Three-dimensional motion of a thin vortex filament with axial velocity, embedded in an inviscid incompressible fluid, is investigated. The deflections of the core centreline are not restricted to be small compared with the core radius. We first derive the equation of the vortex motion, correct to the second order in the ratio of the core radius to that of curvature, by a matching procedure, which recovers the results obtained by Moore & Saffman (1972). An asymptotic formula for the linear dispersion relation is obtained up to the second order. Under the assumption of localized induction, the equation governing the self-induced motion of the vortex is reduced to a nonlinear evolution equation generalizing the localized induction equation. This new equation is equivalent to the Hirota equation which is integrable, including both the nonlinear Schrödinger equation and the modified KdV equation in certain limits. Therefore the new equation is also integrable and the soliton surface approach gives the N-soliton solution, which is identical to that of the localized induction equation if the pertinent dispersion relation is used. Among other exact solutions are a circular helix and a plane curve of Euler's elastica. This local model predicts that, owing to the existence of the axial flow, a certain class of helicoidal vortices become neutrally stable to any small perturbations. The non-local influence of the entire perturbed filament on the linear stability of a helicoidal vortex is explored with the help of the cutoff method valid to the second order, which extends the first-order scheme developed by Widnall (1972). The axial velocity is found to discriminate between right- and left-handed helices and the long-wave instability mode is found to disappear in a certain parameter range when the successive turns of the helix are not too close together. Comparison of the cutoff model with the local model reveals that the non-local induction and the core structure are crucial in making quantitative predictions.

This work concerns the combined oscillatory flow and current in a circular, smooth pipe. The study comprises wall shear stress measurements, and laser-Doppler-anemometer velocity and turbulence measurements. Three kinds of pipes were used, with diameters D=19 cm, 9 cm, and 1.1 cm, enabling the influence of the parameter R/δ to be studied in the investigation (R/δ ranging from about 3 to 53), where R is the radius of the pipe, and δ is the Stokes layer thickness. The ranges of the two other parameters of the combined flow processes, namely the current Reynolds number, Rec, and the oscillatory-flow boundary-layer (i.e. the wave–boundary layer) Reynolds number, Rew, are: Rec=0−1.6×105, and Rew=0−7×106. The transition to turbulence in the combined flow case occurs at a current Reynolds number larger than the conventional value, ca. 2×103, depending on Rew, and R/δ. A turbulent current can be laminarized by superimposing an oscillatory flow. The overall average value of the wall shear stress (the mean wall shear stress) may retain its steady-current value, it may decrease, or it may increase, depending on the flow regime. The increase (which can be as much as a factor of 4) occurs when the combined flow is in the wave-dominated regime, while the oscillatory-flow component of the flow is in the turbulent regime. The component of the wall shear stress oscillating around the mean wall shear stress can also increase with respect to its oscillatory-flow-alone value. For this to occur, the originally laminar oscillatory boundary layer needs to become a fully developed turbulent boundary layer, when a turbulent current is superimposed. This increase can be as much as O(3–4). The velocity profiles across the cross-section of the pipe change near the wall when an oscillatory flow is superimposed on a current, in agreement with the results of the wall shear stress measurements. The period-averaged turbulence profiles across the cross-section of the pipe behave differently for different flow regimes. When the two components of the flow are equally significant, the turbulence profile appears to be different from those corresponding to the fundamental cases; the level of turbulence increases (only slightly) with respect to those experienced in the fundamental cases.

A photographic study of the wakes of slugs rising in tubes of 19 mm and 52 mm internal diameter is presented. The dependence of the flow pattern in the wake upon the Reynolds number of the rising slug, R, is established for different slug lengths. Values of R covered in this study are in the range 25 to 1.3 × 104. For low values of R the flow pattern in the wake is laminar and axisymmetric and values of wake length and wake volume could be determined from the photographs: these values were correlated with the other variables in the system by means of dimensional analysis.

Direct numerical simulations (DNS) were carried out to investigate the laminar–turbulent transition for a flared cone at Mach 6 at zero angle of attack. The cone geometry of the flared cone experiments in the Boeing/AFOSR Mach 6 Quiet Tunnel (BAM6QT) at Purdue University was used for the simulations. In the linear regime, the largest integrated spatial growth rates ($N$-factors) for the primary instability were obtained for a frequency of approximately $f=300~\text{kHz}$. Low grid-resolution simulations were carried out in order to identify the azimuthal wavenumber that led to the strongest growth rates with respect to the secondary instability for a fundamental and subharmonic resonance scenario. It was found that for the BAM6QT conditions the fundamental resonance is much stronger compared to the subharmonic resonance. Subsequently, for the case which led to the strongest fundamental resonance onset, detailed investigations were carried out using high-resolution DNS. The simulation results exhibit streamwise streaks of very high skin friction and of high heat transfer at the cone surface. Streamwise ‘hot’ streaks on the flared cone surface were also observed in the experiments carried out at the BAM6QT facility using temperature sensitive paint. The presented findings provide strong evidence that the fundamental breakdown is a dominant and viable path to transition for the BAM6QT conditions.

We study finite-inertia effects on the collision rate of bidisperse heavy particles in a turbulent gas, using direct numerical simulations and kinematic descriptions. As shown previously for a monodisperse system (Sundaram & Collins 1997; Wang, Wexler & Zhou 2000), a statistical mechanical description of the average collision kernel consists of two parts, namely a description of the relative velocity between two colliding particles (the turbulent transport effect) and of the non-uniform particle distribution due to dynamic interaction of particles with coherent vortex structures (the accumulation effect). We first show that this description remains valid and accurate for a bidisperse system involving two groups of particles of inertial response time τp1 and τp2, respectively. Numerical results for the turbulent transport effect and the accumulation effect have been obtained as a function of τp1 and τp2. Interestingly, the accumulation effect in a bidisperse system is bounded above by that of a monodisperse system. An explanation for this observation is given, in terms of the correlation between concentration fields of the two size groups. Simulations show that particles from two size groups were found in different regions of a vortex, thus reducing the net accumulation effect in a bidisperse system. The turbulent transport effect, on the other hand, is bounded below by the level in a monodisperse system, due to a differential inertia effect. The above observations imply that the size polydispersity enhances the turbulent transport effect but weakens the accumulation effect, relative to a monodisperse system.

A simple eddy–particle interaction (EPI) model was developed and shown to give a reasonable prediction of the collision kernel, except for a small parametric region where both τp1 and τp2 are on the order of the ow Kolmogorov time τk and thus the accumulation effect must be included. A more accurate model incorporating both the turbulent transport effect and the accumulation effect has also been developed. The model would provide an upper bound on the collision rates for a non-dilute bidisperse system, since turbulence modulation and particle-particle interactions are not considered in this model.

Finally, some consideration is given to the effect of nonlinear drag on the collision kernel. The results show that the drag nonlinearity can increase the collision kernel slightly (less than 10%) at large particle inertia.

Thermoacoustic systems can oscillate self-excitedly, and often non-periodically, owing to coupling between unsteady heat release and acoustic waves. We study a slot-stabilized two-dimensional premixed flame in a duct via numerical simulations of a $G$-equation flame coupled with duct acoustics. We examine the bifurcations and routes to chaos for three control parameters: (i) the flame position in the duct, (ii) the length of the duct and (iii) the mean flow velocity. We observe period-1, period-2, quasi-periodic and chaotic oscillations. For certain parameter ranges, more than one stable state exists, so mode switching is possible. At intermediate times, the system is attracted to and repelled from unstable states, which are also identified. Two routes to chaos are established for this system: the period-doubling route and the Ruelle–Takens–Newhouse route. These are corroborated by analyses of the power spectra of the acoustic velocity. Instantaneous flame images reveal that the wrinkles on the flame surface and pinch-off of flame pockets are regular for periodic oscillations, while they are irregular and have multiple time and length scales for quasi-periodic and aperiodic oscillations. This study complements recent experiments by providing a reduced-order model of a system with approximately 5000 degrees of freedom that captures much of the elaborate nonlinear behaviour of ducted premixed flames observed in the laboratory.

Modern linear control theory has recently been established as a viable tool for developing effective, spatially localized convolution kernels for the feedback control and estimation of linearized Navier–Stokes systems. In the present paper, the effectiveness of these kernels for significantly expanding the basin of attraction of the laminar state in a subcritical nonlinear channel flow system is quantified using direct numerical simulations for a range of Reynolds numbers ($Re_{\CL}=2000, 3000 {\rm and} 5000$) and for a variety of initial conditions of physical interest. This is done by quantifying the change in the transition thresholds (see Reddy et al. 1998) when feedback control is applied. Such transition thresholds provide a relevant measure of performance for transition control strategies even in the nonlinear regime. Initial flow perturbations with streamwise vortices, oblique waves, and random excitations over an array of several Fourier modes are considered. It is shown that the minimum amplitude of these initial flow perturbations that is sufficient to excite nonlinear instability, and thereby promote transition to turbulence, is significantly increased by application of the control feedback. The kernels used to apply the feedback are found to decay exponentially with distance far from the origin, as predicted by the analysis of Bamieh, Paganini & Dahleh (2002). In the present paper, it is demonstrated via numerical simulation that truncation of these spatially localized convolution kernels to spatially compact kernels with finite non-zero support does not significantly degrade the effectiveness of the control feedback. In addition to the new state-feedback control results, exponential convergence of a localized physical-space state estimator with wall measurements is also demonstrated. The estimator and the full-state feedback controller are then combined to obtain a wall-information-based linear compensator. The compensator performance is also quantified, and key issues related to improving the performance of this compensator, which is degraded compared with the full-state feedback controller, are discussed.

We present results from an experimental investigation of turbulent transport and molecular mixing of a Sc [Gt ] 1 conserved scalar in the fully developed self-similar far field of a steady, axisymmetric, momentum-driven, free turbulent jet issuing into a quiescent medium. Our experiments cover the axial range from the jet exit to 350 diameters downstream, and span the range of Reynolds numbers from 1500 to 20000. Flow visualizations of the scalar concentration field directly verify the presence of an underlying characteristic large-scale organization in the jet far field essentially consistent with a simplified conceptual picture proposed in an earlier study (Dahm & Dimotakis 1987). High-resolution imaging measurements of successive instantaneous scalar concentration profiles in the jet support the presence of such a large-scale organization and provide details of its implications for mixing. These results also establish the proper similarity scaling for the mean concentration in the jet far field and give the scaling constant on the jet centreline as 5.4. We also present conserved scalar concentration p.d.f.s throughout the jet far field, and introduce a chemical reaction method for measuring the p.d.f.s with potentially molecular resolution. The amount of unmixed ambient fluid that reaches the jet centreline is found to decrease with increasing Reynolds number over the range investigated. The distribution of mixed fluid compositions in the concentration p.d.f. also appears to change over this range of Reynolds numbers, indicating that some aspects of large Schmidt number mixing in the jet far field have not yet become Reynolds number independent.

Experimental investigations of natural convection in a porous layer placed between two horizontal and isothermal plane surfaces have revealed a new type of convection as the Rayleigh number Ra* increases: fluctuating convection. A numerical study carried out on a two-dimensional model in order to simulate this phenomenon shows that, apart from the influence of the Rayleigh number, the aspect ratio A (length/height) of a vortical cell is the most important parameter for the occurrence of this type of convection. These quasi-periodic fluctuations induce important variations in the temperature field and in the streamlines. The total heat transport, as defined by the Nusselt number Nu*, varies within limits which may be separated by 80% of the mean value. Using the Galerkin method it is possible to deduce the conditions for the onset of convection from a state of pure conduction and also to define the critical conditions for the development of fluctuating convection from another perturbed state. A physical interpretation of the results is given for each type of convection. The results seem to agree with the experimental and numerical results obtained by different authors.

Based on theoretical analysis and laboratory data, we proposed a unified two-parameter wave spectral model as $\phi(n) = \frac{\beta g^2}{n^m n_0^{5-m}} {\rm exp} \left\{-\frac{m}{4}\left(\frac{n_0}{n}\right)^4\right\}$ with β and m as functions of the internal parameter, the significant slope η of the wave field which is defined as $\sect = \frac{(\overline{\zeta^2})^{\frac{}1{2}}}{\lambda_0},$ where $\overline{\zeta^2}$ is the mean squared surface elevation, and λ0, n0 are the wavelength and frequency of the waves at the spectral peak. This spectral model is independent of local wind. Because the spectral model depends only on internal parameters, it contains information about fluid-dynamical processes. For example, it maintains a variable bandwidth as a function of the significant slope which measures the nonlinearity of the wave field. And it also contains the exact total energy of the true spectrum. Comparisons of this spectral model with the JONSWAP model and field data show excellent agreements. Thus we established an alternative approach for spectral models. Future research efforts should concentrate on relating the internal parameters to the external environmental variables.

A new model of surface drift currents is constructed using the full nonlinear equations of motion. This model includes the balance between Coriolis forces due to the mean and wave-induced motions and the surface wind stresses. The approach used in the analysis is similar to the work by Craik & Leibovich (1976) and Leibovich (1977), but the emphasis is on the mean motion rather than the small-scale time-dependent part of the Langmuir circulation. The final result indicates that surface currents can be generated by both the direct wind stresses, as in the classical Ekman model, and the Stokes drift, derived from the surface wave motion, in an interrelated fashion depending on a wave Ekman number E defined as \[ E = \Omega/\nu_ek^2_0, \] where Ω is the angular velocity of the earth's rotation, νe, the eddy viscosity and k0, the wavenumber of the surface wave at the spectral peak. When E [Lt ] 1, the Langmuir mode dominates. When E [Gt ] 1, inertial motion results. The classical Ekman drift current is a special case even under the restriction E ≃ 1. On the basis of these results, a new model of the surface-layer movements for future large-scale ocean circulation studies is presented. For this new model both the wind stresses and the sea-state information are crucial inputs.

We consider the decay of an initial discontinuity in a polytropic gas in a Minkowski space–time (the special relativistic Riemann problem). In order to get a general analytical solution for this problem, we analyse the properties of the relativistic flow across shock waves and rarefactions. As in classical hydrodynamics, the solution of the Riemann problem is found by solving an implicit algebraic equation which gives the pressure in the intermediate states. The solution presented here contains as a particular case the special relativistic shock-tube problem in which the gas is initially at rest. Finally, we discuss the impact of this result on the development of high-resolution shock-capturing numerical codes to solve the equations of relativistic hydrodynamics.

The reflection of shock waves over straight reflecting surfaces in steady flows was investigated experimentally using the supersonic wind tunnel of Laboratoire d'Aerothermique du CNRS, Meudon, France. The results for a flow Mach number M0 = 4.96 contradict the state of the art regarding the regular [harr ] Mach reflection transition in steady flows. Not only was a hysteresis found to exist in this transition, but, unlike previous reports, regular reflection configurations were found to be stable in the dual-solution domain in which theoretically both regular and Mach reflection are possible.

The paper presents a model for particle–particle and particle–wall collisions during interface-resolving numerical simulations of particle-laden flows. The accurate modelling of collisions in this framework is challenging due to methodological problems generated by interface approach and contact as well as due to the greatly different time scales involved. To cope with this situation, multiscale modelling approaches are introduced avoiding excessive local grid refinement during surface approach and time step reduction during the surface contact. A new adaptive model for the normal forces in the phase of ‘dry contact’ is proposed, stretching the collision process in time to match the time step of the fluid solver. This yields a physically sound and robust collision model with modified stiffness and damping determined by an optimization scheme. Furthermore, the model is supplemented with a new approach for modelling the tangential force during oblique collisions which is based on two material parameters: a critical impact angle separating rolling from sliding and the friction coefficient for the sliding motion. The resulting new model is termed the adaptive collision model (ACM). All proposed sub-models only contain physical parameters, and virtually no numerical parameters requiring adjustment or tuning. The new model is implemented in the framework of an immersed boundary method but is applicable with any spatial and temporal discretization. Detailed validation against experimental data was performed so that a general and versatile model for arbitrary collisions of spherical particles in viscous fluids is now available.

Experiments to investigate the formation mechanism and frequency of vortex shedding from a sphere in uniform shear flow were conducted in a water channel using flow visualization and velocity measurement. The Reynolds number, defined in terms of the sphere diameter and approach velocity at its centre, ranged from 200 to 3000. The shear parameter K, defined as the transverse velocity gradient of the shear flow non-dimensionalized by the above two parameters, was varied from 0 to 0.25. The critical Reynolds number beyond which vortex shedding from the sphere occurred was found to be lower than that for uniform flow and decreased approximately linearly with increasing shear parameter. Also, the Strouhal number of the hairpin-shaped vortex loops became larger than that for uniform flow and increased as the shear parameter increased.

The formation mechanism and the structure of vortex shedding were examined on the basis of series of photographs and subsequent image processing using computer graphics. The range of Reynolds number in the present investigation, extending up to 3000, could be classified into three regions on the basis of this study, and it was observed that the wake configuration did not differ substantially from that for uniform flow. Also, unlike the detachment point of vortex loops in uniform flow, which was irregularly located along the circumference of the sphere, the detachment point in shear flow was always on the high-velocity side.

A transition scenario initiated by streamwise low- and high-speed streaks in a flat-plate boundary layer is studied. In many shear flows, the perturbations that show the highest potential for transient energy amplification consist of streamwise-aligned vortices. Due to the lift-up mechanism these optimal disturbances lead to elongated streamwise streaks downstream, with significant spanwise modulation. In a previous investigation (Andersson et al. 2001), the stability of these streaks in a zero-pressure-gradient boundary layer was studied by means of Floquet theory and numerical simulations. The sinuous instability mode was found to be the most dangerous disturbance. We present here the first simulation of the breakdown to turbulence originating from the sinuous instability of streamwise streaks. The main structures observed during the transition process consist of elongated quasi-streamwise vortices located on the flanks of the low-speed streak. Vortices of alternating sign are overlapping in the streamwise direction in a staggered pattern. The present scenario is compared with transition initiated by Tollmien–Schlichting waves and their secondary instability and by-pass transition initiated by a pair of oblique waves. The relevance of this scenario to transition induced by free-stream turbulence is also discussed.

The dynamics of a thin film of Newtonian fluid coating the inner surface of an elastic circular tube is analysed. This problem is motivated by an interest in the closure of small airways of the lungs either by formation of a liquid bridge, the collapse of the airway wall or a combination of both processes. Liquid bridge formation is due to the destabilization of the liquid film that coats the inner surface of airways, while wall collapse can be due to either the high surface tension of the air–liquid interface or the flexibility of the wall.

Nonlinear evolution equations for the film thickness and wall position are derived using lubrication theory, but an accurate representation of the curvatures of both the liquid and wall interfaces is employed which is valid for thick films. These approximations allow closure to be predicted. In addition, these approximations are justified by comparison with rigid-wall results obtained by solving the full Navier–Stokes equations and because fluid inertia only becomes important in the very late stages of closure. The linear stability of these equations is examined using normal-mode analysis for infinitesimal disturbances and the nonlinear stability is investigated by solving the governing equations numerically using the method of lines. Solutions show that there is a critical film thickness, strongly dependent on fluid and wall properties, above which unstable waves grow to form liquid bridges. The critical film thickness decreases with increasing surface tension or wall compliance since waves grow faster. Even for relatively stiff airways, the volume of fluid in the liquid lining required for closure can be approximately 70% of the volume for the rigid-tube case. Wall damping is an important effect only when the airway is sufficiently compliant. Airway closure occurs more rapidly with increasing unperturbed film thickness, surface tension and wall flexibility and decreasing wall damping.

We present an experimental study of drop impact on a solid surface in the spreading regime with no splashing. Using the space–time-resolved Fourier transform profilometry technique, we can follow the evolution of the drop shape during the impact. We show that a self-similar dynamical regime drives the drop spreading until the growth of a viscous boundary layer from the substrate selects a residual minimal film thickness. Finally, we discuss the interplay between capillary and viscous effects in the spreading dynamics, which suggests a pertinent impact parameter.

Convection in water above ice penetrates into the stably stratified region above the density maximum at 4 °C. Two-dimensional penetrative convection in a Boussinesq fluid confined between free boundaries has been studied in a series of numerical experiments. These included cases with a constant temperature at both boundaries as well as cases with a fixed average flux at the lower boundary. Steady convection occurs at Rayleigh numbers below the critical value predicted by linear theory. At high Rayleigh numbers, resonant coupling between convection and gravitational modes in the stable layer excites finite amplitude oscillations. The problem can be described by a simplified model which allows for distortion of the mean temperature profile and balances the convected and conducted flux. This model explains the finite amplitude instability and predicts the Nusselt number as a function of Rayleigh number. These predictions are in excellent agreement with the computed results.

An erodible surface exposed to supercritical flow often devolves into a series of steps that migrate slowly upstream. Each step delineates a headcut with an associated hydraulic jump. These steps can form in a bed of cohesive material which, once eroded, is carried downstream as washload without redeposition. Here the case of purely erosional, one-dimensional periodic, or cyclic steps in cohesive material is considered. The St. Venant shallow-water equations combined with a formulation for sediment erosion are used to construct a complete theory of the erosional case. The solution allows wavelength, wave height, migration speed and bed and water surface profiles to be determined as functions of imposed parameters. The analysis also admits a solution for a solitary step, or single headcut of self-preserving form.