Turbulent boundary layer separation control using plasma actuator at Reynolds number 2000000

Zhang Xin,Huang Yong,Wang Xunnian,Wang Wano,Tang Kun,Li Huaxing

aSchool of Aeronautics,Northwestern Polytechnical University,Xi'an 710072,China

bChina Aerodynamics Research and Development Center,Mianyang 621000,China

Turbulent boundary layer separation control using plasma actuator at Reynolds number 2000000

Zhang Xina,*,Huang Yongb,Wang Xunnianb,Wang Wanbob,Tang Kunb,Li Huaxingb

aSchool of Aeronautics,Northwestern Polytechnical University,Xi'an 710072,China

bChina Aerodynamics Research and Development Center,Mianyang 621000,China

An experimental investigation was conducted to evaluate the effect of symmetrical plasma actuators on turbulent boundary layer separation control at high Reynolds number.Compared with the traditional control method of plasma actuator,the whole test model was made of aluminum and acted as a covered electrode of the symmetrical plasma actuator.The experimental study of plasma actuators' effect on surrounding air,a canonical zero-pressure gradient turbulent boundary,was carried out using particle image velocimetry (PIV) and laser Doppler velocimetry (LDV) in the 0.75 mX0.75 m low speed wind tunnel to reveal the symmetrical plasma actuator characterization in an external flow.A half model of wing-body configuration was experimentally investigated in the Ø 3.2 m low speed wind tunnel with a six-component strain gauge balance and PIV.The results show that the turbulent boundary layer separation of wing can be obviously suppressed and the maximum lift coefficient is improved at high Reynolds number with the symmetrical plasma actuator.It turns out that the maximum lift coefficient increased by approximately 8.98%and the stall angle of attack was delayed by approximately 2°at Reynolds number 2X106.The effective mechanism for the turbulent separation control by the symmetrical plasma actuators is to induce the vortex near the wing surface which could create the relatively largescale disturbance and promote momentum mixing between low speed flow and main flow regions.©2016 Chinese Society of Aeronautics and Astronautics.Production and hosting by Elsevier Ltd.This is an open access article under the CC BY-NC-ND license(http://creativecommons.org/licenses/by-nc-nd/4.0/).

1.Introduction

Currently,active flow control (AFC) technology is a popular method that can improve the aircraft aerodynamics performance without added drag and also liberate designers from the restriction of traditional aerodynamic design.

Dielectric barrier discharge (DBD) plasma flow control method has been extensively proved to be a promising AFC technology1in the field of separation control for wing,2,3lift enhancement4,5and separation flow control for airfoil,6boundary layer control,7–10separation control for lowpressure turbine blades,11control of bluff body,12–14control of compressor cascade,15,16control of broadband noise,17and control of jet diffusers.18Compared with the traditional AFC technology,the benefits of plasma actuator are obvious,such as simple structure without moving parts and rapid response.Remarkable reviews on plasma flow control technology have been published lately.19–21

Separation control of wing using plasma actuators at lowmiddle Reynolds number was investigated by numerous researchers with experimental and computational methods.22–32By numerical simulation method,Asada and Fujii made an investigation on the laminar separation control by a plasma actuator around NACA 0015 at Reynolds number 63000 recently.33They indicated that the plasma actuation can promote the transition to turbulence at the laminar separation shear layer more effectively and the n the turbulent mixing led to early reattachment of the separated flow.The effect of the location and operating conditions of the plasma actuator was widely studied by Sato et al.34It was found that the most effective location is just upstream of the natural separation point.The deep stall flow control was investigated by Aono et al.at Reynolds number 260000.35They carried out an optimization study of the burst frequency of the plasma actuator and proved the applicability of the plasma actuator for control of the laminar separated flow at middle Reynolds number.It was suggested that the optimum burst frequency becomes higher than that at low Reynolds number.From these investigations,the promotion of the turbulent transition at the shearlayer is the key for effective laminar-separation control with the plasma actuator at low-middle Reynolds number.35

However,some researchers could question the applications of the plasma actuator.Because the Reynolds number of real airplane is usually over 106that could be beyond the plasma authority from the literature.In order to meet the requests of engineering application,the plasma authority must be improved at higher Reynolds number.In this high Reynolds number zone,flow is often separated in turbulence but not in laminar.Therefore,the separation characteristics are different from the laminar separation.It is not clear whether the same tactic for the laminar separation control at low-middle Reynolds number could be suitable for the turbulence separation control.

The purpose of this investigation was to study the turbulent boundary layer separation control over a wing-body configuration using a symmetrical plasma actuator at Reynolds number 2X106.The integral control effect of the symmetrical plasma actuator is described by the force measurements and particle image velocimetry (PIV) results.Moreover,this paper highlights how the induced airflow by the plasma actuator interacts with the boundary layer with laser Doppler velocimetry (LDV) technology and smoke visualization,and discusses the controlling mechanism of the plasma actuator.

2.DBD plasma actuator

Fig.1 Electrodes configuration of DBD plasma actuators.

Classical DBD plasma actuator comprises an upper electrode and a lower electrode that are separated by insulating material.One of the electrodes is exposed to the ambient air and the other is completely encapsulated by dielectric material(Fig.1(a)).When a high voltage power supply is applied to the two electrodes,ionized air(plasma)at the edge of the exposed electrode is produced.The plasma covers the region projected by the covered electrode.As a result of the electric field gradient,the plasma produces a body force36–38that acts on the surrounding air and induces the airflow in the direction from the upper electrode towards the lower electrode.39,40The process of ionizing the air at this configuration is referred to as DBD.Fig.2(a)presents the plasma discharge image for the ordinary plasma actuator.It can be seen that a single bluish line in the darkness is close to the edge of the exposed electrode.

Compared with the typical one,this work used the testing model made of met al and copper foil tape as a lower electrode and an upper electrode respectively(Fig.1(b)).Therefore,there are two bright lines on both sides of the exposed electrode for symmetrical actuator(Fig.2(b)).These discharge photographs are consistent with DBD theory as mentioned above that the glow usually occurs at the edge of exposed electrode and spreads to the projected area of covered electrode.Since the lower electrode covers the whole upper electrode,the symmetrical actuator could produce the double bluish lines.The induced velocity field of the plasma actuator at the test arrangement will be discussed in more details later in this paper.

3.Plasma actuator characterization

3.1.Experimental setup

(1)Wind tunnel

The experiments were carried out in an open-circuit low speed wind tunnel with a test section 1.05 mX0.75 mX 0.75 m at China Aerodynamics Research and Development Center (CARDC),which is capable of generating a maximum wind speed of 55 m/s with the turbulence intensity level of less than 0.3%.It is a draw down wind tunnel driven by a centrifugal fan.

Fig.2 Photography of plasma actuators in action.

(2)Testing model

A smooth plate across the wind tunnel extended 0.15 m above the lower ground.The flat plate was made of aluminum with super-elliptic leading edge and sharp trailing edge which can be adjusted to change the streamwise pressure gradient(Fig.3,where U∞is the incoming flow velocity).It is 1 m in length and 0.1 m in thickness.In order to make sure that a thick boundary is turbulent boundary at the measurement point,the wind speed kept constant at 5 m/s.At the transition of the plate,the boundary layer was affected by a 50 mm wipe trip of distributed sand grain roughness.

Experiments were first carried out to depict the turbulent boundary layer flow over the plate.Boundary layer measurements on the plate were implemented using constant temperature hot wire anemometry.The measurement position was at xp=837 mm,where xpis the streamwise distance along the boundary development plate as measured from the boundary layer trip.Fig.4 shows the mean velocity profiles of boundary layer,where u is the streamwise velocity of boundary layer,δ is the boundary layer thickness.The boundary layer mean velocity profiles u/U∞are well consistent with the typical zero pressure gradient turbulent boundary layer profile41(Fig.4).

(3)Plasma actuator and high voltage power supply

Fig.3 Schematic of turbulent boundary layer development plate.

Fig.4 Mean velocity profiles of boundary layer.

Fig.5 Schematic of experimental set-up for PIV measurements.

The plasma actuator was arranged such that the downstream edge of the exposed electrode was 762 mm from the boundary layer trip(Fig.3).The plate model as the covered electrode(cathode)was connected to the ground.The insulating material which was mounted on the plate was 0.05 mm thick.The exposed electrode(anode)was made of a 2 mm wide copper foil tape which was 280 mm in length and 0.05 mm in thickness.

The plasma actuator was excited by an alternating current power supply.The frequency of the power is from 0.1 to 6 kHz and the voltage amplitude ranges from 0 to 10 kV.

(4)PIV systems

A laser system was positioned on top of the test section and provided a light sheet to the center plane of the actuator mounted on the flat plate(Fig.5).The laser produced double pulses with a 25 μm interval at a maximum repetition rate of 15 Hz.PIV images were captured by a PIV CAM 10–30 digital camera which was set to view an area of 70 mmX40 mm.The air inside was seeded using olive oil droplets of nominally 1 μm diameter produced by a TSI atomizer.Data processing was performed with TSI Insight 6 software,using a crosscorrelation algorithm to generate vectors over a 20X20 pixel interrogation area with 50%overlap to an accuracy of 3%–5%.

(5)LDV system

Besides that,the induced wall normal velocity profiles downstream of the actuator were obtained using a Dantec Dynamics Fiber Flow LDV system.The fiber optic LDV system was performed in 180°backscatter mode.A BSA F60 Flow Processor and BSA Flow Software Version were used to measure the Doppler bursts.In order to obtain the data near the wall,only the horizontal to the component of the 2D 60 mm probe was used.To get the induced velocity profiles,the fiber optic probe was traversed normal to the surface of the plate.

3.2.Experimental results

(1)Plasma actuator characterization in quiescent air

The induced velocity field by the plasma actuator in quiescent air was acquired using two-component PIV.Fig.6 shows time-averaged velocity fields above the actuator which was applied AC voltage U=8 kVp-pand a frequency f=3 kHz in still air.Upand Vpare the induced velocity components by plasma actuator in the x direction and y direction,respectively.The results show that the plasma actuator attracts surrounding airflow towards the wall and the n emanates this induced airflow tangentially away from the exposed electrode.The DBD plasma induced wall jet is different from a classic laminar wall jet because there is no mass added to the flow.39In this paper,the insulating electrode was the plate and the exposed electrode was aligned in the spanwise direction.Therefore,the plasma actuator induced a bi-directional wall-jet tangential to the dielectric surface away from the exposed electrodes.The maximum velocity is around 3.9 m/s.

The mean velocity profiles whose location was at 20 mm downstream of the actuator by LDV technology are shown in Fig.7,Upmaxis the maximum velocity induced by plasma actuator in the x direction.The induced jet was shown ejecting away from the edge of the exposed electrode.Meanwhile there was a strong normal velocity component above the plasma actuator,because the air was drawn towards the plasma actuator and the n was accelerated by the body force.The maximum speed of induced airflow increases with the increasing applied voltage.

(2)Plasma actuation with external flow

The effect of plasma actuator on the boundary was studied by a great deal of researchers in the openly published papers.But,the investigation on the dynamic process of the mutual effect between induced airflow by plasma actuator and boundary layer is relatively less.This section studies this problem by the PIV and the LDV.The incoming flow speed was 5 m/s and the other testing condition was the same as the experiments of plasma actuator characterization in still air.The plasma actuator was synchronized to the CCD camera.

Fig.8(a)shows instantaneous velocity field around the actuator.It can be seen from Fig.8 that the momentum from downstream induced jet adds to the boundary layer and the upstream induced jet is against boundary layer flow.Then the upstream induced jet interacts with boundary layer flow and wraps around to induce the vortex as in Fig.8(b).

Fig.6 Time-averaged velocity field above the actuator in still air.

Fig.7 Velocity profiles above plasma actuator.

Fig.8 Instantaneous velocity field above actuator.

The vortex was very stable when the plasma was actuated.The flow field can be considered as a steady-state phenomenon.When the plasma turned off,the vortex rapidly shrunk to the wall,taking less than 0.1 s to disappear entirely.

The induced vortex is a highlight in the induced airflow field by the plasma actuator.Compared to the ordinary plasma actuator,the symmetrical plasma actuator is both jet actuator and vortex actuator.It can produce relatively large-scale perturbations and facilitate momentum exchange between boundary layer and main flow zone.4

Fig.9 shows the turbulent boundary layer velocity profiles which were at 75 mm downstream of the plasma actuator.The results about the baseline flow with the LDV system are in agreement with hot wire measurements shown previously in Fig.4.The results show that momentum in the boundary layer increase through the downstream induced jet.Mean-while,the mean velocity profile due to plasma actuator is still developing at 75 mm downstream of the exposed electrode.

Fig.9 Boundary layer velocity profiles.

The corresponding streamwise component turbulence intensity profile is shown in Fig.10.Up′is the fluctuating velocity induced by plasma actuator in the x direction.The results suggest that the turbulence intensity increases near the boundary layer.That is indicative of increased mixing of momentum because of the induced vortex.This result agrees well with previous PIV results shown in Fig.8(b).

4.Separation control

In this section,the turbulent boundary layer separation control over the wing-body configuration with the symmetrical plasma actuator was studied.Force measurements,PIV and tufts flow visualization were used to characterize the effect of actuation.

4.1.Experimental setup

(1)Wind tunnel

Fig.10 Turbulence intensity profiles.

Fig.11 Schematic of wing-body configuration model in test section.

The experimental investigation on wing-body configuration model was carried out in the Ø 3.2 m low speed wind tunnel located in the CARDC.It is a single-return continuous tunnel with the open test section.The test section is 5 m in length and a round cross-section with 3.2 m diameter.The wind velocity range is 10 m/s to 115 m/s.

(2)Experimental model and support equipment

In order to improve the Reynolds number,a half model with a swept was adopted.The wing has a 25°swept leading edge with a chord length of 510 mm and a spanwise length of 1890 mm.The airfoil profile shape of wing was the SC(2)-0714 supercritical airfoil.The test model was made of aluminum.Fig.11 shows the side view of the test model.Lift and drag were measured using a six-component balance that was mounted under the ground.The model was installed on the force balance which was supported by the supporting bracket.The angular position of the test model was driven by the computer controlled turntable.

(3)Plasma actuator

Fig.12 shows the wing-body con figuration model in the wind tunnel.The total wing was encapsulated by the kapton film and the copper foil tape as upper electrode was placed at 1%of the chord from the leading edge of the wing.The copper foil electrode was 5 mm in width and had rounded corners.It did not span the whole length of the supercritical wing,terminating at 30 mm from the wingtip to avoid point discharges.The power supply was the same as that in the plasma actuator characterization experiments.

4.2.Experimental results

(1)Force measurement results

Firstly,the force measurement results were used to evaluate the macro effect of the symmetrical plasma actuator on the separation control.This paragraph shows the force measurement results with different Reynolds numbers.These results centred on separation flow control for a series of angles of attack including αs,the stall angle of attack.The freestream flow speed ranged from 15 m/s to 60 m/s and corresponding Reynolds number of the wing varied from 0.5X106to 2X106based on the mean aerodynamic chord length.The plasma actuator was located at 1%of the chord from the leading edge of the wing and was applied AC voltage of 8 kVp-pwith the frequency of 3 kHz.

Fig.12 Photograph of wing-body configuration model in wind tunnel.

Fig.13 Mean lift and drag coefficients as a function of angle of attack.

The mean lift and drag coefficients CLand CDas a function of angle of attack for the steady plasma actuator at different Reynolds numbers are shown in Fig.13.The corresponding values for the baseline wing are also included for reference in each data graph.This section focuses on the capacity of the plasma actuator which controls flow separation at post stall angles of attack.Therefore,these pictures only show angles of attack for αs-4°≤ α ≤ αs+8°.

Fig.14 Maximum lift coefficient and stall angle of attack vs Reynolds number.

Fig.15 Maximum difference of lift coefficient and angle of attack.

Without actuation,the separation flow occurs in natural post stall conditions,which can be verified by the rapid increase in drag and decrease in lift.It is also found that the plasma actuator can produce dramatically lift enhancement and drag reduction in natural post stall conditions,indicating that the separated flow could reattach on the surface.

Fig.16 Convergence plot for lift coefficient at angle of attack 16°(Re=2.0X106).

Fig.17 Velocity distribution around wing without and with actuation at angle of attack 16°(Re=2.0X106).

The control effect of plasma actuator was appraised on the basis of the maximum total lift coefficient as well as the maximum difference between the lift coefficient with the plasma actuator and that of the nature flow.The former is usually the standard of value for a virtual slat.The latter is a potential valuefor flight control at high angles of attack.The difference between CLmax(on)and CLmax(off)is shown in Fig.14(a),where CLmax(on)is the maximum lift coefficient induced by the plasma actuator and CLmax(off)is the maximum lift coefficient without control.Their respective αCLmax(on)values are plotted in Fig.14(b),where the baseline stall angle of attack αCLmax(off)has been subtracted off to represent the stall angle of attack where the maximum lift happened.

Fig.18 Time-averaged distributions of TKE without or with actuation at angle of attack 16°(Re=2.0X106).

Fig.19 Photograph of supercritical wing with and without fluttering tufts.

Fig.14(a)shows that the plasma actuator brings in a very comparable maximum lift enhancement.The maximum lift coefficient has an increase of 8.98%,up to Reynolds number 2X106.For these,the maximum lift occurred at a 2°larger angle of attack than the baseline stall angle of attack(Fig.14(b)).Meanwhile,the results indicate that with the increase of the Reynolds number,the increased maximum lift remains above 7.4%and the delayed stall angle of attack is kept at a steady value of 2°.

Fig.20 Time evolution of flow field around wing with plasma actuator on(α =12°).

Fig.21 Momentum coefficient of plasma jet vs incoming flow velocity.

The other evaluation of the plasma actuator performance is presented in Fig.15.Fig.15 shows the maximum change in the lift coefficient compared to the baseline and the angle of attack where this happens.The maximum change in the lift coefficient,(ΔCL)max,generally occurs at larger nature stall angle of attack because of the rapid drop of the nature lift compared to that with the plasma actuator.This is clear by the large values of α(△CL)maxthat increase with increasing Reynolds number.

At 6°post angle of attack,the lift increased by approximately 42%using plasma actuator at Reynolds number 2X106.This is meaningful if the plasma flow control technology could be applied to flight control during high angle of attack flight.

Overall,the force measurement results indicate that the control ability of the symmetrical plasma actuator does not decrease with Reynolds numbers and thus the aerodynamic performance is improved at high Reynolds numbers.It seems that the influence of Reynolds number on the plasma actuator authority is different from the typical plasma actuator.36

To demonstrate the result's validity,the error analysis was conducted by repeated experiments.Fig.16 presents the lift coefficient versus the number of data samples with or without actuation at angle of attack 16°.With the actuation,the data is stable.However,because the flow around the surface of the wing is instable after stalling angle of attack,the data is fluctuating.But the scope of data vibration is much less than the lift augment and is approximately 0.004.

(2)PIV results

Besides that,the PIV experiments were employed to confirm the force measurement data.Considering the paper length,we only include a few PIV results here.Fig.17 presents the time-averaged velocity field around the wing without or with actuation.Without control,the flow separation occurs at the leading edge of the wing(Fig.17(a)).With the actuators turned on,there is obviously a remarkable effect on the global flow structure.The plasma actuators substantially reduce the extent of the separation region,and result in a valid lift increase(Fig.17(b)).The PIV results are in good agreement with the force measurement results.

As mentioned above (Fig.8(b)),the induced vortex generated by the plasma actuator can energetically interact with the turbulent boundary flow;hence it is worth analyzing the turbulent kinetic energy (TKE),as defined by TKE=0.5XHere,U′is the fluctuating velocity components in the x direction,and V′is the fluctuating velocity components in the y direction.Fig.18 reveals the distributions of the time-averaged TKE.From the enlarged TKE distribution of the no control case,the high energy region above the wing surface is obvious.It indicates that the velocity fluctuation increased and the turbulent separation happened in this region(Fig.18(a)).Due to the symmetrical plasma actuation,the turbulence intensity increased in the near wall area but reduced throughout most of the outer portion of boundary layer(Fig.18(b)).The symmetrical plasma actuator results in enhancing exchange of momentum in the boundary layer and prevents separation around the wing surface.

(3)Fluorescence tufts results

Another proof that the separation can be suppressed by the plasma actuator comes from fluorescence tufts results.The fluorescent tufts were used to visualize flow reattachment.Small tufts were pasted slightly on the surface of the wing(Fig.17),corresponding to the experimental conditions of PIV experiments.Fig.19(a)shows a photograph of the supercritical wing with fluttering tufts,indicating the reverse flow condition,and Fig.19(b)shows smooth(non-fluttering)tufts,which suggest the flow separation is reattached by plasma actuator.

5.Controlling mechanism

From the experimental results of plasma actuator characterization(Fig.8(b)),it is a clear indication that the symmetrical plasma actuator has two mechanical phenomena(wall jet and induced vortex)which can affect boundary layer and modify the flow filed near the surface of testing model,leading to the control of the separation.First,smoke visualization experiments were performed by the continuous laser sheet to verify the PIV results and obtain the detail of the flow field around the plasma actuator.Fig.20 shows a series of images of the flow field above the actuator.The plasma actuator was driven by a sinusoidal signal with frequency of 3 kHz and voltage amplitude of 8 kVp-p.The flow speed was 2 m/s.The sampling frequency of the high-speed camera was 500 frames per second.The plasma jet coexists with the induced vortex at some point (Fig.20).Meanwhile,it can be found through Figs.20(a)–20(e)that the induced vortex rolls up to form a coherent structure.With time,the induced vortex grew and created relatively large-scale disturbances.Then,the vortex could bring the high momentum from the main stream flow and enable the flow to withstand the adverse pressure gradient without separating.

To figure out the question which is more important for controlling the separation between wall jet and induced vortex,this paper introduces the momentum coefficient of plasma jet,as defined byHere, ρ is the local air density,and Mpis the momentum flux by plasma actuator,whereA is the total area of each PIV image.Fig.21 presents the relationship between Cμand incoming flow velocity.Cμdecreases with the increase of the incoming flow velocity.But the force measurement results suggest that the control authority of the symmetrical plasma actuator did not decrease with Reynolds numbers(Figs.14 and 15).It is speculated that the induced vortex may play an important role in controlling the separation flow.

6.Conclusions

The results of the plasma flow control investigations shown in this paper demonstrate the control effect of symmetrical plasma actuators on turbulent boundary layer separation control at high Reynolds number.Compared with the typical DBD plasma configuration,the covered electrode is the whole test model.Therefore,the actuator has two induced airflow in reverse directions on each edge of the exposed electrode.

An experimental investigation of flow control by symmetrical plasma actuator has been carried out in a turbulent boundary layer flow around a plate at a freestream velocity of 5 m/s.The PIV results reveal that in addition to being a jet actuator,the plasma actuator at the test configuration is also a vortex actuator.By analyzing the variation of momentum coefficient with incoming flow speed,we can see that the induced vortex exerted a remarkable effect on separation flow control.It created relatively large-scale disturbances and promoted momentum mixing between low speed and main flow regions.The LDV results show that the actuation led to increased streamwise component turbulence intensity in boundary layer due to the formation of spanwise vortices.

Turbulent boundary layer separation control by the plasma actuator around a wing-body configuration at high Reynolds numbers has also been investigated.A force balance and time-resolved PIV were used to measure the lift and drag coefficients and to study the velocity fields,respectively.The results show that the control authority did not decrease with increasing Reynolds number.As the Reynolds number increased,the increased maximum lift remains above 7.4%and the delayed stall angle is kept at a steady value of 2°.Up to Reynolds number 2X106,the maximum lift increased by 8.98%and the maximum change in the lift coefficient increased by 42%at a 6°post stall angle of attack.Therefore,the symmetrical plasma actuators are potential valuefor replacing leading edge slat and controlling separation flow at high angles of attack.Future research will focus on flight experimentation on the aerodynamic enhancement of different wing geometries using plasma actuator at Reynolds numbers typical of aircraft takeoff and landing.

Acknowledgement

This work was supported by the Exploration Foundation of Weapon Systems(No.7130711).

1.Li YH,Wu Y,LiJ.Review of the investigation on plasma flow control in China.Int J Flow Control 2012;4(1):1–18.

2.Sosa R,Artana G,Moreau E,Touchard G.Stall control at high angle of attack with plasma sheet actuators.Exp Fluids 2007;42(1):143–67.

3.Goeksel B,Greenblatt D,Rechenberg I.Steady and unsteady plasma wall jets for separation and circulation control.3nd AIAA flow control conference.Reston:AIAA;2006.

4.Zhang X,Huang Y,Wang WB.Unmanned air vehicle flow separation control using dielectric barrier discharge plasma at high wind speed.Sci China-Phys Mech Astron 2014;57(6):1160–8.

5.Benard N,Moreau E,Griffin J.Slope seeking for autonomous lift improvement by plasma surface discharge.Exp Fluids 2010;48(5):791–808.

6.Little J,Nishihara M,Adamovich I.High-lift airfoil trailing edge separation control using a single dielectric barrier discharge plasma actuator.Exp Fluids 2010;48(3):521–37.

7.Jukes TN,Choi KS,Johnson GA.Turbulent boundary-layer control for drag reduction using surface plasma.2nd AIAA flow control conference.Reston:AIAA;2004.

8.Roupassov D,Zavyalov I,Starikovskii AY.Boundary layer separation plasma control using low-temperature non-equilibrium plasma of gas discharge.44th AIAA aerospace sciences meeting and exhibit.Reston:AIAA;2006.

9.Nie CQ,Li G,Zhu JQ.Investigation of dielectric barrier discharge plasma flow control.Sci China Tech Sci 2008;51(7):1064–72.

10.Zhang PF,Liu AB,Wang JJ.Flow structures in flat plate boundary layer induced by pulsed plasma actuator.Sci China Tech Sci 2010;53(10):2772–82.

11.Huang J,Corke TC,Thomas FO.Plasma actuators for separation control of low-pressure turbine blades.AIAA J 2006;44(1):51–7.

12.Do H,Kim W,Mungal MG.Bluff body flow separation control using surface dielectric barrier discharges.45th AIAA aerospace sciences meeting and exhibit.Reston:AIAA;2007.

13.Jukes TN,Choi KS.Flow control around a circular cylinder using pulsed dielectric barrier discharge surface plasma.Phys Fluids 2009;21(8):4103–15.

14.Sun Q,Cheng BQ,Li YH.Computational and experimental analysis of Mach 2 air flow over a blunt body with plasma aerodynamic actuation.Sci China Tech Sci 2013;56(11):795–802.

15.Li YH,Wu Y,Zhou M.Control of the corner separation in a compressor cascade by steady and unsteady plasma aerodynamic actuation.Exp Fluids 2010;48(6):1015–23.

16.Li G,Xu YJ,Lin JQ.Control of endwall secondary flow in a compressor cascade with dielectric barrier discharge plasma actuation.Sci China Tech Sci 2009;52(12):3715–21.

17.Li Y,Zhang X,Huang X.The use of plasma actuators for bluff body broadband noise control.Exp Fluids 2010;49(2):367–77.

18.Benard N,Bonnet JP,Touchard G,Moreau E.Flow control by dielectric barrier discharge actuators:jet mixing enhancement.AIAA J 2008;46(9):2293–305.

19.Moreau E.Airflow control by non-thermal plasma actuators.J Phys D Appl Phys 2007;40(3):605–36.

20.Corke TC,Post ML,Orlov DM.SDBD plasma enhanced aerodynamics:concepts,optimization and applications.Pro Aerosp Sci 2007;43(7):193–217.

21.Wang JJ,Choi KS,Feng LH,Jukes TN,Whalley RD.Recent developments in DBD plasma flow control.Prog Aerosp Sci 2013;62:52–78.

22.Post ML,Corke TC.Separation control using plasma actuators:dynamic stall vortex control on oscillating airfoil.AIAA J 2006;44(12):3125–35.

23.Rizzetta DP,Visbal MR.Large eddy simulation of plasma-based control strategies for bluff body flow.AIAA J 2009;47(3):717–29.

24.Benard N,Jolibois J,Moreau E.Lift and drag performances of an axisymmetric airfoil controlled by plasma actuator.J Electrostatics 2009;67(3):133–9.

25.Mabe JH,Calkins FT,Wesley B.Single dielectric barrier discharge plasma actuators for improved airfoil performance.J Aircraft 2009;46(3):847–55.

26.Vorobiev AN,Rennie RM,Jumper EJ.Experimental investigation of lift enhancement and roll control using plasma actuators.J Aircraft 2008;45(4):1315–21.

27.Rizzetta DP,Visbal MR.Numerical investigation of plasma-based control for low Reynolds number airfoil flows.AIAA J 2011;49(2):411–25.

28.Vorobiev A,Rennie RM,Jumper EJ.Lift enhancement by plasma actuators at low Reynolds numbers.J Aircraft 2013;50(1):12–9.

29.Greenblatt D,Goksel B,Schule CY.Dielectric barrier discharge flow control at very low flight Reynolds numbers.AIAA J 2008;46(6):1528–41.

30.Greenblatt D,Schuele CY.Mechanism of flow separation control using plasma actuation.Phys Fluids 2012;24:077102.

31.Zhao GY,Li YH,Liang H.Flow separation control on swept wing with nanosecond pulse driven DBD plasma actuators.Chin J Aeronaut 2015;28(2):368–76.

32.Han MH,Li J,Niu ZG.Aerodynamic performance enhancement of a flying wing using nanosecond pulsed DBD plasma actuator.Chin J Aeronaut 2015;28(2):377–84.

33.Asada K,Fujii K.Burst frequency effect of DBD plasma actuator on the control of separated flow over an Airfoil.6th AIAA flow control conference.Reston:AIAA;2012.

34.Sato M,Okada K,Nonomnura T.Massive parametric study by LES on separated- flow control around airfoil using DBD plasma actuator at Reynolds number 63,000.43rd AIAA fluid dynamics conference.Reston:AIAA;2013.

35.Aono H,Okada K,Nonomura T.Effects of burst frequency and momentum coefficient of DBD actuator on control of deep-stall flow around NACA0015 at Reynolds number 260000.52nd aerospace sciences meeting.Reston:AIAA;2014.

36.Wu Y,Li YH,Jia M.Experimental investigation into characteristics of plasma aerodynamic actuation generated by dielectric barrier discharge.Chin J Aeronaut 2010;23(1):39–45.

37.Liu ZF,Wang LZ,Fu S.Study of flow induced by sine wave and saw tooth plasma actuators.Sci China-Phys Mech Astron 2011;54(11):2033–9.

38.Che XK,Nie WS,Hou ZY.High altitude plasma flow control simulation through ground experiment.Acta Aeronaut Astronaut Sin 2015;36(2):441–8[Chinese].

39.Whalley R,Choi KS.Starting,traveling and colliding vortices:dielectric-barrier discharge plasma actuator in quiescent air.Phys Fluids 2010;22:91–105.

40.Che XK,Shao T,Nie WS,Yan P.Numerical simulation on a nanosecond-pulse surface dielectric barrier discharge actuator in near space.J Phys D:Appl Phys 2012;45:145201.

41.Klebanoff PS.Characteristics of turbulence in a boundary layer with zero pressure gradient.Washington,D.C.:NASA;1995.Report No.:NACA TR-1247.

42.Zuo W,Gu YS,Cheng KM,Liu Y.An experimental investigation on separation control of an airfoil by beveled slit synthe tic jet actuator.J Exp Fluid Mech 2014;28(6):45–50[Chinese].

43.Feng LH,Jukes TN,Choi KS.Flow control over a NACA 0012 airfoil using dielectric barrier discharge plasma actuator with a Gurney flap.Exp Fluids 2012;52(6):1533–46.

Zhang Xinis a Ph.D.student at School of Aeronautics,Northwestern Polytechnical University.His main research field is flow separation control using plasma actuator.

Huang Yongis an associate professor of China Aerodynamics Research and Development Center.His current research interests are flow control technology using plasma actuator and blowing and low speed full-span TPS test.

Wang Xunnianis a professor of China Aerodynamics Research and Development Center.His current research interests include acoustic noise and flow separation control using active flow control technology.

Li Huaxingis a professor of Northwestern Polytechnical University.His main research interests include fore-body flow separation control and flow control on a slender delta wing at high angle of attack.

28 August 2015;revised 10 September 2015;accepted 15 March 2016

Available online 26 August 2016

Dielectric barrier discharge;

Flow control;

Plasma actuator;

Turbulent boundary layer;

Wing-body configuration

*Corresponding author.Tel.:+86 816 2461286.

E-mail addresses:lookzx@mail.ustc.edu.cn(X.Zhang),dragonhy@163.com(Y.Huang).

Peer review under responsibility of Editorial Committee of CJA.