In the last 30 days

• Open/close Lift generation by a two-dimensional symmetric flapping wing: immersed boundary-lattice Boltzmann simulations

  Keigo Ota et al 2012 Fluid Dyn. Res. 44 045504 Tag this article

  Article References Full text PDF (2.96 MB) Enhanced article HTML

  Two-dimensional (2D) symmetric flapping flight is investigated by an immersed boundary-lattice Boltzmann method (IB-LBM). In this method, we can treat the moving boundary problem efficiently on the Cartesian grid. We consider a model consisting of 2D symmetric flapping wings without mass connected by a hinge with mass. Firstly, we investigate the effect of the Reynolds number in the range of 40–200 on flows around symmetric flapping wings under no gravity field and find that for high Reynolds numbers ( Re ⩾ 55), asymmetric vortices with respect to the horizontal line appear and the time-averaged lift force is induced on the wings, whereas for low Reynolds numbers ( Re ⩽ 50), only symmetric vortices appear around the wings and no lift force is induced. Secondly, the effect of the initial position of the wings is investigated, and the range of the initial phases where the upward flight is possible is found. The effects of the mass and flapping amplitude are also studied. Finally, we carry out free flight simulations under gravity field for various Reynolds numbers in the range 60 ⩽  Re ⩽ 300 and Froude numbers in the range 3 ⩽  Fr ⩽ 60 and identify the region where upward flight is possible.

• Open/close Nonlinear streak computation using boundary region equations

  J A Martín and C Martel 2012 Fluid Dyn. Res. 44 045503 Tag this article

  Article References Full text PDF (699 KB) Enhanced article HTML

  The boundary region equations (BREs) are applied for the simulation of the nonlinear evolution of a spanwise periodic array of streaks in a flat plate boundary layer. The well-known BRE formulation is obtained from the complete Navier–Stokes equations in the high Reynolds number limit, and provides the correct asymptotic description of three-dimensional boundary layer streaks. In this paper, a fast and robust streamwise marching scheme is introduced to perform their numerical integration. Typical streak computations present in the literature correspond to linear streaks or to small-amplitude nonlinear streaks computed using direct numerical simulation (DNS) or the nonlinear parabolized stability equations (PSEs). We use the BREs to numerically compute high-amplitude streaks, a method which requires much lower computational effort than DNS and does not have the consistency and convergence problems of the PSE. It is found that the flow configuration changes substantially as the amplitude of the streaks grows and the nonlinear effects come into play. The transversal motion (in the wall normal-streamwise plane) becomes more important and strongly distorts the streamwise velocity profiles, which end up being quite different from those of the linear case. We analyze in detail the resulting flow patterns for the nonlinearly saturated streaks and compare them with available experimental results.

• Open/close Motion of droplets under shear flow with suction

  Masato Makino and Osamu Sano 2012 Fluid Dyn. Res. 44 045501 Tag this article

  Article References Full text PDF (540 KB) Enhanced article HTML

  The motion of a droplet through an incompressible viscous fluid is investigated on the basis of the Stokes approximation. A numerical study of the transportation of the droplet was carried out using the two-dimensional boundary element method. The system is composed of the shear flow region and the Poiseuille flow region. The droplet moves in the shear flow and is sucked into the Poiseuille flow region if the suction pressure is stronger than the shear flow. The criterion for the suction is investigated. The dependences on the surface tension, the ratio of the inside–outside viscosities of the droplet, magnitudes of shear rate and suction pressure gradient and the initial position of the droplet are elucidated.

• Open/close The finite-difference lattice Boltzmann method and its application in computational aero-acoustics

  Michihisa Tsutahara 2012 Fluid Dyn. Res. 44 045507 Tag this article

  Article References Full text PDF (797 KB) Enhanced article HTML

  The application of the finite-difference lattice Boltzmann method in computational aero-acoustics is reviewed, mainly on the basis of the work of the author and his colleagues. Some models of thermal and isothermal fluids are described and the constraints for recovering the Euler equations and the Navier–Stokes equations are described. The arbitrary Lagrangian Eulerian technique is used for high Mach number flows and for simulations of moving bodies. A model of gas–liquid two-phase fluid is introduced in which the density difference is 800 times and the sound velocity difference is 4 times. Some applications of aero-acoustic problems are briefly described and the simultaneous simulation of underwater sound and sound propagating in air is also presented. The difference between the thermal model and the isothermal model is shown in the aero-acoustic problems.

• Open/close Lattice Boltzmann methods for moving boundary flows

  Takaji Inamuro 2012 Fluid Dyn. Res. 44 024001 Tag this article

  Article References Full text PDF (1.62 MB) Enhanced article HTML

  The lattice Boltzmann methods (LBMs) for moving boundary flows are presented. The LBM for two-phase fluid flows with the same density and the LBM combined with the immersed boundary method are described. In addition, the LBM on a moving multi-block grid is explained. Three numerical examples (a droplet moving in a constricted tube, the lift generation of a flapping wing and the sedimentation of an elliptical cylinder) are shown in order to demonstrate the applicability of the LBMs to moving boundary problems.

• Open/close Collision of a vortex ring on granular material. Part I. Interaction of the vortex ring with the granular layer

  Naoya Masuda et al 2012 Fluid Dyn. Res. 44 015501 Tag this article

  Article References Full text PDF (4.93 MB) Enhanced article HTML

  An experimental study was carried out on the normal impact of the vortex ring on the granular layer. The Reynolds number of the vortex ring ranged from 1000 to 6000, whereas the distance to the granular layer (glass beads of diameter 0.10 mm) varied between 2 and 13 times of the diameter of the vortex ring generator nozzle. The velocity field was visualized by the sodium fluorescein and was measured by particle image velocimetry or particle tracking velocimetry. The flow field in the vicinity of the granular surface was also visualized by an almost mono-layer of fine particles deposited on the main granular layer, which elucidated the development of the secondary and tertiary vortex rings. Generally speaking, the characteristics of the granular layer are found to be similar to a liquid plane for the impact of a vortex ring of a smaller impulse, whereas it is closer to a solid plane for a vortex ring of a larger impulse. For the vortex ring with a much larger impulse, the granular layer is engraved, whose patterns reflect the deformation of the primary and/or secondary vortex ring due to the interaction with the granular layer (as will be described in part II).

• Open/close Viscous propagation of two-dimensional non-Newtonian gravity currents

  M R Chowdhury and F Y Testik 2012 Fluid Dyn. Res. 44 045502 Tag this article

  Article References Full text PDF (740 KB) Enhanced article HTML

  This paper presents the results of a detailed experimental and theoretical investigation on the viscous propagation of non-Newtonian gravity currents. Laboratory gravity currents are generated in a horizontal rectangular tank by releasing a constant flux of high-concentration fluid mud suspensions that exhibit profound non-Newtonian (shear thinning) behavior. Experimental observations on the propagation of fluid mud gravity currents revealed that viscous propagation of these currents was typically preceded by two phases as expected: an initial momentum-driven horizontal buoyant wall jet and a buoyancy-driven inertial phase. The experimental transition times, t ^**, and positions, x ^**, at which fluid mud gravity currents transition into viscous propagation phase were determined. The experimental data that correspond to the viscous propagation of fluid mud gravity currents (i.e. experimental time, t ⩾  t ^**, and front position, x _N  ⩾  x ^**) were used to evaluate the predictive capabilities of two well-known mathematical modeling approaches: the lubrication theory approximation and the box-model approaches. Regarding the lubrication theory approximation, a recently developed self-similarity solution for viscous propagation of power-law gravity currents that has not been experimentally evaluated was used. Regarding the box-model approach, a viscous box-model solution for two-dimensional (2D) non-Newtonian gravity currents was developed. The evaluation of these models using experimental data revealed that both models were in good agreement with the experimental observations, despite several simplifying assumptions embedded in each. Given its more advanced mathematical development, the lubrication theory approximation model provides a more complete description of a gravity current (i.e. shape and velocity variation along the gravity current) than the box model at the expense of a relatively simple computational effort.

• Open/close The flow in a divergent channel driven by stretching the inclined walls

  M B Zaturska and W H H Banks 2012 Fluid Dyn. Res. 44 045505 Tag this article

  Article References Full text PDF (625 KB) Enhanced article HTML

  We consider the steady two-dimensional flow of an incompressible viscous fluid between the walls of a divergent channel; the walls are assumed to be stretching, and further, they meet at an angle 2 α, where 0 <  α ⩽  π. A similarity solution is sought and a Reynolds number, R, based on the motion of the walls is introduced. The apparently simple case when R = 0 is discussed but its relevance is only found later. For general R the form of the solution results in a cubic equation with the discriminant Δ. Three cases arise: (i) Δ = 0, (ii) Δ < 0 and (iii) Δ > 0. It is found that in the case (i) there is a one-parameter family of solutions. In the cases (ii) and (iii) solutions are found in terms of Jacobian elliptic functions with parameter m. In the case (ii) for each value of m a one-parameter family of solutions is again found. In case (iii) there are multiple solutions in the interval 0 <  α ⩽  π. Furthermore, from a small- m analysis it is found that the solution in the case when R = 0 is dependent on the nonlinear inertia terms. All the solutions found are symmetric.

• Open/close Coherent structure statistics in the wake of a sharp-edged bluff body placed vertically in a shallow channel

  Arindam Singha and Ram Balachandar 2011 Fluid Dyn. Res. 43 055504 Tag this article

  Article References Full text PDF (2.39 MB)

  Proper orthogonal decomposition and closed-streamline-based vortex identification algorithm has been applied to extract information on coherent structures downstream of a sharp-edged bluff body, placed vertically in a shallow open channel flow. The friction effect of the bed was found to reduce the size and strength of the coherent structures close to the bed. The free surface weakly decreases the size and strength of the prevalent coherent structures. The population trends of coherent structure statistics with downstream distance indicate the phenomenon of merging and spatial growth of coherent structures with time and space. But the rates of merging and growth were found to be different at different vertical locations over the flow depth. The distribution functions of the location, size and strength in the transverse direction were examined to shed light on the dynamics of these structures.

• Open/close Remodeling of vascular endothelial cells exposed to fluid shear stress: experimental and numerical approach

  Toshiro Ohashi and Masaaki Sato 2005 Fluid Dyn. Res. 37 40 Tag this article

  Article References Full text PDF (1.02 MB)

  Vascular endothelial cells are an interface between blood vessel walls and blood flow, and play important roles in physiological functions. Since endothelial cell responses to fluid shear stress have been implicated in the localization of atherosclerosis, the effect of shear stress on endothelial cell morphology and functions has been exclusively studied. After applying fluid shear stress, cultured endothelial cells show marked elongation and orientation in the flow direction. In addition, thick stress fibers of actin filaments appear and align along the cell long axis. Thus, the endothelial cell morphology is closely related to the cytoskeletal structure. The purpose of this review is to summarize endothelial cell responses to fluid flow which have been studied, focusing on the changes in cell shape and cytoskeletal structure. Numerical studies to simulate local flow field at the cellular level and the resulting intracellular stresses are also reviewed.