Control of the corner separation in a compressor cascade by steady and unsteady plasma aerodynamic actuation

This paper reports experimental results on using steady and unsteady plasma aerodynamic actuation to control the corner separation, which forms over the suction surface and end wall corner of a compressor cascade blade passage. Total pressure recovery coefficient distribution was adopted to evaluate the corner separation. Corner separation causes significant total pressure loss even when the angle of attack is 0°. Both steady and unsteady plasma aerodynamic actuations suppress the corner separation effectively. The control effect obtained by the electrode pair at 25% chord length is as effective as that obtained by all four electrode pairs. Increasing the applied voltage improves the control effect while it augments the power requirement. Increasing the Reynolds number or the angle of attack makes the corner separation more difficult to control. The unsteady actuation is much more effective and requires less power due to the coupling between the unsteady actuation and the separated flow. Duty cycle and excitation frequency are key parameters in unsteady plasma flow control. There are thresholds in both the duty cycle and the excitation frequency, above which the control effect saturates. The maximum relative reduction in total pressure loss coefficient achieved is up to 28% at 70% blade span. The obvious difference between steady and unsteady actuation may be that wall jet governs the flow control effect of steady actuation, while much more vortex induced by unsteady actuation is the reason for better control effect.     

Scaling of flow separation on a pitching low aspect ratio plate

We use two different dye injection approaches, in two different water tunnels, to visualize the formation and subsequent evolution of leading-edge vortices and related separated structures, for a pitching low aspect ratio plate. The motion is a smoothed linear pitch ramp from 0° to 40° incidence, brief hold, and return to 0°, executed at reduced pitch rates ranging from 0.1 to 0.35 and about various pivot locations. All cases evince a leading edge vortex with pronounced axial flow, which leads to formation of large-scale, three-dimensional flow structures, culminating in a large vortical structure centered at the wing symmetry plane. Pitch is also compared to plunge, where the plunge-induced angle of attack is taken as the geometric pitch incidence angle, ignoring pitch-rate effects. At successively increasing values of convective time C/U, the three-dimensional patterns of the flow structure are remarkably similar for the pitching and plunging motions. The similarity of these patterns persists, though they are shifted in time, for variation of either the location of the pitching axis or the dimensionless pitch rate.     

Calibration of a four-hole pyramid probe and area traverse measurements in a short-duration transonic turbine cascade tunnel

A four-hole pyramid probe has been calibrated for use in a short-duration transonic turbine cascade tunnel. The probe is used to create area traverse maps of total and static pressure, and pitch and yaw angles of the flow downstream of a transonic annular cascade. This data is unusual in that it was acquired in a short-duration (5 s of run time) annular cascade blowdown tunnel. A four-hole pyramid probe was used which has a 2.5 mm section head, and has the side faces inclined at 60° to the flow to improve transonic performance. The probe was calibrated in an ejector driven, perforated wall transonic tunnel over the Mach number range 0.5–1.2, with pitch angles from -20° to + 20° and yaw angles from-23° to +23°. A computer driven automatic traversing mechanism and data collection system was used to acquire a large probe calibration matrix (～ 10,000 readings) of non dimensional pitch, yaw, Mach number, and total pressure calibration coefficients. A novel method was used to transform the probe calibration matrix of the raw coefficients into a probe application matrix of the physical flow variables (pitch, yaw, Mach number etc.). The probe application matrix is then used as a fast look-up table to process probe results. With negligible loss of accuracy, this method is faster by two orders of magnitude than the alternative of global interpolation on the raw probe calibration matrix. The blowdown tunnel (mean nozzle guide vane blade ring diameter 1.1 m) creates engine representative Reynolds numbers, transonic Mach numbers and high levels (≈ 13%) of inlet turbulence intensity. Contours of experimental measurements at three different engine relevant conditions and two axial positions have been obtained. An analysis of the data is presented which includes a necessary correction for the finite velocity of the probe. Such a correction is non trivial for the case of fast moving probes in compressible flow.     

Wake visualization of a heaving and pitching foil in a soap film

Many fish depend primarily on their tail beat for propulsion. Such a tail is commonly modeled as a two-dimensional flapping foil. Here we demonstrate a novel experimental setup of such a foil that heaves and pitches in a soap film. The vortical flow field generated by the foil correlates with thickness variations in the soap film, which appear as interference fringes when the film is illuminated with a monochromatic light source (we used a high-frequency SOX lamp). These interference fringes are subsequently captured with high-speed video (500 Hz) and this allows us to study the unsteady vortical field of a flapping foil. The main advantage of our approach is that the flow fields are time and space resolved and can be obtained time-efficiently. The foil is driven by a flapping mechanism that is optimized for studying both fish swimming and insect flight inside and outside the behavioral envelope. The mechanism generates sinusoidal heave and pitch kinematics, pre-described by the non-dimensional heave amplitude (0–6), the pitch amplitude (0°–90°), the phase difference between pitch and heave (0°–360°), and the dimensionless wavelength of the foil (3–18). We obtained this wide range of wavelengths for a foil 4 mm long by minimizing the soap film speed (0.25 m s⁻¹) and maximizing the flapping frequency range (4–25 Hz). The Reynolds number of the foil is of order 1,000 throughout this range. The resulting setup enables an effective assessment of vortex wake topology as a function of flapping kinematics. The efficiency of the method is further improved by carefully eliminating background noise in the visualization (e.g., reflections of the mechanism). This is done by placing mirrors at an angle behind the translucent film such that the camera views the much more distant and out-of-focus reflections of the black laboratory wall. The resulting high-quality flow visualizations require minimal image processing for flow interpretation. Finally, we demonstrate the effectiveness of our setup by visualizing the vortex dynamics of the flapping foil as a function of pitch amplitude by assessing the symmetry of the vortical wake.     

Large Eddy Simulation of Flow Control Around a Cube Subjected to Momentum Injection

The concept of Momentum Injection (MI) through Moving Surface Boundary layer Control (MSBC) applied to a cubic structure is numerically studied using Large Eddy Simulation at a Reynolds number of 6.7×10⁴. Two small rotating cylinders are used to add the momentum at the front vertical edges of the cube. Two configurations are studied with the yaw angle of 0° and 30°, respectively, with ratio of the rotation velocity of cylinders and the freestream velocity of 2. The results suggest that MI delays the boundary layer separation and reattachment, and thus reduces the drag. A drag reduction of about 6.2 % is observed in the 0° yaw angle case and about 44.1 % reduction in the 30° yaw angle case. In the case of 0° yaw angle, the main change of the flow field is the disappearance of the separation regions near the rotating cylinders and the wake region is slightly changed due to MI. In the 30° yaw angle case, the flow field is changed a lot. Large flow separations near one rotating cylinder and in the wake is significantly reduced, which results in the large drag reduction. Meanwhile, the yaw moment is increased about 50.5 %.     

Flow characteristics and flow-induced forces of a stationary and rotating triangular cylinder with different incidence angles at low Reynolds numbers

In this paper, the problem of two-dimensional fluid flow past a stationary and rotationally oscillating equilateral triangular cylinder with a variable incident angle, Reynolds number, oscillating amplitude, and oscillating frequency is numerically investigated. The computations are carried out by using a two-step Taylor-characteristic-based Galerkin (TCBG) algorithm. For the stationary cases, simulations are conducted at various incident angles of α=0.0–60.0° and Reynolds numbers of Re=50–160. For the oscillation cases, the investigations are done at various oscillating amplitudes of θ_max=7.5–30.0° and oscillating frequencies of F_s/F_o=0.5–3.0 considering two different incidence angles (α=0.0°, 60.0°) and three different Reynolds numbers (Re=50, 100, 150). The results show that the influences of key parameters (incidence angle, Reynolds number, oscillating amplitude, and oscillating frequency) are significant on the flow pattern and hydrodynamic forces. For the stationary cases, at smaller angle of incidence (α≤30.0°), Reynolds number has a large impact on the position of the separation points. When α is between 30.0° and 60.0°, it was found that the separation points are located at the rear corners. From a topological point of view, the diagram of flow pattern is summarized, including two distinct patterns, namely, main separation and vortex merging. A deep analysis of the influence of Reynolds number and incidence angles on the mean pressure coefficient along the triangular cylinder surface is presented. Additionally, for the oscillating cases, the lock-on phenomenon is captured. The dominant flow patterns are 2S mode and P+S mode in lock-on region at α=0.0°. It is found at α=60.0°, however, that the flow pattern is predominantly 2S mode. Furthermore, except for the case of F_s/F_o=2.0, the mean drag decreases as the oscillating amplitude increases for each Reynolds number at α=0.0°. At α=60.0°, the minimum mean drag for F_s/F_o=1.5 is lower than that for stationary case, and occurs at θ_max=15.0° (Re=100) and θ_max=22.5° (Re=150), respectively. Finally, the effect of Reynolds number on a rotational oscillation cylinder is elucidated.     

Effect of flow regime change from subsonic to transonic on the air loads of an oscillating airfoil

In this research, the effect of flow regime change from subsonic to transonic on the air loads of a pitching NACA0012 airfoil is investigated. To do this, the effect of change in flow regime on the lift and pitching moment coefficients hysteresis cycles is studied. The harmonic balance approach is utilized for numerical calculation due to its low computational time. Verifications are also made with previous works and good agreements are observed. The assessment of flow regime change on the aforementioned hysteresis cycles is accomplished in the Mach number range of M=0.65–0.755. The reduced frequency and pitch amplitude also vary from k=0.03 to 0.1 and α₀=1–2.51°, respectively. Results show that the effect of increase in Mach number is to increase and decrease the lift coefficient during downstroke and upstroke, respectively, whereas at low reduced frequencies, the effect of increase in Mach number may lead to a reverse manner when airfoil moves toward its extremum angle of attack. Results also reveal that as the pitch amplitude varies, the shape of lift coefficient hysteresis cycle depends more on the pitch amplitude than on the appearance of shock. It is shown that as the Mach number increases, the incidence angles correspond to the extremum pitching moment, and depending on the reduced frequency, lie between zero and extremum angle of attack. These incidence angles shift toward the extremum angle of attack as the reduced frequency decreases. Results also show that the increase in pitch amplitude at low Mach number, in such a way that leads to the formation of shock around the extremum angle of attack, causes the extremum pitching moment to appear around these angles and at high Mach number, depending on the reduced frequency, the extremum pitching moment incidence angles would be between zero and extremum incidence angle.     

Influence of pitching angle of incidence on the dynamic stall behavior of a symmetric airfoil

This study presents the influence of pitch angle of an airfoil on its near-field vortex structure as well as the aerodynamic loads during a dynamic stall process. Dynamic stall behavior in a sinusoidally pitching airfoil is usually analyzed at low to medium reduced frequencies and with the maximum angle of attack of the airfoil not exceeding 25°. In this work, we study dynamic stall of a symmetric airfoil at medium to high reduced frequencies even as the maximum angle of attack goes from 25° to 45°. The evolution and growth of the laminar separation bubble, also known as a dynamic stall vortex, at the leading edge and the trailing edge are studied as the pitch cycle goes from the minimum to the maximum angle of attack. The effect of reduced frequencies on the vortex structure as well as the aerodynamic load coefficients is investigated. The reduced frequency is shown to be a bifurcation parameter triggering period doubling behavior. However, the bifurcation pattern is dependent on the variation of the pitch angle of incidence of the airfoil.     

二维扩压叶栅非定常分离流控制途径探索

二维扩压叶栅非定常黏性数值模拟结果表明,在一定攻角范围内,叶片前缘点附近的周期性吹吸气激励能有效控制混乱的非定常分离流．详细研究了非定常激励频率、幅值、位置对流场的影响．满足一定条件的非定常激励能够使流动由无序变为有序,时均气动性能提高。     

Forces on oscillating foils for propulsion and maneuvering

Experiments were performed on an oscillating foil to assess its performance in producing large forces for propulsion and effective maneuvering. First, experiments on a harmonically heaving and pitching foil were performed to determine its propulsive efficiency under conditions of significant thrust production, as function of the principal parameters: the heave amplitude, Strouhal number, angle of attack, and phase angle between heave and pitch. Planform area thrust coefficients of 2.4 were recorded for 35° maximum angle of attack and efficiencies of up to 71.5% were recorded for 15° maximum angle of attack. A plateau of good efficiency, in the range of 50–60%, is noted. A phase angle of 90–100° between pitch and heave is found to produce the best thrust performance. Also, the introduction of higher harmonics in the heave motion, so as to ensure a sinusoidal variation in the angle of attack produced much higher thrust coefficient at high Strouhal numbers. Second, experiments on a harmonically oscillating foil with a superposed pitch bias, as well as experiments on impulsively moving foils in still water, were conducted to assess the capability of the foil to produce large lateral forces for maneuvering. Mean side force coefficients of up to 5.5, and instantaneous lift coefficients of up to 15 were recorded, demonstrating an outstanding capability for maneuvering force production.     

Aerodynamics of laminar separation flutter at a transitional Reynolds number

Experimental observations of self-sustained pitch oscillations of a NACA 0012 airfoil at transitional Reynolds numbers were recently reported. The aeroelastic limit cycle oscillations, herein labelled as laminar separation flutter, occur in the range 5.0×10⁴≤Re_c≤1.3×10⁵. They are well behaved, have a small amplitude and oscillate about θ=0°. It has been speculated that laminar separation leading to the formation of a laminar separation bubble, occurring at these Reynolds numbers, plays an essential role in these oscillations. This paper focuses on the Re_c=7.7×10⁴ case, with the elastic axis located at 18.6% chord. Considering that the experimental rig acts as a dynamic balance, the aerodynamic moment is derived and is empirically modelled as a generalized Duffing–van-der-Pol nonlinearity. As expected, it behaves nonlinearly with pitch displacement and rate. It also indicates a dynamically unstable equilibrium point, i.e. negative aerodynamic damping. In addition, large eddy simulations of the flow around the airfoil undergoing prescribed simple harmonic motion, using the same amplitude and frequency as the aeroelastic oscillations, are performed. The comparison between the experiment and simulations is conclusive. Both approaches show that the work done by the airflow on the airfoil is positive and both have the same magnitude. The large eddy simulation (LES) computations indicate that at θ=0°, the pitching motion induces a lag in the separation point on both surfaces of the airfoil resulting in negative pitching moment when pitching down, and positive moment when pitching up, thus feeding the LCO.     

Experimental study of pitching and plunging airfoils at low Reynolds numbers

Measurements of the unsteady flow structure and force time history of pitching and plunging SD7003 and flat plate airfoils at low Reynolds numbers are presented. The airfoils were pitched and plunged in the effective angle of attack range of 2.4°–13.6° (shallow-stall kinematics) and ?6° to 22° (deep-stall kinematics). The shallow-stall kinematics results for the SD7003 airfoil show attached flow and laminar-to-turbulent transition at low effective angle of attack during the down stroke motion, while the flat plate model exhibits leading edge separation. Strong Re-number effects were found for the SD7003 airfoil which produced approximately 25 % increase in the peak lift coefficient at Re = 10,000 compared to higher Re flows. The flat plate airfoil showed reduced Re effects due to leading edge separation at the sharper leading edge, and the measured peak lift coefficient was higher than that predicted by unsteady potential flow theory. The deep-stall kinematics resulted in leading edge separation that led to formation of a large leading edge vortex (LEV) and a small trailing edge vortex (TEV) for both airfoils. The measured peak lift coefficient was significantly higher (~50 %) than that for the shallow-stall kinematics. The effect of airfoil shape on lift force was greater than the Re effect. Turbulence statistics were measured as a function of phase using ensemble averages. The results show anisotropic turbulence for the LEV and isotropic turbulence for the TEV. Comparison of unsteady potential flow theory with the experimental data showed better agreement by using the quasi-steady approximation, or setting C(k) = 1 in Theodorsen theory, for leading edge–separated flows.     

The flow over a thin airfoil subjected to elevated levels of freestream turbulence at low Reynolds numbers

Micro Air Vehicles (MAVs) can be difficult to control in the outdoor environment as they fly at relatively low speeds and are of low mass, yet exposed to high levels of freestream turbulence present within the Atmospheric Boundary Layer. In order to examine transient flow phenomena, two turbulence conditions of nominally the same longitudinal integral length scale (Lxx/c?=?1) but with significantly different intensities (Ti?=?7.2?% and 12.3?%) were generated within a wind tunnel; time-varying surface pressure measurements, smoke flow visualization, and wake velocity measurements were made on a thin flat plate airfoil. Rapid changes in oncoming flow pitch angle resulted in the shear layer to separate from the leading edge of the airfoil even at lower geometric angles of attack. At higher geometric angles of attack, massive flow separation occurred at the leading edge followed by enhanced roll up of the shear layer. This lead to the formation of large Leading Edge Vortices (LEVs) that advected at a rate much lower than the mean flow speed while imparting high pressure fluctuations over the airfoil. The rate of LEV formation was dependent on the angle of attack until 10° and it was independent of the turbulence properties tested. The fluctuations in surface pressures and consequently aerodynamic loads were considerably limited on the airfoil bottom surface due to the favorable pressure gradient.     

RUPTURE OF THIN METAL TUBES BY NORMAL AND OBLIQUE IMPACT OF BLUNT CONICALNOSED MISSILES: EXPERIMENTS

Impact tests at both normal and oblique angles of incidence were conducted on thin mild tubes using a moderate size of 90° conical-nosed missiles. The minimum impact speed that generated cracks through the thickness of the wall, termed the speed for rupture, was measured, and various modes of rupture were identified. For a thin tube hit by a missile at a normal angle of obliquity at the speed for rupture, the contact region spreads across the nose of the missile, and the transverse shear deformation is predominant in the final failure process. If the angle of obliquity is 30°, the missile pierces a hole through the wall of the tube. At the speed for rupture, the kinetic energy of the missile for oblique angle 30° is only about 45% that required for plugging at a normal angle of obliquity. The project sponsored by National Natural Science Foundation of China(No. 19842001, 19872048) and Scientific Research Foundation for Returned Overseas Chinese Scholars of State Education Commission and Shanxi Province of China.     

Stability analysis and transition prediction of hypersonic boundary layer over a blunt cone with small nose bluntness at zero angle of attack

Stability and transition prediction of hypersonic boundary layer on a blunt cone with small nose bluntness at zero angle of attack was investigated. The nose radius of the cone is 0.5 mm; the cone half-angle is 5°, and the Mach number of the oncoming flow is 6. The base flow of the blunt cone was obtained by direct numerical simulation. The linear stability theory was applied for the analysis of the first mode and the second mode unstable waves under both isothermal and adiabatic wall condition, and eN method was used for the prediction of transition location. The N factor was tentatively taken as 10, as no experimentally confirmed value was available. It is found that the wall temperature condition has a great effect on the transition location. For adiabatic wall, transition would take place more rearward than those for isothermal wall. And despite that for high Mach number flows, the maximum amplification rate of the second mode wave is far bigger than the maximum amplification rate of the first mode wave, the transition location of the boundary layer with adiabatic wall is controlled by the growth of first mode unstable waves. The methods employed in this paper are expected to be also applicable to the transition prediction for the three dimensional boundary layers on cones with angle of attack.     

Study of the flow around surface-mounted pyramids

The turbulent flow around square-based wall-mounted pyramids in thin and thick boundary layers was experimentally investigated as a function of the pyramid apex angle, ‘, and angle of attack, f, based on mean flow surface patterns, pressure and velocity measurements. For thin boundary layers, wake periodicity is observed. For slender pyramids (15°<‘<75°), the periodic formation and shedding of vortices is observed. The shedding frequency scales with the length scale L=w'(w'/h)^-1/4, where h is the pyramid height and w' is the frontal (projected) base width. For broad pyramids, wake periodicity exists but cannot be related to vortex shedding. Vortex shedding appears related to the existence of a double vortex structure along the side faces of the slender pyramids. The location of the separation point upstream and the attachment point downstream of the obstacle also scales with L. For thick boundary layers, no wake periodicity is observed and the mean flow structure in the wake differs from that seen for thin boundary layers. The location of the separation and attachment points scales only approximately with L.     

An analysis of the structure of laminar separation bubbles in swept infinite geometries

The influence of sweep on the general structure of short separation bubbles in strictly laminar flow fields on swept infinite geometries is investigated by theoretical analysis and direct numerical simulations (DNS). In this situation the ‘independence principle’ of the Navier–Stokes equations for incompressible flow enforces a unique topology for the mean flow, which includes the much better understood unswept separation bubbles as a special case: swept laminar separation bubbles form leading edge parallel streamtubes with a spanwise outflow and a helical motion inside directed parallel to the separation line. If chordwise inflow conditions are kept constant, their cross-sections stay independent of a rising sweep angle, as the spanwise velocity field is then merely superimposed over the unchanged flow of the corresponding unswept case. Their mean flow field follows strict scaling rules that may be derived analytically from the generic 45°-solution, as confirmed by DNS-results for a series of pressure-induced separation bubbles subjected to a systematic variation of the sweep angle between 0° and 60°.     

Evolution of vortex structures in an electromagnetically excited separated flow

Time periodic wall parallel Lorentz forces have been used to excite the separated flow on the suction side of an inclined flat plate. Experiments for a Reynolds number of 10⁴ and an angle of attack of α = 13° are reported. The controlled flow is characterised by a small number of relatively large scale vortices, which are related to the control mechanism. The influence of the main parameters, i.e. the excitation frequency, amplitude and wave form on the suction side flow structures was investigated by analysing time resolved particle image velocimetry (TR-PIV) measurements using continuous wavelet analysis for vortex detection and characterisation. Statistical analysis of the coherent structures of the flow was performed on a large amount of data samples.     

Numerical study of vortex-breakdown over a slender wing in transonic flow at high incidence

The transonic flowfields and vortex-breakdown over a slender wing with the angle of attack from 10° to 28° are studied numerically, and the emphasis is on the secondary separation and the charateristics of vortex-breakdown. The results indicated that: (a) TVD schemes have strong capability for capturing vortices in three-dimensional transonic separated flow at large angle of attack. (b) The development of secondary vortices is more complex than that of leading-edge ones, and is affected by wing's configuration, angle of attack and compressibility simultaneously, and the effect of compressibility is more severe at low angle of attack. (c) The starting angle of attack for vortex-breakdown (when vortex bursting point crosses trailing-edge) is about 18° forM∞=0.85, then the bursting point moves upstream quickly with increasing angle of attack. (d) At α=24°, breakdown occurs over most part of upper side, and the wing begins to stall. Therefore, there is a large lag of angle of attack between the beginning of vortex-breakdown and the stall of the wing. (e) This lag increase with the decreasing of Mach number.     

Parametric study of separation and transition characteristics over an airfoil at low Reynolds numbers

Time-resolved surface pressure measurements are used to experimentally investigate characteristics of separation and transition over a NACA 0018 airfoil for the relatively wide range of chord Reynolds numbers from 50,000 to 250,000 and angles of attack from 0° to 21°. The results provide a comprehensive data set of characteristic parameters for separated shear layer development and reveal important dependencies of these quantities on flow conditions. Mean surface pressure measurements are used to explore the variation in separation bubble position, edge velocity in the separated shear layer, and lift coefficients with angle of attack and Reynolds number. Consistent with previous studies, the separation bubble is found to move upstream and decrease in length as the Reynolds number and angle of attack increase. Above a certain angle of attack, the proximity of the separation bubble to the location of the suction peak results in a reduced lift slope compared to that observed at lower angles. Simultaneous measurements of the time-varying component of surface pressure at various spatial locations on the model are used to estimate the frequency of shear layer instability, maximum root-mean-square (RMS) surface pressure, spatial amplification rates of RMS surface pressure, and convection speeds of the pressure fluctuations in the separation bubble. A power-law correlation between the shear layer instability frequency and Reynolds number is shown to provide an order of magnitude estimate of the central frequency of disturbance amplification for various airfoil geometries at low Reynolds numbers. Maximum RMS surface pressures are found to agree with values measured in separation bubbles over geometries other than airfoils, when normalized by the dynamic pressure based on edge velocity. Spatial amplification rates in the separation bubble increase with both Reynolds number and angle of attack, causing the accompanying decrease in separation bubble length. Values of the convection speed of pressure fluctuations in the separated shear layer are measured to be between 35 and 50% of the edge velocity, consistent with predictions of linear stability theory for separated shear layers.