Subsonic Freestream Research Articles

Exploration of High-Frequency Actuation on Gust Response Using Large Eddy Simulation

Abstract Mitigation of gust-induced separation and aerodynamic loads using a high-frequency blowing/suction slot is demonstrated using high-order implicit large eddy simulation (LES). This approach was previously shown to be effective at alleviating dynamic stall on pitching wings. A NACA0012 wing section operating at a transitional chord-based Reynolds number of Rec=500,000, subsonic freestream Mach number of M∞=0.1, and angles of attack of α = 4 deg and 12 deg is subjected to discrete 1-cos transverse gusts. Gust-induced stall is demonstrated and then active flow control (AFC) is applied to cases vulnerable to gust-induced stall. The flow control strategy is shown to be effective at stall suppression during gust encounter, thereby providing partial alleviation of gust induced loads. This approach is most effective at attenuating pitching moment increment.

Numerička analiza tegljajućeg nadzvučnog dvokrilca u uslovima van projektovanja sa zaklopkom zadnje ivice

Supersonic biplanes can achieve low-boom and low-drag supersonic flights. In the present study, aerodynamic analysis of a two-dimensional stagger Bussmann biplane (staggered upper element by 0.5c) at zero degrees angle of attack with trailing edge flap was investigated with the help of computational fluid dynamics (CFD) tools. Due to the wave cancellation effect, the Busemann biplane delivers a positive drag reduction at design supersonic Mach values. However, when operating outside of its intended parameters, it performs worse, and the wave cancellation effect has no beneficial effects on reducing drag. Another issue with the Busemann biplane is flow chocking, which produces a potent bow shock wave in front of the aircraft. This paper attempts to address low aerodynamics efficiency problems during take-off through numerical simulation of a staggered Busemann biplane with trailing edge flaps at zero degrees angle of attack. It was confirmed that the staggered biplane airfoil with flap has better aerodynamic performance during take-off at lower subsonic free stream Mach numbers. In 2D wings, the effect of flow chocking and hysteresis as starting problems, which arise when the biplanes accelerate from low Mach numbers, is reduced by using the suitable dimension and angle of rotation of the flap, and the flap is effective in settling these issues.

Transition Prediction of Boundary Layers in the Presence of Backward-Facing Steps

A set of available engineering models for the prediction of transition to turbulence is evaluated via comparisons with a high-quality experiment involving a backward-facing step (BFS) on a flat plate at subsonic freestream conditions. The streamwise shift in the transition location is monitored as the step height and flow speed are varied across step-height-to-local-displacement-thickness ratios of . We apply the -factor method based on the linear amplification of boundary-layer instabilities to laminar two-dimensional basic states. -factor correlations that rely on quasi-parallel linear stability theory (LST), the parabolized stability equations, and the harmonic linearized Navier–Stokes equations (HLNSE) yield good agreement with the experimental data, except if the onset of transition occurs slightly downstream of the BFS. For LST, the neglect of nonparallel flow effects results in transition locations that are further upstream compared to HLNSE. In contrast, models that use auxiliary transport equations and transition correlations based on local parameters fail to capture the physics associated with a BFS for moderate step heights. Specifically, the results obtained with the model indicate that it is unable to account for the flow history effects. Results show the amplification-factor-transport model also fails to capture the BFS effects in spite of accounting for the basic-state distortion near the step.

Wake response downstream of a spanwise-oscillating hemispherical turret

An experimental Pressure Sensitive Paint (PSP) study of the unsteady pressure fields downstream of and on an oscillating hemispherical turret at subsonic freestream Mach numbers between 0.3 and 0.6 and Reynolds numbers of about 2 million is presented. The oscillating turret consisted of a hemispheric shell, mounted on an aluminum rectangular plate and is designed to oscillate in the spanwise direction at a single frequency, which lies within the frequency range of the dominant wake motion. The resonant-based aero-elastic response of the turret to the flow resulted in intermittent turret oscillations. Pressure fluctuations in the wake were found to be less energetic for the oscillating turret and the size of the recirculating region was increased, compared to the stationary turret. POD-based modal analysis of the pressure fields revealed that the wake dynamics is characterized by global anti-symmetric and symmetric shedding and showed that the oscillating turret primarily suppressed the global anti-symmetric shedding. The analysis of the unsteady forces acting on the turret demonstrated that the oscillating turret reduces the spanwise component of the unsteady force, while keeping other force components mostly unchanged. Intermittent nature of the spanwise force, which is responsible for the intermittent motion of the oscillating turret, was observed and discussed.

Computational evaluation of geometric effects on aerodynamic performance of circulation control airfoils

The geometric effects of Coanda trailing edges on the aerodynamic performance of an airfoil are numerically evaluated for a range of different freestream Mach numbers and momentum coefficients. A Circulation control (CC) airfoil with a circular trailing edge (ACTE) proves to have better control effectiveness at low subsonic freestream speeds (Mach = 0.1). A CC airfoil having an elliptic trailing edge (AETE) outperforms the ACTE at high subsonic flow conditions. The occurrence of C μ-stall for the AETE is greatly postponed, and meanwhile the maximum net lift coefficient increment achieved for the AETE (Δ C L = 0.51) is slightly higher than that of the ACTE (Δ C L = 0.50) at Mach 0.6. Compared to the ACTE, the AETE is found to have better control consistency at different operating velocities and better control stability when the Coanda jet is supersonic. Through careful consideration of the aerodynamic performance and the control effects, the most appropriate axial ratio for an AETE ellipse is within the interval from 1.5 to 2. Finally, the flow field instability phenomenon and the jet detachment induced by the supersonic Coanda jet are investigated. A self-sustained shock-wave instability phenomenon without jet detachment is first observed in this paper.

Experimental and Numerical Investigations of the Characteristics of Pulsed Thermal Actuators

The purpose of the study is to numerically, experimentally, and analytically investigate the characteristics of plasma pulsed thermal actuators (PT actuators) and to assess their possibilities in controlling flow around airfoils, wings, and configurations at large subsonic freestream velocities. For the PT actuators of the types considered the mathematical models adequately describing their effect on flow past bodies are developed. The characteristics of a prototype PT actuator are experimentally investigated on a specially developed rig. A new type of the PT actuator equippedwith a channel (PTC actuator) is proposed; it is designed to operate at a high pulse repetition frequency and at large flow velocities. Numerical investigations show that the PTC actuators are free of the essential and fundamental shortcoming of the PT actuators which consists in the working zone superheating at high pulse repetition frequencies.

Numerical and experimental investigation of the means for reducing the aeroacoustic loads in an extended rectangular cavity at subsonic and transonic freestream velocities

The results of a comprehensive investigation including numerical calculations and experiments with models in a wind tunnel and a vehicle under flight conditions aiming to find the ways of reducing the pressure fluctuation levels in an extended cavity at subsonic and transonic freestream velocities are presented. It is shown that the reduction of these loads can be achieved using the means which have demonstrated their effectiveness for cavities with the open-type flows, for example, a permeable deflector and the bevelling of the rear wall but only in the case of a given combination of their geometric parameters. The mechanisms of the action of these devices on the flow, thanks to which the intensity of the wave disturbances generated by the rear wall is reduced, the instability wave growth in the mixing layer behind the deflector is limited, and the fluctuation level in the cavity decreases, are investigated. The results of numerical investigations of the flow in a cavity with a permeable deflector are apparently among the first.

Staggered Approach for Fluid-Structure Interaction Phenomena of an AGARD 445.6 Wing Using Commercial CFD/CSM Software

This paper presents the development of an efficient time-marching simulation technology for fluid-structure interaction applications using commercial computational fluid dynamics (CFD) and computational solid mechanics (CSM) software and in-house modules. By splitting the multidisciplinary coupling problem into three major analysis fields (the fluid field, the structure field, and the moving mesh pseudofield) one can solve each field using already available and validated software and the coupling between the different analysis fields can be accomplished by in-house software (FORTRAN and C codes). This simulation technology is able to predict both unsteady interaction phenomena and nonlinear aeroelastic problems in reasonable time, taking advantage of a high-performance computing network. Commercial CFD software was used to solve the Navier-Stokes equations and the chosen turbulence models of the fluid field while the structural field was solved using commercial CSM software. The transfer of information from the fluid field to the structural field and vice versa is computed by an in-house interpolation module, which reduces the user interaction to a minimum initialization. The third field, the moving mesh, was solved with another in-house code, which employs the spring network analogy and the elastic material analogy. The advancement of the solution is controlled by means of FORTRAN and C codes that manage the run/wait conditions of software modules based on the staggered procedures proposed in previous research. The proposed aeroelastic simulation technology was validated with the experimental data available for the advisory group for aerospace research and development (AGARD) 445.6 standard aeroelastic configuration for dynamic response at subsonic and transonic free stream Mach numbers.

On an Acoustic Oscillation Energy for Shear Flows

The pressure perturbation and its derivative for an acoustic wave in a uniformly moving medium are canonical variables whose Hamiltonean is the acoustic energy. In the present paper is introduced a generalized acoustic Hamiltonean or oscillation energy that: (i) is conserved in a uniform flow, when it reduces to the acoustic energy; (ii) is not conserved in an homentropic unidirectional shear flow, in which case it has minimum at the critical layer where the Doppler shifted frequency vanishes. These properties are illustrated in the case of an unidirectional shear flow with an hyperbolic tangent velocity profile, representing a boundary layer over a flat rigid or impedance wall. The exact solution in the boundary layer can be matched to plane waves in the free stream, and reflection and transmission coefficients are defined for a incident wave of unit amplitude. If there is no critical layer in the boundary layer, a single powers series solutions covers the whole flow region from the free stream up to the wall; it is shown that for a rigid wall the modulus of the reflection coefficient is unity, but the argument is not zero, i.e. all sound is reflected with a phase shift. Also it is shown that, in this case, of a boundary layer without critical layer over a rigid wall, the phase of the transmission coefficient is one-half the phase the phase of the reflection coefficient, i.e. sound takes the same time to cross the boundary layer both ways, bouncing from the wall in between. In more general cases, viz. impedance wall and/or boundary layer with a critical layer, the modulus and phase of the reflection and transmission coefficients are calculated as functions of the angle of incidence, free stream Mach number and boundary layer thickness measured on a wavelength scale. When a critical layer is present in the boundary layer, the two pairs of solutions around it and the free stream must be matched, in order to obtain the acoustic field across the whole boundary layer, and calculate the reflection and transmission coefficients. The special case of critical layer at the free stream is also discussed. The cases discussed include boundary layers with and without a critical layer, subsonic and supersonic free streams with one zone of silence and one or two propagation zone(s), and rigid, inductive, reactive and mixed impedance walls. The generalized acoustic Hamiltonean or oscillation energy is also used to obtain a simple approximation for the modulus of the transmission factor, which is reasonably accurate not far from normal incidence.

Tomographic Particle Image Velocimetry Measurements in a Bluff-Body Wake Flow

I N the wake of bluff bodies, flow separation occurs along with phenomena like recirculation, increased turbulence levels, periodic shedding processes, and low-pressure regions. For flow bodies, which cannot avoid such geometries, base drag and buffeting effects can vitally reduce performance and efficiency. This holds true for axisymmetric, cylindrical bodies placed parallel to freestream direction, like the generic rocket configuration considered in the current work. Similar geometries were studied extensively for subsonic freestream conditions, with early studies concentrating on static or time-averaged quantities [1,2]. Following the evolution of both experimental and numerical methods, the unsteadiness of the wake turned into focus. Many studies reported a macroscopic vortex shedding mode at a reduced frequency of StD 0:2 [3–5], that was also detected in hot-wire and dynamic base pressure measurements regarding the present case [6]. The instability of the shear layer is assumed to be a key factor in the shedding mechanism and the wake topology [4]. Correspondingly, Pastoor et al. [7] used zero-net-massflux actuation to actively manipulate the shear layer of a bluff body over wide Reynolds ranges. A possible overall drag reduction of 15% was reported. Perret [8] investigated the wake of a quasitwo-dimensional (2-D) cylinder by means of two-component (2C) particle image velocimetry (PIV) for ReD 1:25 10 [8]. Within the shear layer, a vortex population was detected using the swirling strength criterion. The corresponding structures were both small (<5% of the model diameter) and of high intensity compared to the macroscopic Karman vortices, and, thus, connected to the Kelvin– Helmholtz instabilitymechanism. It was also pointed out that for 2-D measurements, only the planar swirling motion can be taken into account and may be affected or biased by the out-of-plane velocity and inclined vortex axes [8]. Fewer results can be found regarding three-dimensional (3-D) test cases. Deck and Garnier [5] performed both detached and large eddy simulations (DES/LES) on the wake flow of a cylindrical, generic launch vehicle model. The bluff base geometry was prolonged by a Presented at the 41st AIAA Fluid Dynamics Conference and Exhibit, Honolulu, Hawaii, June 27–30, 2011; received 27 November 2011; revision received 10 April 2012; accepted for publication 11 April 2012. Copyright © 2012 by Institute of Aeronautics and Astronautics, RWTH Aachen University. Published by the American Institute of Aeronautics and Astronautics, Inc., with permission. Copies of this paper may be made for personal or internal use, on condition that the copier pay the $10.00 per-copy fee to the Copyright ClearanceCenter, Inc., 222RosewoodDrive,Danvers,MA01923; include the code 0001-1452/12 and $10.00 in correspondence with the CCC. Research Engineer, Wuellnerstr 7. Member AIAA. Senior Research Engineer, Wuellnerstr 7. AIAA JOURNAL Vol. 50, No. 12, December 2012

Contributions of turbulence to subsonic cavity flow wall pressures

Poisson-type pressure sources in the flow over open cavities were examined for several subsonic free stream conditions. Wall-mounted pressure transducers and two-component particle image velocimetry (PIV) were employed synchronously to obtain data at Mach 0.19, 0.29, 0.39, 0.58, and 0.73 for a rectangular cavity with a length-to-depth ratio of 6. The modified quadratic stochastic estimation was used to yield a time-resolved, low-dimensional representation of the velocity field using the PIV and pressure measurements. A conformal transformation was applied to the cavity geometry to allow the integration of Poisson’s equation giving a time-resolved estimate of the surface pressure fluctuations. The use of Poisson’s equation allowed the contribution of the mean-shear and nonlinear velocity sources of the surface pressure loads to be evaluated independently, showing that the resonant features of the cavity flow (Rossiter modes) are dominant in the mean-shear contribution and that nonlinear interactions become increasingly significant above Mach 0.4.

Properties of subsonic open cavity flow fields

Flow over an open cavity was studied for several different subsonic free stream Mach numbers ranging from 0.19 to nearly 0.73. Velocity field information was acquired through an application of particle image velocimetry, while the fluctuating surface pressure was acquired through a linear array of surface pressure sensors. These data were acquired on the centerline of the cavity which had a length to depth ratio of 6 and a turbulent boundary layer upstream of its leading edge. Over the range of free stream Mach numbers the fluctuating surface pressure spectra in the cavity exhibited different behavior ranging from no apparent resonance to resonance being dominated by the second or third Rossiter modes. The broadband levels of surface pressure spectra with strong resonant tones collapse with scaling by the flow dynamic pressure. Velocity measurements reveal that the center of circulation of the flow within the cavity moves from the aft wall towards the center of the cavity with increasing Mach number. The trend in the mean flow was also apparent in the mean square fluctuating velocities although to a lesser extent. Application of the proper orthogonal decomposition was performed independently on the velocity fields from each of the different free stream cases yielding a spatially dependent basis set. Using a similarity parameter between these spatial orthogonal basis functions it was shown that the first two modes for all of the cases were quantitatively similar. Furthermore this analysis showed some higher less energetic modes that were similar between the cases.

Two noncircular cross-section bodies and a high wing-body configuration at incidence in a low subsonic free stream

This experimental study examines the flow characteristics over a square cross-section body with sharp edges, a rectangular one with rounded edges, and a high wing-body configuration, in a low subsonic free stream. Pressure, velocity, force measurements and flow visualization provide a picture of the flow behaviour, locally and globally. Flow visualization and velocity measurements clearly depict the presence of two axial counter-rotating vortices in the leeside of the first two bodies, being mainly responsible for the asymmetric loading at nonzero roll angles, maximizing the side-force at a roll angle of about 25°. For all body orientations there is always a recirculation region at the nose-afterbody junction leeside area, the extent of which depends mainly on the roll angle. Pressure gradients take high values at the corners of the after body cross-section, even when these are rounded. No asymmetries were practically detected for a zero roll angle and pitch angles up to 20° for the examined three models. The wing-body configuration exhibits a higher lift slope and a more negative zero lift angle, compared to the wing-alone case, and the side-force increases monotonically with the roll angle, without showing any maximum in contrast to the other two models.

The effect of viscosity on the free stream in transonic flow over a corner point

An asymptotic analysis of the system of Navier–Stokes equations for describing the flow which arises from the subsonic free stream in the neighbourhood of the vertex of a convex corner with curvilinear generatrices is presented for Reynolds numbers approaching infinity. It is assumed that, in limiting non-viscous flow, the subsonic free stream reaches the velocity of sound at the vertex of the corner and, in the first approximation, is described by the Vaglio–Laurin solution. It is shown that the flow can have a different form depending on the value of the pressure gradient, which is formed in the neighbourhood of the corner point. However, irrespective of the steady form of the flow, as a result of the interaction of the Vaglio–Laurin flow with the boundary layer, the latter induces perturbations in the outer flow, which “rounds off” the vertex of the corner when there is a transonic flow around it.

Acoustic excitation of stationary streamwise structures in a separation region on a straight wing

An effect of sound on the flow with a transitional separation bubble on a wing model at low subsonic freestream velocity was experimentally investigated. Using hot-wire measurements and liquid-crystal visualization, stationary streamwise structures were found to appear in the reattachment region. Quantitative characteristics of the disturbances upcoming during the breakdown to turbulence were measured.

On the structure of vortex breakdown on a delta wing

This paper presents a detailed description of the vortex breakdown phenomenon on a delta wing. While there have been a number of earlier studies which have given us a broad, overall picture of these interesting and important flow fields, none has so far yielded the kind of fine structure considered here, in spite of using very fine grids. The present work has been made possible by the use of a new type of grid, called the embedded conical grid, which exploits the special nature of the flow field. Based on the Euler equations, the flow field past a 70 degrees sharp–edged delta wing in a subsonic free–stream is resolved here using this grid. The breakdown is found to begin as a rapid deceleration of the inner core due to a sudden increase in the adverse axial pressure gradient. This leads to a sudden broadening of the core and the formation of a deficit velocity region which becomes intensified causing axial flow reversal. The near–axis vorticity lines get tilted away from the vortex axis and their spiralling direction is reversed. The effect of artificial dissipation and grid coarseness on the computed vortex breakdown has been examined. The results presented here, though obtained for the delta wing geometry, are expected to add to our understanding of vortex breakdown in other flow situations as well.

Accurate development of leading-edge vortex using an embedded conical grid

The vortex formation and its development near the apex of a sharp-edged delta wing are poorly represented on conventional grids because of dissimilar length scales of the flow and the grid. The problem is overcome by having an embedded conical grid surrounding the wing. Computed Euler solutions on such an embedded conical grid and on a conventional H-O grid are presented for a range of incidence for a subsonic freestream. The solutions on the embedded conical grid show that the vortex is well resolved right from the apex, even at a very low incidence. The vortex breakdown is well predicted on the embedded conical grid, including the case where the breakdown has just moved upstream of the trailing edge.

Numerical investigation of dynamic stall of an oscillating wing

The unsteady three-dimensional flowfield over an oscillating wing is investigated with the numerical solution of the compressible, time-dependent, Reynolds-averaged Navier-Stokes equations. Spatial discretization is performed with a third-order accurate, upwind-biased, vertex-based, finite volume scheme. An alternative direction implicit, iterative scheme is used for the time integration. The high Reynolds number turbulent flow behavior is modeled with a one-equation turbulence model. The effect of subiterations, time step and grid density on the accuracy of the computed solutions is investigated. It is found that scaling of the time step with the angular velocity of the motion produces accurate solutions at a reduced computational cost. The computational domain over an aspect ratio 5 wing with rounded tip and NACA-0015 airfoil sections is discretized with a single-block grid. The light stall flowfield over the wing oscillating in a subsonic freestream with a mean angle of attack of 11 deg and an amplitude of 4.2 deg is computed. The structure of the separated, unsteady flowfield is investigated and comparisons with available experimental data are performed.

Numerical investigation of high incidence flow over a double-delta wing

The vortical flowfield over a double-delta wing configuration, consisting of a sharp leading-edge 76-deg sweep strake and a 40-deg sweep wing section is investigated numerically. The governing equations are solved with a partially upwind, finite difference, two-factor algorithm. The leeward-side vortex system resulting from the strake and wing vortices is investigated for a subsonic freestream speed of M^ = 0.22, high Reynolds number turbulent flow at various angles of incidence. At low angles of attack the strake and wing vortices remain separate over the wing section, whereas for flows at higher angles of attack the two vortices merge and vortex breakdown develops. Vortex breakdown appears initially in the trailing-edge region of the wing section. As the angle of attack increases, bursting occurs further upstream closer to the strake section. The effect of numerical grid density is investigated, and the solutions are compared with available experimental data. The computed surface pressures are in good agreement with the experimental measurements for the lower angles of attack, but the agreement deteriorates as the angle of attack increases.

Transonic flows of Bethe—Zel'dovich—Thompson fluids

Bethe–Zel'dovich–Thompson fluids are ordinary, single-phase fluids in which the fundamental derivative of gasdynamics is negative over a finite range of temperatures and pressures. We examine the steady transonic flow of these fluids over two-dimensional thin wings and turbine blades. The free-stream state is taken to be in the neighbourhood of one of the zeros of the fundamental derivative. It is shown that a modified form of the classical transonic small-disturbance equation governs such flows. Critical Mach number estimates are provided which take into account the non-monotone variation of the Mach number with pressure predicted in previous investigations. Critical Mach numbers well over 0.95 are predicted for conventional airfoil sections. Numerical solutions reveal substantial reductions in the strength of the compression shocks occurring in supercritical flows. Further new results include the prediction of expansion and compression shocks in the same flow and compression bow shocks in flows with subsonic free streams.