, Pengel K. Key results from a higher harmonic control aeroacoustic rotor test (HART). J Am Helicopter Soc 1997;42(1). [8] Lowson MV, Ollerhead JB. Studies of helicopter rotor noise. USAAVLABS Technical Report 68-60, 1968. [9] Widnall SE. Helicopter noise due to blade}vortex interaction. J Acoust Soc Amer 1971;50(1) (part 2). [10] Lowson MV. Directionality of helicopter BVI noise. European Rotorcraft Forum, St. Petersburg, Russia, 1995. [11] Sim BW, George AR. Development of a rotor aerodynamic load prediction scheme for blade}vortex interaction noise study. American Helicopter Society 51st Annual Forum, Ft. Worth, TX, 1995. [12] Sim BW, Schmitz FH, Acoustic phasing and ampli"cation e!ects of single rotor helicopter blade}vortex interactions. American Helicopter Society 55th Annual Forum. Montreal, Canada, 1999. [13] Schmitz FH, Boxwell DA, Lewy DA, Dahan C. Model- to full-scale comparisons of helicopter blade}vortex interaction noise. J Am Helicopter Soc 1984;29(2). [14] Shockey GA, Williamson TW, Cox CR. Helicopter aerodynamics and structural loads survey. 32nd Annual Forum of American Helicopter Society, Washington, DC, 1996. [15] Lorber PF. Aerodynamic results of a pressure-instrumented model rotor test at the DNW. 46th Annual Forum of the American Helicopter Society, Washington, DC, 1990.

113

[16] Mercker E, Pengel K. Flow visualization of helicopter blade tip vortices. 18th European Rotorcraft Forum, paper number 26. France: Avignon, 1992. [17] Horner MB, Saliveros E, McD Galbraith RA. An examination of vortex convection e!ects during blade}vortex interaction. Aeronaut J 1992:373}9. [18] Kitaplioglu C, Caradonna FX. A study of blade}vortex interaction aeroacoustics utilizing an independently generated vortex. AGARD Fluid Dynamics Panel Symposium on Aerodynamics and Aeroacoustics of Rotorcraft. Germany: Berlin, 1994. [19] Ehrenfried K. Meier GEA Obermeier F. Sound produced by vortex-airfoil interaction. 17th European Rotorcraft Forum, paper C91-63, Berlin, Germany, 1991. [20] Lee S, Bershader D. Head-on parallel blade*vortex interaction. AIAA J 1994;32(1):16}22. [21] Ffowcs Williams JE, Hawkings DL. Sound generation by turbulence and surfaces in arbitrary motion. Philosophical Trans Roy Soc London. Ser A 1969;264:321}42. [22] Farassat F. Theory of noise generation from moving bodies with an application to helicopter rotors. NASA TR R-451, 1975. [23] Beaumier P, Prieur J, Rahier G, Spiegel P, Demargne A, Kube R, van der Wall BG, Schultz KJ, Splettstoesser W, Tung C, Gallman J, Yu YH, Brooks TF, Burley CL, Boyd DD. E!ect of higher harmonic control on helicopter rotor blade*vortex interaction noise: prediction and initial validation. 75th Fluid Dynamics Panel Symposium on Aerodynamics and Aeroacoustics of Rotorcraft, Berlin, Germany, 1994. [24] Nakamura Y. Prediction of blade*vortex interaction noise from measured blade pressure. Seventh European Rotorcraft and Powered Lift Aircraft Forum, paper number 32, Garmisch-Partenkirchen, Germany, 1981. [25] Nishimura H, Kondo, N, Nakamura H, Tsujiuchi T, Yamakawa E, Aoyama T, Saito S. Comparison between calculated rotor noise and experimental data obtained by DNW test. 24th European Rotorcraft Forum, Marseilles, France, 1998. [26] Ochi A, Shima E, Yamakawa E, Aoyama T, Saito S. Aerodynamic and aeroacoustic analysis of BVI by moving overlapped grid method. 24th European Rotorcraft Forum, Marseilles, France, 1998. [27] Tung C, Kube R, Brooks TF, Rahier G. Prediction and measurement of blade}vortex interaction. J Aircraft 1998;35(2):260}6. [28] Yu YH, Tung C, Gallman J, Schultz KJ, van der Wall B, Spiegel P, Michea B. Aerodynamics and acoustics of rotor blade*vortex interactions. J Aircraft 1995;32(5):970}7. [29] Yu YH, Gmelin B, Heller H, Philippe JJ, Mercker E, Preisser JS. HHC aeroacoustics rotor test at the DNW * the joint German/French/US HART project. 20th European Rotorcraft Forum, Netherlands: Amsterdam, 1994. [30] Murashige A, Kobiki N, Tsuchihashi A, Nakamura H, Inagaki K, Yamakawa E. ATIC aeroacoustic model rotor test at DNW. Heli Japan 98, Gifu, Japan, 1998. [31] Young L. Tilt rotor aeroacoustic model (TRAM): a new rotorcraft research facility. Heli Japan 98, Japan: Gifu, 1998.

Jpas=76=EM=VVC=BG 114

Y.H. Yu / Progress in Aerospace Sciences 36 (2000) 97}115

[32] Lim JW, Tung C, Yu YH. Prediction of blade}vortex interaction airloads with higher-harmonic pitch controls using the 2GCHAS program. Heli Japan 98, Japan, Gifu, 1998. [33] Gallman JM, Tung C, Yu YH, Low SL. Prediction of blade}vortex interaction noise with applications to higher harmonic control. 15th AIAA Aeroacoustics Conference, AIAA 93-4331, CA, Long Beach, 1993. [34] Yu YH, Tung C, Low S. Blade aeroelastic e!ect on rotor blade}vortex interaction noise. 52nd American Helicopter Society Forum, Washington, DC, 1996. [35] Lee CS, Saberi H, Ormiston RA, Aerodynamic and numerical issues for coupling CFD into comprehensive rotorcraft analysis. 53rd Annual American Helicopter Society Forum, Virginia, Virginia Beach, 1997. [36] Childress O. National rotorcraft noise reduction program overview. NATO-CCMS Symposium on Noise Aspects of Rotary Wing Aircraft, 1991. [37] Schultz K-J, Splettstoesser W, Junker B, Wagner W. A parametric wind tunnel test on rotorcraft aerodynamics and aeroacoustics (HELISHAPE) * test procedures and representative results. 22nd European Rotorcraft Forum, Brighton, UK, 1996. [38] Splettstoesser WR, Niesl G, Cenedese F, Nitti F, Papanikas DG. Experimental results of the European HELINOISE aeroacoustic rotor test in the DNW. 19th European Rotorcraft Forum. Cernobbio, Italy, 1993. [39] M. Bebesel, Ploz G, Schoell E. Aerodynamic and aeroacoustic layout of the ATR (Advanced Technology Rotor). American Helicopter Society 55th Annual Forum. Montreal, Canada, 1999. [40] Schimke D, Janker P, Blaas A, Kube R, Schewe G, Kessler C. Individual blade control by servo-#ap and blade root control * a collaborative research and development program. 23rd European Rotorcraft Forum. Dresden, Germany, 1997. [41] Allongue M, Marze HJ, Potdevin F. The Quiet helicopter-from research to reality. American Helicopter Society 55th Annual Forum. Montreal, Canada, 1999. [42] Brooks TF. Studies of blade}vortex interaction noise reduction by rotor blade modi"cation. Proceedings of the 1993 National Conference on Noise Control, Williamsburg, VA, 1993. p. 57}66. [43] Hoad DR. Evaluation of helicopter noise due to blade}vortex interaction for "ve tip con"gurations. NASA TP-1608, 1979. [44] Martin RM, Connor AB, Wind-tunnel acoustic results of two rotor models with several tip designs, NASA TM87698, 1986. [45] Marze HJ, Philippe JJ. A quiet helicopter * a research program today, a reality tomorrow. 20th European Rotorcraft Forum, Paper 9, Amsterdam, Netherlands, 1994. [46] Nakamura H, Nishimura H, Kondo N, Yamakawa E, Aoyama T, Saito S. E!ect of blade geometry on BVI noise in various #ight conditions. Heli Japan 98. Gifu, Japan, 1998. [47] Brocklehurst A, Pike AC. Reduction of BVI noise using a vane tip. American Helicopter Society Aeromechanics Specialists Conference, San Francisco, CA, 1994. [48] Teves D, Niesl G, Blaas A, Jacklin SA. The role of active control in future rotorcraft. 21st European Rotorcraft Forum, Saint-Petersburg, 1995.

[49] Yu YH, Gmelin B, Splettstoesser W, Philippe JJ, Prieur J, Brooks T. Reduction of helicopter blade}vortex interaction noise by active rotor control technology. Prog Aerospace Sci 1997;33:647}87. [50] Brooks TF, Booth ER, Jolly JR, Yeager WT, Wilbur ML. Reduction of blade}vortex interaction noise through higher harmonic pitch control. Am Helicopter Soc J 1990;35(1):86}91. [51] Kobiki N, Tsuchihashi A, Murashige A, Yamakawa E. Elementary study for the e!ect of HHC and active #ap on blade}vortex interaction. Heli Japan 98, Gifu, Japan, 1998. [52] Kobiki N, Tsuchihashi A, Murashige A, Yamakawa E. Study for the e!ect of active #ap on blade}vortex interaction. Heli Japan 98, Gifu, Japan, 1998. [53] Kube R, Splettstoesser WR, Wagner W, Seelhorst U, Yu YH, Boutier A, Micheli F, Mercker E. Initial results from the higher harmonic control aeroacoustic rotor test (HART) in the German}Dutch wind tunnel. 75th Fluid Dynamics Panel Symposium on Aerodynamics and Aeroacoustics of Rotorcraft, Berlin, Germany, 1994. [54] Kube R, van de Wal B, Schultz KJ. Mechanisms of vibration and BVI noise reduction by HHC. 20th European Rotorcraft Forum, Amsterdam, The Netherlands, 1994. [55] Polychroniadis M. Generalized higher harmonic control * ten years of aerospatiale experience. 16th European Rotorcraft Forum, Glasgow, UK, 1990. [56] Splettstoesser WR, Lehmann G, van der Wall B. Higher harmonic control of a helicopter rotor to reduce blade}vortex interaction noise. 15th European Rotorcraft Forum, 1989. [57] Splettstoesser WR, Schultz KJ, Kube R, Brooks TF, Booth ER, Niesl G, Streby O. A higher harmonic control test in the DNW to reduce impulsive BVI noise. J Am Helicopter Soc 1994;39(4). [58] Ham ND. Individual blade control and its application. Annual Forum of the American Helicopter Society, 1983. [59] Morbitzer D, Arnold UTP, Muller M. Vibration and noise reduction through individual blade control experimental and theoretical results. 24th European Rotorcraft Forum, Marseilles, France, 1998. [60] Niesl G, Arnaud, G. Low noise design of the EC135 Helicopter. 52nd Annual Forum of the American Helicopter Society, Washington DC, 1996. [61] Richter P, Eisbrecher HD, Kloeppel V. Design and "rst tests of individual blade control actuators. 16th European Rotorcraft Forum, Glasgow, UK, 1990. [62] Scholl E, Gembler W, Bebesel M, Splettstoesser W, Kube R, Pongratz R. Noise reduction by blade root actuation * analysis of #ight and wind tunnel tests. 24th European Rotorcraft Forum, Marseilles, France, 1998. [63] Splettstoesser WR, Schultz KJ, van der Wall B, Buchholz H, Gembler W, Niesl G. The e!ect of individual blade pitch control on BVI noise * comparison of #ight test and simulation results. 24th European Rotorcraft Forum, Marseilles, France, 1998. [64] Jacklin SA, Blaas A, Teves D, Kube R. Reduction of helicopter BVI noise, vibration, and power consumption through individual blade control. 51st American Helicopter Society Forum, Fort Worth, TX, 1995.

Jpas=76=EM=VVC=BG Y.H. Yu / Progress in Aerospace Sciences 36 (2000) 97}115 [65] Niesl G, Swanson SM, Jacklin SA, Blaas A, Kube R. E!ect of individual blade control on noise reduction. AGARD Conference on Aerodynamics and Aeroacoustics of Rotorcraft, Berlin, Germany, 1994. [66] Swanson SM, Jacklin SA, Blaas A, Niesl G, Kube R. Acoustic results from a full-scale wind tunnel test evaluating individual blade control. 51st American Helicopter Society Forum, Fort Worth, TX, 1995. [67] Marcolini MA, Booth ER, Tadghighi H, Hassan AA, Smith CD, Becker LE. Control of BVI noise using an active trailing edge #ap. American Helicopter Society Vertical Lift Aircraft Design Conference, San Francisco, CA, 1995. [68] Straub FK, A feasibility study of using smart materials for rotor control. 49th Annual Forum of the American Helicopter Society, Louis, MO, 1993. [69] Chopra I. Development of a smart rotor. 19th European Rotorcraft Forum, Cernobbio, Italy, 1993. [70] Chopra I. Application of smart structures technology to rotorcraft systems. Smart Mater Struct 1996;5(1). [71] Chopra I. Status of application of smart structures technology to rotorcraft systems. Royal Aeronautical Society Conference on Innovation in Rotorcraft Technology, London, UK, 1997. [72] Schimke D, Janker P, Wendt V, Junker B. Wind tunnel evaluation of a full scale piezoelectric #ap control unit. 24th European Rotorcraft Forum, Marseilles, France, 1998. [73] Spangler RL, Hall SR. Piezoelectric actuators for helicopter rotor control. 31st AIAA Structures, Structural Dynamics and Material Conference, 1990. [74] Straub FK. Application of smart materials to helicopter rotor active control. SPIE Symposium on Smart Structures and Materials, San Diego, CA, 1997.

115

[75] Fulton MV, Ormiston RA. Hover testing of a smallscale rotor with on-blade elevons. 53rd Forum of the American Helicopter Society, Virginia Beach, VA, 1997. [76] Prechtl EF, Hall SR. Design of a high e$ciency discrete servo-#ap actuator for helicopter rotor control. 1997 SPIE Conference on Smart Structures, San Diego, CA, 1997. [77] Koratkar NA, Chopra I. Design, fabrication and testing of a Mach scaled rotor model with trailing-edge #aps. American Helicopter Society 55th Annual Forum, Montreal, Canada, 1999. [78] Chen PC, Chopra I. Development of a smart rotor blade with induced strain actuation of blade twist. AIAA J 1997;35(1):6}10. [79] Rodgers JP, Hagood NW. Design and manufacture of an integral-twist actuated rotor blade. 38th AIAA/ASME/ASCE/AHS Structures, Structural Dynamics and Materials Conference, Kissimmee, FL, 1997. [80] Bernhard A, Chopra I. Development of a smart moving blade tip activated by a piezo induced bending-torsion coupled beam. SPIE Conference on Smart Structures and Integrated Systems, San Diego, CA, 1996. [81] Wells VL, Glinka AT. The #ip tip } a novel method for reducing blade}vortex interaction noise. Royal Aeronautical Society Conference on Innovation in Rotorcraft Technology, London, UK, 1997. [82] Amerigo MP, Baeder JD. Feasibility of arbitrary pitching motion controlled by piezoceramic actuators to reduce blade}vortex interaction noise. American Helicopter Society 51st Annual Forum, Ft. Worth, TX, 1995. [83] Nitzsche F. Tailoring the dynamic response of helicopter rotors using smart materials. AIAA/ASME Adaptive Structures Forum, AIAA 95-1083, New Orleans, LA, 1995.