1.3.1 Gurney Flap

The Gurney flap is a simple device, such as a flat plate, which can be easily attached to the pressure surface of an airfoil, as shown in Figure 1.3. The size of the Gurney flap is only of the order of about 1% chord length of the airfoil, but can significantly increase the lift coefficient. There have been numerous works to show its effects on airfoils, wings and aircrafts, which has been reviewed by Wang et al. (Reference Wang, Li and Choi2008). Under certain conditions, both the lift coefficient and lift-to-drag ratio could be increased, though there is a drag penalty. Thus, the Gurney flap shows great potential application to shorten the takeoff/landing distance of aircraft.

Figure 1.3 Schematic of a Gurney flap mounted on the pressure surface of an airfoil.

1.3.2 Vortex Generator

A vortex generator is usually a small device placed onto the wall, which can induce a streamwise vortex. There are many devices that can be used as a vortex generator, such as flat plate, wishbone, doublet, airfoil, wedge, ramp, etc. The devices are usually arranged in one row along the spanwise direction to induce a set of vortices, either in a co-rotating manner where each device is inclined in the same direction (Figure 1.4(a)), or a counter-rotating manner where a pair of devices induces vortices with different rotation directions (Figure 1.4(b)). The induced streamwise vortices have the great ability to enhance momentum mixing, which is beneficial for separation flow control. Thus, vortex generators have been applied in various fields for flow control, ranging from low-speed to high-speed fields, which has been reviewed by Lin (Reference Lin2002) and Lu et al. (Reference Lu, Li and Liu2012).

(a) Co-rotating arrangement;

(b) counter-rotating arrangement

Figure 1.4 Schematic of vortex generators.

1.3.3 Bump

A bump is usually used to control the shock wave of a transonic airfoil and thus to reduce its drag. It is placed in the downstream of the normal shock wave over the airfoil suction surface, as shown in Figure 1.5. The bump could induce λ-shock structures that can reduce the impact of shock waves by replacing a single normal shock wave with several shock legs. The total pressure loss through a series of oblique shock waves is usually smaller than that across a normal shock wave, thus the wave drag can be greatly reduced. Milholen and Owens (Reference Milholen and Owens2005) and Li et al. (Reference Li, Zhang, Chen and Chen2011) found that drag reduction of around 12%–15% could be achieved at Mach numbers of 0.73 to 0.78. Besides, the bump could also be used to delay buffet onset and expand buffet boundary (such as Tian et al. Reference Tian, Liu and Peng2011).

Figure 1.5 Schematic of an airfoil with a bump.

1.3.4 Cavity

A cavity with trapped vortex is usually used for the control of airfoils. A cavity of suitable shape is positioned along the spanwise direction over the suction surface of the airfoil, as shown in Figure 1.6. A large-scale vortex will be induced in the cavity, creating a recirculation region closed by the dividing streamline (Lasagna et al. Reference Lasagna, Donelli, De Gregorio and Luso2011). The flow over the airfoil suction surface will be modified. In particular, the flow over the airfoil with a cavity separates before the forward edge of the cavity at high angles of attack. The separated flow displays a strong interaction with the cavity, causing the flow to shed smaller-scale vortical structures than the airfoil without a cavity. Thus, the near wake becomes narrower and the lift-to-drag ratio is increased with cavity control, in comparison with the clean airfoil (Olsman and Colonius Reference Olsman and Colonius2011). In order to stabilize the trapped vortex, additional blowing or suction control could be used, such as that conducted by Iollo and Zannetti (Reference Iollo and Zannetti2001) and Olsman et al. (Reference Olsman, Willems, Hirschberg, Colonius and Trieling2011).

Figure 1.6 Schematic of an airfoil with a cavity.

1.3.5 Roughness

There have been numerous works to show the effects of roughness on laminar flow, flow transition, and turbulent flow, such as those reviewed by Jiménez (Reference Jimenez2004). It is well known that roughness may promote flow transition, and one example is shown in Figure 1.7(a) for the golf ball, which has global dimples to make it a roughness surface. The dimples over the golf surface could accelerate flow transition to turbulence, which enhances momentum mixing to enable fluids to resist adverse pressure gradient. Thus, the flow may separate at around 120° from the front stagnation point, in comparison with the separation angle of about 80° for the laminar flow over a smooth ball. Thus, the golf ball has a smaller pressure-difference drag and could move further than a smooth one.

Figure 1.7

(a) Effect of dimples on a sphere.

(b) Schematic of roughness elements placed on a flat plate.

However, pioneer work conducted by Fransson et al. (Reference Fransson, Talamelli, Brandt and Cossu2006) indicated that a row of cylindrical roughness elements placed on a flat plate (Figure 1.7(b)) with specific height and spacing could effectively delay flow transition. In particular, an original turbulent flow could be changed to a laminar flow. Subsequent studies have also indicated the effect of the roughness elements on the bypass transition and the turbulent flow.

1.3.6 Small Disturbance

A small disturbance may trigger the instability of the local flow to further influence the global field. Some examples of disturbance are the trip wire (Figure 1.8(a)) and the small rod (Figure 1.8(b)). A trip wire is usually used to promote flow transition to turbulence for different objects, including a cylinder (such as Ekmekci and Rockwell Reference Ekmekci and Rockwell2010) and a sphere (such as Son et al. Reference Son, Choi, Jeon and Choi2011). However, it is mostly used for the boundary layer experiment, when a turbulent flow is needed.

Figure 1.8

(a) Schematic of a circular cylinder with a trip wire

(b) and a small rod in the upstream position

A rod is usually a much smaller cylinder in comparison with the scale of the controlled bluff bodies, including a square cylinder (such as Zhang et al. Reference Zhang, Wang, Lu and Mi2005), a circular cylinder (such as Wang et al. Reference Wang, Zhang, Lu and Wu2006; Zhang et al. Reference Zhang, Wang and Huang2006a), a disc (such as Zhang et al. Reference Zhang, Gao and Wang2006b; Wang et al. Reference Wang, Pan, Choi, Gao and Lian2013b), etc. The small rod could be placed upstream or downstream of the bluff body with a staggered angle. The wake induced by the small rod may interact with the flow over the bluff body, which could result in a global variation of the bluff body’s wake and a reduction in drag. For example, Zhang et al. (Reference Zhang, Wang, Lu and Mi2005) and Wang et al. (Reference Wang, Zhang, Lu and Wu2006) found that there were six different wake modes induced by a small rod upstream of the bluff body, namely cavity flow, wake impinging, wake merging, wake splitting, weak boundary layer interaction mode and negligible interaction. In particular, a maximum drag reduction by about 98% was found for the cavity flow mode.

1.3.7 Bleed

Bleed control is achieved by machining narrow slots across the controlled body. Fluids would flow across the slot and form a localized jet from the exit. One application is to control the circular or square cylinder, and there are several different arrangements of the slot. One slot could be across the front and rear stagnation points (such as Fu and Rockwell Reference Fu and Rockwell2005; Baek and Karniadakis Reference Baek and Karniadakis2009), as shown in Figure 1.9(a). Also, two slots can be used, which could be parallel to the streamwise direction (such as Aydın et al. Reference Aydin, Cetiner and Unal2010), as shown in Figure 1.9(b), or with some inclined angle (such as Shi and Feng Reference Shi and Feng2015). For all cases, the jets issuing from the slots will interact with the boundary layer or shear layer, leading to an ideal control effect. Baek and Karniadakis (Reference Baek and Karniadakis2009) and Shi and Feng (Reference Shi and Feng2015) indicated that the bleed control could convert the original asymmetric vortex shedding mode to the symmetric one, which is beneficial for the suppression of vortex-induced vibration.

(a) One slot across the front and rear stagnations points;

(b) two parallel slots; and

(c) two inclined slots.

Figure 1.9 Schematic of a circular cylinder with bleed control.

1.3.8 Splitter Plate

A splitter plate is a flat plate placed upstream (Figure 1.10(a)), downstream (Figure 1.10(b)), or both upstream and downstream (Figure 1.10(c)) of the object along the streamwise direction. Note that it is not necessary to connect the splitter plate to the controlled body surface; there could be a distance between them. Celik et al. (Reference Celik, Akdag, Gunes and Beskok2008) indicated that the splitter plate upstream of the cylinder changed the flow dynamics of the downstream cylinder in the formation region. The splitter plate downstream of the cylinder may restrict the mutual interaction between the upper and lower shear layer, resulting in an elongated recirculation region, a larger vortex formation length, a weaker wake vortex strength, and a reduced drag coefficient, which have been found by Hwang et al. (Reference Hwang, Yang and Sun2003), Akilli et al. (Reference Akilli, Sahin and Tumen2005), Serson et al. (Reference Serson, Meneghini, Carmo, Volpe and Gioria2014), among others. Hwang and Yang (Reference Hwang and Yang2007) indicated that the upstream splitter plate reduced the stagnation pressure, while the downstream one increased the base pressure by suppressing vortex shedding. Thus, the combined effect with both upstream and downstream plates caused a significant drag reduction on the cylinder. Note that the splitter plate could be either rigid or flexible.

(a) A splitter plate placed at upstream position;

(b) a splitter plate placed at downstream position; and

(c) two splitter plates placed at both upstream and downstream positions.

Figure 1.10 Schematic of a circular cylinder with splitter plates.

1.3.9 Polymer

Polymer is a transitional passive control technique for turbulent drag reduction. When the polymer additives are dropped in the fluids, the turbulent flow in the near-wall region might force the polymer to roll up chains, which are stretched in the mean flow direction, as shown in Figure 1.11. In this case, polymer chains exhibit characteristic length scales associated with the turbulent structures, which is related to drag reduction. It was found that minute concentrations of polymers could reduce the drag in turbulent flows by up to 80% (Procaccia et al. Reference Procaccia, L’vov and Benzi2008). Thus, the study of drag reduction by polymer is important for potential applications in engineering, such as oil transportation, heating and cooling systems, etc.

(a) and its response to stretching by turbulent motions

(b) (Jovanović et al. Reference Jovanović, Pashtrapanska, Frohnapfel, Durst and Koskinen2006).

Reproduced with permission of The American Society of Mechanical Engineers.

Figure 1.11 Schematic of polymer in solution at equilibrium

1.3.10 Biomimetic Techniques

Some specific features on organisms are shown in Figure 1.12. They might be related to high aerodynamic performance, and some biomimetic flow control techniques are proposed accordingly.

(a) Wing tip of an albatross.

(b) Self-adaptive hairy flaps of a falcon (Brücker and Weidner Reference Brücker and Weidner2014).

Reproduced with permission, copyright © 2014 Elsevier Ltd. All rights reserved.

(c) Combed serrations on the leading edge of the owl feather (Bachmann and Wagner Reference Bachmann and Wagner2011).

Reproduced with permission from John Wiley and Sons

(d) Corrugation of a dragonfly forewing (Kesel Reference Kesel2000).

Reproduced with permission, copyright © The Company of Biologists Limited 2000

(e) Air bubbles releasing from a penguin.

(f) Compliant skin of a dolphin (Allen and Bridges Reference Allen and Bridges2003).

Reproduced with permission, © John Wiley and Sons, 2003

(g) Tubercles on the leading edge of a humpback whale.

(h) Riblet on sharkskin (Liu and Li Reference Liu and Li2012).

Reproduced with permission from Elsevier

(i) Riblets on a saguaro (Talley et al. Reference Talley, Iaccarino, Mungal and Mansour2001)

Reproduced with permission

(j) Water droplet sitting on the lotus leaf (upper) and micrographs of lotus leaf surface to show the microstructures (Bhushan and Jung Reference Bhushan and Jung2011).

Reproduced with permission, copyright © 2011 Elsevier Ltd. All rights reserved.

Figure 1.12 Some specific features on organisms that might be related to high aerodynamic performance.

One can observe the deflection of the wing tip of a bird during its cruise stage (Figure 1.12(a)) and the raising of a bird’s feathers during its landing (Figure 1.12(b)), which are considered to be the reason why birds can improve their aerodynamic performance. It has been found that the wing tip could reduce the lift-induced drag of the aircraft, which has been widely used in modern airplanes as a well-developed flow control technique. On the other hand, researchers have proved the effects of hairy coating on bluff bodies and airfoils, confirming the function of raising feathers during bird landing.

The owl has a special structure of combed serrations at the leading edge of its feathers (Figure 1.12(c)), which are not observed in most other birds. The combed serrations are found to be of importance to reduce noise by functioning as a vortex generator to attach the flow on the wing suction surface. Thus, the owl has a great ability to fly silently. The leading-edge serration could be used as an effective control approach for noise reduction, as has been indicated by Ito (Reference Ito2009) and Narayanan et al. (Reference Narayanan, Chaitanya, Haeri, Joseph, Kim and Polacsek2015), and reviewed by Choi et al. (Reference Choi, Park, Sagong and Lee2012).

Insect wings are usually not smooth but corrugated, as shown in Figure 1.12(d). The corrugation allows the insect wing to have the advantages of low mass, high stiffness, and low membrane stress (Meng and Sun Reference Meng. and Sun2013). Some researchers, such as Hu and Tamai (Reference Hu and Tamai2008) and Barnes and Visbal (Reference Barnes and Visbal2013), have pointed out that a corrugation airfoil could have a larger lift coefficient at post stall angles of attack. Thus, the corrugation configuration could be used as an approach to improve the aerodynamic performance of micro air vehicles (MAVs).

A penguin may release air bubbles to reduce drag and thus promotes swimming speed (Figure 1.12(e)). Many researches, such as Xu et al. (Reference Xu, Maxey and Karniadakis2002), Elbing et al. (Reference Elbing, Winkel, Lay, Ceccio, Dowling and Perlin2008), and Jacob et al. (Reference Jacob, Olivieri, Miozzi, Campana and Piva2010), have indicated that the air bubbles injected to the boundary layer could lead to a reduction in the friction drag. Thus, the air bubbles show great potential application for drag reduction in underwater vehicles, such as submarines.

It is found that the skin of a dolphin is compliant (Figure 1.12(f)), which is suggested to be effective in reducing the drag of the dolphin when swimming. Accordingly, a flow control technique, namely compliant coating, has been proposed. There have been many researches to show that compliant coating is capable of substantially delaying laminar-to-turbulence transition and reducing turbulent activities in the turbulent boundary layer. For reviews of this field please refer to Riley et al. (Reference Riley, Gad-El-Hak and Metcalfe1988) and Gad-el-Hak (Reference Gad-el-Hak2002).

The leading edge of the flippers of humpback whales is also not smooth and has bumpy tubercles (Figure 1.12(g)). Experimental and numerical studies have suggested that the leading-edge tubercles are beneficial for aerodynamic performance at high angles of attack. A recent review on this field has been made by Aftab et al. (Reference Aftab, Razak, Rafie and Ahmad2016).

From a close-view, we can see that the sharkskin is comprised of small riblet structures (Figure 1.12(h)), which might be related to the drag reduction of shark swimming. Thus, riblets have been used as a useful drag reduction technique in the past, such as in the so-called sharkskin swimsuit.

Similarly, riblets also occur on one kind of saguaro with typical diameter of the order of 0.5 m and height of 10 m (Figure 1.12(i)). It is suggested that riblets are related to suppression of vortex-induced-vibration, which is beneficial for the saguaro in living up to 150 years of age with high wind velocity in their natural habitat.

It is well known that a water drop has a high static contact angle on the lotus leaf, as shown in Figure 1.12(j). The micrograph shows that the lotus leaf surface is very rough with microstructures, which are formed by papillose epidermal cells covered with epicuticular wax tubules. Such features enable the lotus leaf to self-clean and have low adhesion. Thus, the control idea of superhydrophobic surface is developed, where the artificial surface can be made by microstructures. It has been shown that the superhydrophobic surface could lead to a significant reduction in friction drag for both laminar (such as Busse et al. Reference Busse, Sandham, McHale and Newton2013) and turbulent (such as Daniello et al. Reference Daniello, Waterhouse and Rothstein2009) flows. More details to show the effects can be found in the review papers by Rothstein (Reference Rothstein2010) and Bhushan and Jung (Reference Bhushan and Jung2011).

1.4.1 Oscillation and Flow Perturbation

Oscillation and perturbation in flow are two control approaches that may result in similar effects, that are usually imposed on bluff bodies and airfoils. The oscillation is achieved by forcing the controlled body to move in the streamwise (Figure 1.13(a)) or vertical (Figure 1.13(b)) direction. Similarly, the perturbations imposed in the flow could also be in the streamwise (Figure 1.14(a)) or vertical (Figure 1.14(b)) direction. Some primary conclusions have been made for the flow around an oscillating cylinder or the perturbed flow around a stationary cylinder, as has been summarized by Feng and Wang (Reference Feng and Wang2010, Reference Feng and Wang2014).

(a) and vertical

(b) directions.

Figure 1.13 Schematic of a circular cylinder oscillating in the streamwise

(a) and vertical

(b) directions.

Figure 1.14 Schematic of flow perturbation in the streamwise

First, vortex-synchronization could be induced by both methods, where the vortex shedding frequency is closely related to the excitation frequency. It could be categorized into two groups: vortex synchronization at the excitation frequency and at the subharmonic excitation frequency. Vertical oscillation and vertical flow perturbation may result in vortex synchronization at the excitation frequency, while streamwise oscillation and streamwise flow perturbation usually result in vortex synchronization at the subharmonic excitation frequency.

Second, the vortex-synchronization regime is related to both the excitation frequency and amplitude. The fundamental trend is that a higher excitation amplitude can induce a much larger vortex-synchronization regime. The vortex-synchronization regime induced by vertical oscillation and vertical flow perturbation is usually around the natural frequency, while that for the streamwise cases is usually around twice of the natural frequency.

Thirdly, the near wake dynamics could be modified by oscillation and flow perturbation. In particular, the symmetric vortex shedding mode could be induced by symmetric excitation, such as streamwise oscillation and streamwise flow perturbation.

Similar to the oscillation approach, rotation is another control method for the bluff body, as has been studied by Navrose et al. (Reference Navrose, Meena and Mittal2015). The oscillation method can also be used with a flat plate to control the boundary layer, where the drag could be reduced by up to 45%, as indicated by Choi (Reference Choi2002). Similarly, flexible wall oscillation is another approach for boundary layer control, as conducted by Shen et al. (Reference Shen, Zhang, Yue and Triantafyllou2003).

1.4.2 Acoustic Excitation

Acoustic excitation is a control method based on a sound wave, which could be achieved by a loudspeaker, as shown in Figure 1.15. The basis for such control is the boundary layer receptivity. The acoustic pressure wave interacts with the boundary layer/shear layer and triggers the instability which results in global variations of the flow field. There have been studies to show that the acoustic excitation could increase both the lift and lift-to-drag ratio of the airfoil in the close and post stall regions (such as Yarusevych et al. Reference Yarusevych, Sullivan and Kawall2007; Yang and Spedding Reference Yang and Spedding2013). In addition, the effects of acoustic excitation on the flow around circular cylinders (such as Mohany and Ziada Reference Mohany and Ziada2009) and on the development of vortical structures (such as Leblanc Reference Leblanc2001) and jets (such as Huang et al. Reference Huang, Jufar and Hsu2013) have also been found. The sound press level, excitation frequency, as well as the Reynolds number, determine the effects. However, the experimental conditions, such as the test-section geometry and acoustic resonance, are also of great importance in achieving the ideal effects.

Figure 1.15 Schematic of flow around an airfoil under acoustic excitation.

1.4.3 Jet

The jet, also known as free jet, steady jet or continuous jet, is one of the earliest techniques used for boundary layer flow control. When fluids are issuing from an orifice, flow separates from the edges to form a separated shear layer, which gradually rolls up into vortical structures, as shown in Figure 1.16. The vortices are undergoing flow transition as convecting downstream, resulting in the turbulent flow. The jet can enhance momentum mixing between inner and outer boundary layers, which is beneficial for separation delay. Similarly, suction control is another approach by drawing the low momentum fluids away from the near-wall region. Both blowing and suction can be used for flow control in various fields.

Figure 1.16

(a) Flow pattern of a jet at Re_j = 1620.

(b) Schematic of a jet issuing from a nozzle.

(Todde et al. Reference Todde, Spazzini and Sandberg2009), reproduced with permission, copyright © Springer 2009

1.4.4 Synthetic Jet

The synthetic jet has been widely investigated since the 1990s. A typical synthetic jet actuator includes a cavity, an orifice and an oscillating device, such as a piezoceramic diaphragm or piston, as shown in Figure 1.17. An excitation signal drives the diaphragm or piston to move forward to and backward from the orifice, and thus the fluids are ejected from the orifice and sucked into the cavity, periodically. During the blowing cycle, the separated shear layer formed from the orifice edge grows and develops into vortical structures that are convecting downstream gradually due to the self-induced velocity. During the suction cycle, the vortical structures formed during the previous blowing cycles have moved further downstream and they are hardly influenced by the suction process. Thus, a series of vortical structures are formed periodically during the blowing and suction processes, which could enhance momentum mixing efficiency. The synthetic jet is also called the zero-net-mass-flux jet, as the net flow flux during one period is zero though the net momentum flux is not zero. The synthetic jet has become one of the most important active flow control techniques, and it has shown significant effects in various fields, as have been reviewed by Glezer and Amitay (Reference Glezer and Amitay2002), Luo and Xia (Reference Luo and Xia2005), and Zhang et al. (Reference Zhang, Wang and Feng2008).

Figure 1.17 Schematic of a synthetic jet actuator.

1.4.5 Plasma Actuator

Wide investigation of the plasma actuator for flow control started in the 1990s. The dielectric barrier discharge (DBD) plasma actuator is representative, though several other types have been developed, such as corona discharge actuator and plasma spark-jet actuator. A typical DBD plasma actuator consists of an exposed electrode and an embedded electrode, separated by a dielectric sheet, as shown in Figure 1.18. When the electrodes are supplied with high voltage and frequency, the air over the embedded electrode is ionized and thus a wall jet forms. Accompanying the wall jet, the flow separates from the wall and rolls up into a vortex. These are the two main features of the DBD plasma actuator: inducing a wall jet and a starting vortex. The advantages of plasma control include electronic design with no moving parts, extremely fast response and real-time control ability, low power consumption, low mass, and simple to use, making the plasma actuator one of the most popoular techniques for flow control applications. The main developments can also be found in the review papers by Moreau (Reference Moreau2007), Corke et al. (Reference Corke, Enloe and Wilkinson2010), Wang et al. (Reference Wang, Choi, Feng, Jukes and Whalley2013a), Wu and Li (Reference Wu and Li2015).

Figure 1.18 Schematic of a DBD plasma actuator.

1.4.6 Lorentz Force

The subject of flow control with Lorentz force (or electromagnetic body force) originates from electromagnetism. When a charged particle is moving in the presence of an electric field and a magnetic field, it will experience a Lorentz force. The Lorentz force required for flow control can be created by placing electrodes and magnets side by side, parallel to one another, as shown in Figure 1.19. The direction of the Lorentz force can be changed by varying the direction of either electric field or magnetic field, and its magnitude can be adjusted by varying the strength of either electric or magnetic field. Thus, the Lorentz force could be produced in a sinusoidal waveform and in either spanwise or streamwise direction relative to the free stream. The spanwise Lorentz force could reduce the friction drag of the turbulent boundary layer, though the streamwise Lorentz force may increase it. Moreover, the Lorentz force could reduce flow separation and thus increase the lift coefficient while reducing the drag coefficient of airfoils.

Figure 1.19 Schematic of an arrangement for the creation of the Lorentz force (Adapted from Berger et al. Reference Berger, Kim, Lee and Lim2000).

Reproduced with permission from AIP Publishing.