REVIEW OF DRAG REDUCTION AND AERODYNAMIC NOISE CONTROL OF BLUFF BODIES

SHASHANK JAGTAP NATIONAL INSTITUTE OF TECHNOLOGY CALICUT

1. INTRODUCTION In aerodynamics, a bluff body is one which has a length in the flow direction close or equal to that perpendicular to the flow direction. This spawns the characteristic that the contribution of skin friction is much lower than that of pressure to the integrated force acting on the body. Even a streamlined body such as an airfoil behaves much like a bluff body at large angles of incidence. A circular cylinder is a paradigm often employed for studying bluff body flows. Bluff body flows are characterized by flow separation which produces a region of disturbed flow behind, i.e. the wake. Within the near-wake various forms of flow instabilities, both convective and absolute, may be triggered and amplified. These instabilities are manifested by the generation of two- and three-dimensional unsteady flow structures and eventually turbulence as the Reynolds number (Re = UD/, where U is the incident flow velocity, D is the width of the body and is the kinematic viscosity of the fluid) is progressively increased. The most well-known instability is that leading to the periodic formation and shedding of spanwise vortices which produce an impressive wake pattern named after von Krmn and Benhard. The frequency of vortex shedding is characterized by the Strouhal number, St =(fvsD) / U Which is a function of the Reynolds number, St(Re). The self-sustained wake oscillations associated with vortex shedding are incited through a Hopf bifurcation as predicted from linearized stability theory (Triantafyllou et al., 1987) and remain intact over a wide range of Reynolds numbers from approximately 50 to 106 or even higher. Bluff body flows involve the interaction of three shear layers, namely the boundary layer, the separating free shear layers and the wake. The physics of vortex formation and the near-wake flow have been the subject of many past experimental and numerical studies which have provided a wealth of information. The reader is referred to Bearman (1997) for a pertinent review. The basic mechanism of vortex formation is essentially two dimensional although there are inherent three-dimensional features for Reynolds numbers above 190. As well as contributing to time-averaged aerodynamic forces, vortex shedding is responsible for the generation of fluctuating forces acting on the body. This has several ramifications in practical applications as it may excite flow-induced vibration and acoustic noise. In order to modify the aerodynamic characteristics of bluff bodies, e.g., for control of drag, vibration and noise, it is imperative to control the separated flow in the wake and the dynamics of vortex formation which

are the sources of the fluid forces on the body. Such control can be achieved by both active and passive methods. In active control, it is typical to perturb the wake flow by some sort of excitation mechanism, e.g., rectilinear or rotational oscillations of the body, flow and/or sound forcing, fluid injection and/or blowing (synthetic jets), etc. Some of these methods affect the local flow while others influence the flow globally. The frequency and amplitude of the imposed perturbations provide a means to achieve the desired objective. Passive control is achieved by modification of the cylinders geometry, e.g., by the addition of tabs, strakes or bumps; these mainly influence the three-dimensional flow characteristics. A main characteristic of the wake response to periodic perturbations is the vortex lock-on Phenomenon whereby the imposed perturbation takes over control of the vortex formation and shedding over a range of conditions in the amplitudefrequency parameter space. For excitation along the flow direction, lock-on occurs when the driving frequency is near twice that of vortex shedding whereas the perturbations need be near the vortex shedding frequency for perturbations transverse to the flow direction. Lock-on is accompanied by a resonant amplification of both mean and fluctuating forces acting on the body but other types of wake response can take place, e.g., different modes of vortex formation can be incited such as symmetric (Konstantinidis & Balabani, 2007). These interactions provide a means for wake flow control. This paper presents issues related to the fluid dynamics of bluff bodies in steady and Time-dependent flows. The importance of vortex shedding for the generation of aerodynamic forces is exemplified with reference to numerical simulations of two-dimensional fluid flow about a circular cylinder at low Reynolds numbers. Results are presented for the flow patterns in the near-wake and fluid-induced forces exerted on the cylinder in response to flows with superimposed harmonic and non-harmonic perturbations in velocity. Implications for wake control are also discussed. It seems reasonable to concentrate on the stable range of Reynolds number (laminar wake regime) where the effects of simple changes can be studied without the complications attending the presence of turbulent flow. 2. PASSIVE VENTILATION The drag of bluff bodies is determined mainly by the pressure drag caused by flow separation. The separation region, or wake, often extends over a larger cross-section than the cross-section of the body. This leads to a big pressure difference between front and rear surfaces. Additionally, the shear layers of the separated flow tend to be unstable, and concentration of vorticity results in huge vortices in the wake. Because of this unsteadiness the drag forces are also unsteady with respect to magnitude and direction, thus generating vibrations of the body.

A simple method to overcome these problems is creating passive ventilation of the wake region. The bypass flow has mainly two effects. The first is a simple pressure increase in the near-wake regime by mass addition of the vented air. The second is an addition of momentum from the jet emanating from the rear end of the venting bore. This jet flow, which is driven by the pressure difference between the stagnation pressure and the remaining lower pressure level of the separation area, aligns the free shear layers of the wake to itself and draws them much closer to the surface of the bluff body. This leads to a more streamlined shape of the surrounding flow field which finally results in a much lower pressure drag. Also, the jet generates a ring vortex symmetric to the axis close to the rear side of the body which increases the pressure at parts of the rear surface of the body and reduces unsteady flow components in the wake. This lowers not only the drag but also the unsteady forces on the body and thus decreases body vibrations.

3. PLASMA ACTUATORS FOR NOISE CONTROL Over the past few decades, the popularity of air travel has increased significantly, which has led to increase in air traffic and drawn the attention of communities that live near airports to the problem of noise pollution (Raman and McLaughlin 2000). Airframe noise has been widely demonstrated to be a main contributor of aircraft noise during the

approach to landing phase, especially when the aircraft is equipped with modern high bypass ratio engines (Crighton 1991; Macaraeg 1998). Since airframe noise is normally caused by the interactions between the aircraft surface geometry and turbulent flow, various flow control strategies, including passive and active control methods (Gad-el 2000), have been explored to modify the flow structures around the airframe components to reduce the corresponding aerodynamically generated noise. A main source of airframe noise is the landing gears where fairings can be applied, potentially leading to a reduction of radiated noise by shielding the downstream components from the oncoming flow (Dobrzynski et al. 2002; Piet et al. 2005; Molin et al. 2006). This form of passive control is limited by practical considerations such as the need for easy access for maintenance and weight increase. An experimental study was conducted to explore the potential of plasma actuators to attenuate broadband bluff body noise. Both aeroacoustic and aerodynamic measurements were taken on a representative bluff body consisting of a circular cylinder and an oblique strut, in an open jet located in an anechoic chamber and a low speed open-loop wind tunnel. Microphone array measurements confirm that the major noise sources for this configuration were located in the gap area between the cylinder and the downstream strut. The results demonstrated that plasma actuators could reduce the broadband noise generated by the bluff body. Particularly, it was shown that the upstream directed plasma forcing located on the cylinder surface at 90 deg with respect to the approaching flow could effectively attenuate the major broadband noise radiated from the gap area. At the same AC electrical power consumption level, the sliding discharge with additional high DC voltage input is more effective due to its elongated plasma distribution and a higher induced flow momentum when compared to the dielectric barrier discharge. Particle image velocimetry measurements confirm that the noise attenuation with the upstream directed plasma forcing was due to a reduction of the flow speed impinging on the downstream strut. The downstream directed plasma forcing was found to have little effect on the current bluff body configuration in noise control. The broadband noise radiation by the bluff body could be significantly reduced at the free stream speeds up to 30 m/s under the present actuator authority of sliding discharge. As the wind speed was increased above 30 m/s, the current design gradually lost its effectiveness for flow and noise control. Therefore, for a practical application in the flight of much higher speeds (typical approach speed of about 80 m/s) and Reynolds number, the plasma actuation authority must undergo a commensurate increase. The actuation strategy applied in our experiments presented here is far from optimum since only steady actuation and limited actuators were used. Under the current actuator design, only a little bit higher induced wind speeds could be achieved by increasing the AC and DC voltages before a filamentary structure or sparks between the two surface electrodes appeared, after then the induced

speeds dropped dramatically. The shape of the two exposed electrodes for sliding discharge could be optimized to achieve higher DC corona discharge on the surface. The silicon rubber was selected as the dielectric layer for the actuator mainly due to its flexibility to wrap the bluff body, and therefore other materials with improvement in dielectric strength and constant would no doubt improve the actuator performance. Although the 90 deg location with respect to the approaching flow is more effective than other locations, there is no doubt that with more actuators distributed on the surface of the bluff body, not just at 90 deg, would also achieve higher plasma forcing. This benefit with more actuators has been demonstrated in Thomas et al. (2008). The significant effect of the concept using direct upstream plasma forcing presented here demonstrated that the plasma actuator can generate a virtual fairing that partly shields the downstream small component from the oncoming free stream flow, hence reducing the corresponding highfrequency noise. In this case, the plasma works in the same way as the passive control using solid fairings. 4. VORTEX SHEDDING- BUILDINGS A principal feature of bluff bodies is that they create separated flow regions which become the source of vortex shedding. Vortex shedding excitation can create problems in a variety of contexts in wind engineering. A prime example is the across-wind excitation of very tall buildings by vortex excitation. It has been seen that the mean moment is zero but the peak values in the negative and positive directions are at a maximum. It is quite often the case that the highest overall wind loading on a tall slender building results from across-wind vortex excitation which induces a large dynamic response. The resulting motions of the building may cause discomfort to the building occupants and it becomes a major concern of the structural designer and architect as to how they can keep these motions to within acceptable limits. The solution was developed in the form of additional structure, modified shape and the installation of a large 740 ton tuned mass damper near the top of the tower. The shape modification was to soften the building corners. This reduced the base wind-induced base bending moments by approximately 25%, which solved some problematic foundation loads. The building motions were also reduced by the shape change but this was not enough on its own to bring the accelerations to within acceptable limits. Hence the need for a tuned mass damper which has been described by Irwin and Breukelman (2001). An important factor affecting the strength of vortex excitation is the turbulence intensity of the wind. In the example of Taipei 101 there is no structure of comparable size anywhere near it and the nearby terrain is largely flat. Therefore, the turbulence intensity in the top region is quite low. This accentuated vortex shedding. Towers in a

location like downtown Manhattan see an entirely different type of approaching wind, full of turbulence from surrounding buildings. This tends to suppress vortex excitation from the building itself but can lead to strong buffeting effects coming from the wakes of other buildings. An example of where wake buffeting caused motion problems was the 67 storey Park Hyatt Tower in Chicago where the wake of the existing John Hancock Tower caused severe wake buffeting for north easterly wind directions. This was solved through the installation of a 300 ton tuned mass damper in the Park Hyatt Tower. A known solution to vortex excitation is to introduce setbacks into the tower at various levels. The variations in cross-section with height confuse the vortices being shed by the tower. Not only is the across-wind width of the structure varying with height but also is the Strouhal number. Therefore, the shedding frequency varies with height and the aerodynamic excitation loses its coherence. 5. BLUFF BODY BUFFETING Bluff body wake interference (buffeting) is of great practical interest for wind engineering. In order to develop valid modelling for buffeting loading the flow phenomenology is first investigated via Strouhal number, induced force coefficients, flow visualisations and spectral analysis. Extensive analysis of these measurements reveals the flow mechanics and allows a cognitive partition of the buffeting domain for both tandem and staggered cases in smooth, large- and small-scale turbulent flows. 6. WAKE SPLITTER METHOD It was shown in Roshko (1954) that placing a splitter plate of sufficient length behind a circular cylinder would inhibit the formation of a regular vortex pattern downstream of flow separation. The pressure measurements (at c/H 2%) on a flat plate (=90O) in Gaster and Ponsford (1984), however, demonstrate that placing a splitter plate behind the plate will only alter the front face pressure slightly, even though the base pressure is changed significantly from -1.38 to -0.62 by the reduction in the strength of shed vortices. Furthermore, it was emphasized in Gaster and Ponsford (1984) that the theoretical prediction from Parkinson and Jandali (1970) would provide accurate estimates of the front -face pressures when the corresponding base pressure was used.

7. CONCLUSION Above mentioned where few of the methods which were developed by various observers for the reduction of drag and aerodynamic noise. The results obtained were carried out by experimentation and where crossed

checked by theoretically. The passive ventilation method stated above reduces the drag by around 60%. The experimentation with plasma actuators which was conducted by Yong Li , Xin Zhang , Xun Huang; states that the plasma actuators plasma actuator can generate a virtual fairing that partly shields the downstream small component from the oncoming free stream flow, hence reducing the corresponding high-frequency noise. The vortex shedding method for building is from safety and economic point of view. The wake flow can be controlled by perturbations in the incident flow velocity. This control can be exercised by appropriately selecting the frequency and amplitude of the imposed perturbations on both the mean and fluctuating wake flow and, thereby, on the fluid forces on the any bluff body. Whereas harmonic perturbations have been often utilized in the past, the present study also addresses the effect of non-harmonic perturbations, i.e. perturbations whose waveform deviates from a puretone harmonic. The findings of the present study illustrate that more effective wake control can be achieved by adjusting the waveform in addition to the perturbation frequency and amplitude. By stretching the bluff body, such as adding a fairing to the backside of a circular cylinder, the flow over the body changes. The adverse pressure gradient reduces as the body becomes more streamlined. Therefore, drag is also reduced. By making the body asymmetric, forces are generated that can be used in modern aerospace applications.

8. REFERENCES

1. Bearman, P. (1997). Near wake flows behind two- and threedimensional bluff bodies, Journal of Wind Engineering and Industrial Aerodynamics Vol. 71: 3354. 2. Gerrard, J.H. (1966). The three-dimensional structure of the wake of a circular cylinder, Journal of Fluid Mechanics Vol. 25(Part 1): 143164.

3. Griffin, O. M. & Hall, M. S. (1991). Vortex shedding lock-on and flow control in bluff body wakes Review, Journal of Fluids Engineering Vol. 113(No. 4): 526537. 4. Henderson, R. (1995). Details of the drag curve near the onset of vortex shedding, Physics of Fluids Vol. 7(No. 9): 21022104. 5. Konstantinidis, E. & Balabani, S. (2007). Symmetric vortex shedding in the near wake of a circular cylinder due to stream wise perturbations, Journal of Fluids and Structures Vol. 23: 10471063. 6. Konstantinidis, E. & Balabani, S. (2008). Flow structure in the lockedon wake of a circular cylinder in pulsating flow: Effect of forcing amplitude, International Journal of Heat and Fluid Flow Vol. 29(No. 6): 15671576. 7. Konstantinidis, E. & Bouris, D. (2010). The effect of non-harmonic forcing on bluff-body aerodynamics at a low Reynolds number, Journal of Wind Engineering and Industrial Aerodynamics Vol. 98: 245252. 8. Norberg, C. (2003). Fluctuating lift on a circular cylinder: review and new measurements, Journal of Fluids and Structures Vol. 17(No. 1): 5796. 9. Papadakis, G. & Bergeles, G. (1995). A locally modified 2nd-order upwind scheme for Convection terms discretization, International Journal of Numerical Methods for Heat and Fluid Flow Vol. 5( No. 1): 4962. 10. Rhie, C.M. & Chow,W. L. (1983). Numerical study of the turbulent Flow past an airfoil with trailing edge separation, AIAA Journal Vol. 21(No. 11): 15251532. 11. Triantafyllou, G., Kupfer, K. & Bers, A. (1987). Absolute instabilities and self-sustained oscillations in the wakes of circular-cylinders, Physical Review Letters Vol. 59(No.17): 1914-1917. 12. Wu, M. H., Wen, C. Y., Yen, R. H., Weng, M. C. & Wang, A. B. (2004). Experimental and numerical study of the separation angle for flow around a circular cylinder at low Reynolds number, Journal of Fluid Mechanics Vol. 515: 233260. 13. Research at DLR G. ottingen on bluff body aerodynamics, drag reduction by wake Ventilation and active flow control F.-R. Grosche, G.E.A. Meier.Journal of Wind Engineering and Industrial Aerodynamics 89 (2001) 12011218. 14. On the Reynolds number sensitivity of the aerodynamics of bluff bodies with sharp edges G.L. Larosea, A. DAuteuilb. Journal of Wind Engineering and Industrial Aerodynamics 94 (2006) 365376. 15. Bluff body aerodynamics in wind engineering-Peter A. Irwin. Journal of Wind Engineering and Industrial Aerodynamics 96 (2008) 701712. 16. Self-similarity of confined flow past a bluff body-W.W.H. Yeung. Journal of Wind Engineering and Industrial Aerodynamics 96 (2008) 369388.

17. The effect of non-harmonic forcing on bluff-body aerodynamics at a low Reynolds number E. Konstantinidis, D.Bouris. . Journal of Wind Engineering and Industrial Aerodynamics 98 (2010) 245252. 18. Buffeting of two-dimensional bluff Bodies- H. Hangan, B.J. Vickery. . Journal of Wind Engineering and Industrial Aerodynamics 82 (1999) 173}187. 19. Bluff Body Aerodynamics and Wake Control Efstathios Konstantinidis and Demetri Bouris ,Department of Mechanical Engineering, University of Western Macedonia, Bakola & Sialvera, Kozani, Greece. 20. The use of plasma actuators for bluff body broadband noise controlYong Li ,Xin Zhang , Xun Huang. Received: 7 July 2009 / Revised: 20 December 2009 / Accepted: 21 December 2009. Springer-Verlag 2010.