Skin-friction drag reduction in turbulent channel flow based on streamwise shear control

It is known that stretching and intensification of a hairpin vortex by mean shear play an important role to create a hairpin vortex packet, which generates the large Reynolds shear stress associated with skin-friction drag in wall-bounded turbulent flows. In order to suppress the mean shear at the wall for high efficient drag reduction (DR), in the present study, we explore an active flow control concept using streamwise shear control (SSC) at the wall. The longitudinal control surface is periodically spanwise-arranged with no-control surface while varying the structural spacing, and an amplitude parameter for imposing the strength of the actuating streamwise velocity at the wall is introduced to further enhance the skin-friction DR. Significant DR is observed with an increase in the two parameters with an accompanying reduction of the Reynolds stresses and vorticity fluctuations, although a further increase in the parameters amplifies the turbulence activity in the near-wall region. In order to study the direct relationship between turbulent vortical structures and DR under the SSC, temporal evolution with initial eddies extracted by conditional averages for Reynolds-stress-maximizing Q2 events are examined. It is shown that the generation of new vortices is dramatically inhibited with an increase in the parameters throughout the flow, causing fewer vortices to be generated under the control. However, when the structural spacing is sufficiently large, the generation of new vortex is not suppressed over the no-control surface in the near-wall region, resulting in an increase of the second- and fourth-quadrant Reynolds shear stresses. Although strong actuating velocity intensifies the near-wall turbulence, the increase in the turbulence activity is attributed to the generation of counter-clockwise near-wall vortices by the increased vortex transport.     

Skin-friction drag reduction in a high-Reynolds-number turbulent boundary layer via real-time control of large-scale structures

While large-scale motions are most energetic in the logarithmic region of a high-Reynolds-number turbulent boundary layer, they also have an influence in the inner-region. In this paper we describe an experimental investigation of manipulating the large-scale motions and reveal how this affects the turbulence and skin-friction drag. A boundary layer with a friction Reynolds number of 14 400 is controlled using a spanwise array of nine wall-normal jets operated in an on/off mode and with an exit velocity that causes the jets in cross-flow to penetrate within the log-region. Each jet is triggered in real-time with an active controller, driven by a time-resolved footprint of the large-scale motions acquired upstream. Nominally, the controller injects air into large-scale zones with positive streamwise velocity fluctuations; these zones are associated with positive wall-shear stress fluctuations. This control scheme reduced the streamwise turbulence intensity in the log-region up to a downstream distance of more than five times the boundary layer thickness, δ, from the point of actuation. The highest reduction in spectral energy—more than 30%—was found for wavelengths larger than 5δ in the log-region at 1.7δ downstream of actuation, while scales larger than 2δ still comprised more than 15% energy reduction in the near-wall region. In addition, a 3.2% reduction in mean skin-friction drag was achieved at 1.7δ downstream of actuation. Our reductions of the streamwise turbulence intensity and mean skin-friction drag exceed a base line control-case, for which the jet actuators were operated with the same temporal pattern, but not synchronised with the incoming large-scale zones of positive fluctuating velocity.     

An algebraic closure for the DNS of fiber-induced turbulent drag reduction in a channel flow

An algebraic closure for the non-Newtonian Navier–Stokes equations is presented which accounts for the effect of a dilute fiber suspension. The model is intended to be used in simulations of turbulent drag reduction by fiber additives, and can be considered as a computationally efficient alternative to the existing rheological models for fiber suspensions in turbulent wall-bounded flows. It is based on the assumption that the suspended elongated particles are aligned with the local velocity fluctuation vector. The model is proved to be Galilean invariant. One-way coupled simulations and comparison with a direct solution of the underlying Fokker–Planck equation show a considerable improvement over an existing and comparable model. Finally, two-way coupled simulations demonstrate that the model predicts flow statistics that are in very good agreement with those obtained by the moment approximation approach. Interestingly, the model is realistic in terms of the polymer concentration. Using the proposed model, the cost of simulating a drag-reduced flow in terms of CPU-time is slightly more than that of a Newtonian flow.     

Modeling of drag reduction in turbulent channel flow with hydrophobic walls by FVM method and weakly-compressible flow equations

In this paper the effects of hydrophobic wall on skin-friction drag in the channel flow are investigated through large eddy simulation on the basis of weaklycompressible flow equations with the MacCormack's scheme on collocated mesh in the FVM framework. The slip length model is adopted to describe the behavior of the slip velocities in the streamwise and spanwise directions at the interface between the hydrophobic wall and turbulent channel flow. Simulation results are presented by analyzing flow behaviors over hydrophobic wall with the Smagorinky subgrid-scale model and a dynamic model on computational meshes of different resolutions. Comparison and analysis are made on the distributions of timeaveraged velocity, velocity fluctuations, Reynolds stress as well as the skin-friction drag. Excellent agreement between the present study and previous results demonstrates the accuracy of the simple classical second-order scheme in representing turbulent vertox near hydrophobic wall. In addition, the relation of drag reduction efficiency versus time-averaged slip velocity is established. It is also foundthat the decrease of velocity gradient in the close wall region is responsible for the drag reduction. Considering its advantages of high calculation precision and efficiency, the present method has good prospect in its application to practical projects.     

Mechanism of drag reduction by spanwise oscillating Lorentz force in turbulent channel flow

Numerical simulations and experimental research are both carried out to investigate the controlled effect of spanwise oscillating Lorentz force on a turbulent channel flow. The variations of the streaks and the skin friction drag are obtained through the PIV system and the drag measurement system, respectively. The flow field in the near-wall region is shown through direct numerical simulations utilizing spectral method. The experimental results are consistent with the numerical simulation results qualitatively, and both the results indicate that the streaks are tilted into the spanwise direction and the drag reduction utilizing spanwise oscillating Lorentz forces can be realized. The numerical simulation results reveal more detail of the drag reduction mechanism which can be explained, since the spanwise vorticity generated from the interaction between the induced Stokes layer and intrinsic turbulent flow in the near-wall region can make the longitudinal vortices tilt and oscillate, and leads to turbulence suppression and drag reduction.     

The Berlin oil channel for drag reduction research

For drag reduction research an oil channel has been designed and built. It is also well suited for investigations on turbulent flow and in particular on the dynamics of the viscous sublayer near the wall. The thickness of the viscous sublayer (y ⁺= 5) can be varied between 1 and 4 mm. Surfaces with longitudinal ribs (riblets), which are known to reduce drag, can have fairly large dimensions. The lateral spacing of the ribs can lie between 3 and 10 mm, as compared to about 0.5 mm spacing for conventional wind tunnels. It has been proved by appropriate tests that the oil channel data are completely equivalent to data from other facilities and with other mean flow geometries. However, the shear stress data from the new oil channel are much more accurate than previous data due to a novel differential shear force balance with an accuracy of ±0.2%. In addition to shear stress measurements, velocity fluctuation measurements can be carried out with hot wire or hot film probes. In order to calibrate these probes, a moving sled permits to emulate the flow velocities with the fluid in the channel at rest. A number of additional innovations contribute to the improvement of the measurements, such as, e.g., (i) novel adjustable turbulators to maintain equilibrium turbulence in the channel, (ii) a bubble trap to avoid bubbles in the channel at high flow velocities, (iii) a simple method for the precision calibration of manometers, and (iv) the elimination of (Coulomb) friction in ball bearings. This latter fairly general invention is used for the wheels of the calibration unit of the balance. The channel has a cross section of 25 × 85 cm and is 11 m long. It is filled with about 4.5 metric tons of baby oil (white paraffine oil), which is transparent and odorless like water. The kinematic viscosity of the oil is v = 1.2×10^–5 m²/s, and the highest (average) velocity is 1.29 m/s. Thus, the Reynolds number range (calculated with the channel width, 0.25 m) lies between 5,000 and 26,800 for fully established turbulent flow.The material of this paper has been partly presented at the 5^th European Drag Reduction Working Meeting 15^th and 16^th November 1990, London     

Polymer drag reduction with surface roughness in flat-plate turbulent boundary layer flow

Experimental results from a study of surface roughness effects on polymer drag reduction in a zero-pressure gradient flat-plate turbulent boundary layer are presented. Both slot-injected polymer and homogeneous polymer ocean cases were considered over a range of flow conditions and surface roughness. Balance measurements of skin friction drag reduction are presented. Drag reductions over 60% were measured for both the injected and homogeneous polymer cases even with fully rough surfaces. As the roughness increased, higher polymer concentration was required to achieve a given level of drag reduction for the homogeneous case. With polymer injection, increasing surface roughness caused the drag reduction to decrease to low levels more quickly when the polymer expenditure was decreased or the freestream velocity was increased. However, the percent drag reductions on the rough surfaces with polymer injection were often substantially larger than on the smooth surface. Remarkably, in some cases, the skin friction drag force on a rough surface with polymer injection was less than the drag force observed on a smooth surface at comparable conditions. An erratum to this article can be found at     

Microbubble drag reduction in rough walled turbulent boundary layers with comparison against polymer drag reduction

Experiments were conducted in the 12-inch diameter tunnel at the Applied Research Laboratory, Pennsylvania State University using the tunnel wall boundary layer to determine the influence of surface roughness on microbubble drag reduction. To accomplish this, carbon dioxide was injected through a slot at rates of 0.001 m³/s to 0.011 m³/s, and the resulting skin friction drag measured on a 317.5-mm long by 152.4-mm span balance. In addition to the hydrodynamically smooth balance plate, additional plates were covered with roughly 75, 150, and 300 micron grit. Over the speed range tested of 7.6, 10.7, and 13.7 m/s, the roughness ranged from smooth to fully rough. Not only was microbubble drag reduction achieved over the rough surfaces, but the % drag reduction at a given gas flow rate was larger for larger roughness. Scaling of the data is discussed. Comparison against results of a polymer drag reduction experiment, using the same facility, is made. Finally, a measure of the expected persistence of the phenomenon is given.     

超声速钝体逆向喷流减阻的数值模拟研究

为研究逆向喷流技术对超声速钝体减阻的影响,采用标准k-ε湍流模型,通过求解二维Navier-Stokes方程对超声速球头体逆向冷喷流流场进行了数值模拟,并着重分析了喷口总压、喷口尺寸对流场模态和减阻效果的影响。计算结果显示:随着喷流总压的变化,流场可出现两种流动模态,即长射流穿透模态和短射流穿透模态;喷流能使球头体受到的阻力明显减小;存在最大减阻临界喷流总压值(在所研究参数范围内最大减阻可达51.1%);在其它喷流物理参数不变时,随着喷口尺寸的增大,同一流动模态下的减阻效果下降。本文的研究对超声速钝体减阻技术在工程上的应用具有一定的参考价值。     

Prediction of drag reduction effect by streamwise traveling wave-like wall deformation in turbulent channel flow at practically high Reynolds numbers

A fully developed turbulent channel flow controlled by traveling wave-like wall deformation under a constant pressure gradient condition is studied numerically and theoretically. First, direct numerical simulation (DNS) at three different friction Reynolds numbers, ${\text{Re}}_{\tau }=90,$ 180, and 360, are performed to investigate the modification in turbulence statistics and their scaling. Unlike the previous study assuming a constant flow rate condition, suppression of the quasi-streamwise vortices is not observed in either drag decrease cases or drag increase cases. It is found in the drag reduction case, however, that the periodic component of the Reynolds shear stress (periodic RSS) is largely negative in the viscous sublayer and the buffer layer. For the maximum drag reduction case, the set of control parameters is found to be identical in wall units regardless of the Reynolds number, and the resulting mean velocity profiles are also observed to be approximately similar even with an additional case of ${\text{Re}}_{\tau }=720$. Based on this scaling, we propose a semi-empirical formula for the mean velocity profile modified by the present control. With this formula, about $20\%-25\%$ drag reduction effect is predicted even at practically high Reynolds numbers, ${\text{Re}}_{\tau }\sim {10}^5-{10}^6$.     

Bluff-body drag reduction using a deflector

A passive flow control on a generic car model was experimentally studied. This control consists of a deflector placed on the upper edge of the model rear window. The study was carried out in a wind tunnel at Reynolds numbers based on the model height of 3.1 × 10⁵ and 7.7 × 10⁵. The flow was investigated via standard and stereoscopic particle image velocimetry, Kiel pressure probes and surface flow visualization. The aerodynamic drag was measured using an external balance and calculated using a wake survey method. Drag reductions up to 9% were obtained depending on the deflector angle. The deflector increases the separated region on the rear window. The results show that when this separated region is wide enough, it disrupts the development of the counter-rotating longitudinal vortices appearing on the lateral edges of the rear window. The current study suggests that flow control on such geometries should consider all the flow structures that contribute to the model wake flow.     

Robust design and numerical simulation on drag reduction by a mixture film for liquid turbulence in a channel

An important way of increasing the speed and lowering the fuel consumption of ships is by decreasing the frictional drag. One of the most promising techniques for reducing drag is the use of air bubbles. The goal of this investigation is to establish a set of optimum robust parametric levels for drag reduction by a mixture (air–water) film in turbulent channel flow. Based on the conditions laid out by the Taguchi orthogonal array method, turbulent flows, with air bubbles injected into a channel, are simulated using commercial computational fluid dynamics software. The local shear stress on the upper wall is computed to evaluate the efficiency of drag reduction. Many factors can affect drag reduction. The factors investigated in this study are the rate of air injection, bubble size, area of air injection, flow speed, and measured position of the shear stress. These factors have been investigated through the analysis of variance, which has revealed that the rate of air injection and water flow speed dominate the efficiency of drag reduction by a mixture film. According to the results, the drag can be reduced by an average of 83.4%; and when the configuration of the parametric levels is optimum the maximum drag reduction of 88.5% is achieved. Copyright © 2008 John Wiley & Sons, Ltd.     

An investigation of possible mechanisms of heterogeneous drag reduction in pipe and channel flows

When concentrated polymer solutions are injected into the core-region of a turbulent pipe or channel flow, the injected polymer solution forms a thread which preserves its identity far beyond the injection point. The resulting drag reduction is called heterogeneous drag reduction.This study presents experimental results on the mechanism of this type of drag reduction. The experiments were carried out to find out whether this drag reduction is caused by small amounts of polymer removed from the thread and dissolved in the near-wall region of the flow or by an interaction of the polymer thread with the turbulence. The friction behavior of this type of drag reduction was measured for different concentrations in pipes of different cross-sections, but of identical hydraulic diameter. The parameters of the injection, i.e. injector geometry as well as the ratio of the injection to the bulk velocity, were varied. In one set of experiments the polymer thread was sucked out through an orifice and the friction behavior in the pipe was determined downstream of the orifice. In another experiment, near-wall fluid was led into a bypass in order to measure its drag reducing properties. Furthermore, the influence of a water injection into the near-wall region on the drag reduction was studied.The results provide a strong evidence that heterogeneous drag reduction is in part caused by small amount of dissolved polymer in the near-wall region as well as by an interaction of the polymer thread with the turbulence.Nomenclature a channel height - b channel width - c _p concentration of the injected polymer solution - c _R effective polymer concentration averaged over the cross-section - d pipe or hydraulic diameter - d _i injector diameter - DR drag reduction - f friction factor - l downstream distance from injector - L length of a pipe segment - P polymer type - p differential pressure - Re Reynolds number - U bulk velocity - u ^* ratio of injection to bulk velocity - y ⁺ dimensionless wall distance - v kinematic viscosity - density of the fluid - _w wall shear stress     

Skin friction reduction by large air bubbles in a horizontal channel flow

Microbubble and air film methods are believed to be applicable to skin friction reduction in ships. Small bubbles are dispersed into the turbulent boundary layer in the former case, and wide air layers cover the wall surface in the latter case. Previous studies did not specifically address the intermediate case between the microbubble and air film conditions. This study is concerned with the possibility and mechanism of drag reduction using relatively large air bubbles compared to the boundary layer thickness in a horizontal turbulent channel flow. The relationship between local skin friction and the bubble’s interfacial structure is investigated by synchronizing the measurement of wall-shear stress with the image acquisition of bubbles. The bubble sizes range from 2 to 90 mm approximately. As a result, a negative correlation between the local skin friction and the local void fraction is confirmed by the time-resolved measurement. A new observation is the fact that the local skin friction decreases drastically in the rear part of individual large bubbles, and rapidly increases after the bubble’s rear interface passes. This characteristic underlies the bubble-size dependency of the average skin friction in the intermediate bubble size condition.     

The stability of nonisothermal plasma flow in a flat MHD channel

Papers [1, 2] were devoted to questions of the stability of the laminar flow of a conducting fluid in a transverse magnetic field with Hartmann flow. It was assumed in these papers, however, that the transport coefficients are quantities independent of the flow characteristics; in particular, the temperature and the effect of energy dissipation were not taken into account. When these factors are allowed for it turns out that even for relatively small subsonic velocities, when the medium may be regarded as incompressible, the temperature distribution exerts a considerable influence on the dynamic flow characteristics. Papers [3,4] deal with this type of flow in an MHD channel which will be called nonisothermal in what follows. It has been shown that under specific conditions the velocity profiles are grossly deformed, and non-monotonic profiles with inflection points may even appear.However, the influence of nonisothermal flow on stability is not confined to an alteration of the stability criteria as a result of the change in the velocity profile. When energy dissipation and the fact that the transport coefficients are not constant are taken into account new dissipative instability branches appear, as, for example, the overheat instability [5, 8], This article considers the problem of the hydrodynamic stability of a nonisothermal plasma flow in constant crossed electric and magnetic fields in a flat channel with dielectric walls. The system of equations derived in this paper for the perturbations does, of course, take into account all the instability mechanisms mentioned above, but is difficult to solve. The general system of equations may be investigated in two limiting cases corresponding to the overheat and hydrodynamic instabilities.The author is most grateful to V. Kalitenko for writing the computer programs and to S. Filippov for advice and discussions.     

Influence of curvature on drag reduction by opposition control in turbulent flow along a thin cylinder

Opposition controlled fully developed turbulent flow along a thin cylinder is analyzed by means of direct numerical simulations. The influence of cylinder curvature on the skin-friction drag reduction effect by the classical opposition control (i.e., the radial velocity control) is investigated. The curvature of the cylinder affects the uncontrolled flow statistics; for instance, skin-friction coefficient increases while Reynolds shear stress (RSS) and turbulent intensity decrease. However, the control effect in the case of a small curvature is similar to that in channel flow. When the curvature is large, the maximum drag reduction rate decreased. However, the optimal location of the detection plane is the same as that in a flat plate. Further, the drag reduction effect is achieved even on a high detection plane where the drag increases in the flat plate. Although a difference in the drag reduction effect can be observed with a change in the curvature, its mechanism considered in this analysis based on the transport of the Reynolds stress is similar to that of the flat plate.