A new state-of-the-art CFD solver capable of simulating a broad range of complex engineering flows at real-life Reynolds numbers is developed. The method solves the three-dimensional incompressible unsteady Reynolds-averaged Navier-Stokes (URANS) equations closed with statistical turbulence models. Three such models are incorporated in the solver: the standard k - e model with wall functions, the Spalart-Allmaras model and the detached-eddy simulation (DES) model. The numerical solver employs domain decomposition with structured Chimera overset grids to handle complex, multi-connected geometries. The governing equations are discretized with second order accuracy schemes both in space and time. The capabilities and versatility of the numerical method are demonstrated by applying it to simulate two widely different flow problems: a) flow past a geometrical complex array of multiple bridge piers mounted both on a natural river reach and on a flat bed experimental flume; and b) flow in mechanical, bileaflet, prosthetic heart valve with the leaflets fixed in the fully-open position. Overset grid systems with several millions of grid nodes are used and grid-refinement and other numerical dependency studies are carried out to explore the sensitivity of the computed solutions to various numerical parameters. For all simulated cases, large-scale unsteadiness appears naturally as a result of excited mean-flow instabilities and the computed mean flowfields are shown to be in good quantitative agreement with experimental measurements. By analyzing the instantaneous flowfields numerous novel insights into the physics of both flow cases are obtained and discussed extensively. The results of this thesis demonstrate the potential of the new method as a powerful simulation tool for a broad range of cross-disciplinary engineering flow problems and underscore the need for physics-based numerical modeling by integrating CFD with laboratory experimentation.

We performed large eddy simulation of the flow over a series of two- and three-dimensional dune geometries at laboratory scale using the Lagrangian dynamic eddy-viscosity subgrid-scale model. First, we studied the flow over a standard 2D transverse dune geometry, then bedform three-dimensionality was imposed. Finally, we investigated the turbulent flow over barchan dunes. The results are validated by comparison with simulations and experiments for the 2D dune case, while the results of the 3D dunes are validated qualitatively against experiments. The flow over transverse dunes separates at the dune crest, generating a shear layer that plays a crucial role in the transport of momentum and energy, as well as the generation of coherent structures. Spanwise vortices are generated in the separated shear; as they are advected, they undergo lateral instabilities and develop into horseshoe-like structures and finally reach the surface. The ejection that occurs between the legs of the vortex creates the upwelling and downdrafting events on the free surface known as “boils”. The three-dimensional separation of flow at the crestline alters the distribution of wall pressure, which may cause secondary flow across the stream. The mean flow is characterized by a pair of counter-rotating streamwise vortices, with core radii of the order of the flow depth. Staggering the crestlines alters the secondary motion; two pairs of streamwise vortices appear (a strong one, centred about the lobe, and a weaker one, coming from the previous dune, centred around the saddle). The flow over barchan dunes presents significant differences to that over transverse dunes. The flow near the bed, upstream of the dune, diverges from the centerline plane; the flow close to the centerline plane separates at the crest and reattaches on the bed. Away from the centerline plane and along the horns, flow separation occurs intermittently. The flow in the separation bubble is routed towards the horns and leaves the dune at the tips. Barchan dunes induce two counter-rotating streamwise vortices, along each of the horns, which direct high-momentum fluid toward the symmetry plane and low-momentum fluid near the bed away from the centerline. / Thesis (Ph.D, Mechanical and Materials Engineering) -- Queen's University, 2013-05-27 18:58:48.969

No description available.

In this study, application of computational fluid dynamics to ground vehicle aerodynamics was investigated. Two types of vehicle models namely, Ahmed Body and MIRA Notchback Body and their scaled models were used. A commercial software &quot / Fluent&quot / was used and the effects of implementing different turbulence models with wall functions were observed. As a result, an appropriate turbulence model was selected to use in the study. The drag forces, surface pressure distributions and wake formations were investigated in simulation of various test cases available in the literature. The study was extended to simulate the aerodynamics of the vehicles in close-following situation. The results were then compared with available wind tunnel test data.

published_or_final_version / Mechanical Engineering / Master / Master of Philosophy

Many free surface flow problems encountered in hydraulic engineering can be accurately analyzed by utilizing the depth-averaged equations of motion. A consequence of adopting this depth-averaged modeling approach is that closure approximations must be implemented to represent the so-called effective stresses. These effective stresses consist of the depth-averaged viscous stresses, which are usually small and therefore neglected, the depth-averaged turbulent Reynold's stresses, and additional stresses resulting from depth-averaging of the nonlinear 'convective acceleration terms (often called momentum dispersion terms). Attention is focused on examining closure for both the depth-averaged Reynold's stresses and the momentum dispersion terms. In the present study, the penalty function finite element technique is utilized to solve the governing hydrodynamic and turbulence model equations for a variety of flow domains. Alternative momentum dispersion and turbulence closure models are proposed and evaluated by comparing model predictions with experimental data for strongly curved open channel flow. The results of these simulations indicate that the depth-averaged (k-ε) turbulence model yields excellent agreement with experimental observations. In addition, it appears that neither the streamline curvature modification of the depth-averaged (k-ε) model, nor the momentum dispersion models based on the assumption of helicoidal flow in a curved channel, yield significant improvement in model predictions. Overall model predictions are found to be as good as those of a more complex and restricted three dimensional model. / Ph. D.

A numerical study of fundamental aspects of turbulent mixing has been performed,with emphasis on the behavior of passive scalars of low molecular diffusivity (high Schmidt number Sc). Direct Numerical Simulation is used to simulate incompressible, stationary and isotropic turbulence carried out at high grid resolution. Data analyses are carried out by separate parallel codes using up to 1024^3 grid points for Taylor-scale Reynolds number (R_lambda) up to 390 and Sc up to 1024.Schmidt number of order 1000 is simulated using a double-precision parallel code in a turbulent flow at a low Reynolds number of R_lambda 8 to reduce computational cost to achievable level. The results on the scalar spectrum at high Schmidt numbers appear to have a k^{-1} scaling range. In the presence of a uniform mean scalar gradient, statistics of scalar gradients are observed to deviate substantially from Kolmogorov's hypothesis of local isotropy, with a skewness factor remaining at order unity as the Reynolds number increases. However, this skewness decreases with Schmidt number suggesting that local isotropy for scalars at high Schmidt number is a better approximation. Intermittency exponents manifested by three types of two-point statistics of energy and scalar dissipation, i.e., the two-point correlator (chi(x)chi(x+r)), the second-order moment of local scalar dissipation (chi_r^2) and the variance of the logarithmic local scalar dissipation sigma^2_{lnchi_r} are discussed. Several basic issues in differential diffusion between two scalars of different molecular diffusivities transported by the same turbule nt flow, the physical process of scalar spectral transfer and subgrid-scale transfer are also briefly addressed.

No description available.

Most flows in nature and engineering are turbulent, and many are wall-bounded. Further, in turbulent flows, the turbulence generally has a large impact on the behavior of the flow. It is therefore important to be able to predict the effects of turbulence in such flows. The Navier-Stokes equations are known to be an excellent model of the turbulence phenomenon. In simple geometries and low Reynolds numbers, very accurate numerical solutions of the Navier-Stokes equations (direct numerical simulation, or DNS) have been used to study the details of turbulent flows. However, DNS of high Reynolds number turbulent flows in complex geometries is impractical because of the escalation of computational cost with Reynolds number, due to the increasing range of spatial and temporal scales. In Large Eddy Simulation (LES), only the large-scale turbulence is simulated, while the effects of the small scales are modeled (subgrid models). LES therefore reduces computational expense, allowing flows of higher Reynolds number and more complexity to be simulated. However, this is at the cost of the subgrid modeling problem. The goal of the current research is then to develop new subgrid models consistent with the statistical properties of turbulence. The modeling approach pursued here is that of "Optimal LES". Optimal LES is a framework for constructing models with minimum error relative to an ideal LES model. The multi-point statistics used as input to the optimal LES procedure can be gathered from DNS of the same flow. However, for an optimal LES to be truly predictive, we must free ourselves from dependence on existing DNS data. We have done this by obtaining the required statistics from theoretical models which we have developed. We derived a theoretical model for the three-point third-order velocity correlation for homogeneous, isotropic turbulence in the inertial range. This model is shown be a good representation of DNS data, and it is used to construct optimal quadratic subgrid models for LES of forced isotropic turbulence with results which agree well with theory and DNS. The model can also be filtered to determine the filtered two-point third-order correlation, which describes energy transfer among filtered (large) scales in LES. LES of wall-bounded flows with unresolved wall layers commonly exhibit good prediction of mean velocities and significant over-prediction of streamwise component energies in the near-wall region. We developed improved models for the nonlinear term in the filtered Navier-Stokes equation which result in better predicted streamwise component energies. These models involve (1) Reynolds decomposition of the nonlinear term and (2) evaluation of the pressure term, which removes the divergent part of the nonlinear models. These considerations significantly improved the performance of our optimal models, and we expect them to apply to other subgrid models as well. / text

Magnetohydrodynamics describes the motions of an electrically conducting fluid under the influence of magnetic fields. Such flows are encountered in a large variety of applications, from steel industry to heat exchangers of nuclear fusion reactors. <p><p>Here we are concerned with situations where the magnetic field is relatively strong and the flow manifests turbulent motions. The interaction of the fluid with the electromagnetic field is still insufficiently understood and efficient predicting methods are lacking. Our goal is to provide more insight on this problem by making heavy use of numerical methods. In this work, two different classes of problem are investigated. <p><p>First we consider that the turbulent character of the fluid is well developed and that solid boundaries are sufficiently far away to be completely neglected. The main effects of a strong magnetic field in that case are to damp the motion and to homogenize the flow along its direction, leading to a quasi two dimensional state. Using numerical simulations we have studied the dynamics of the flow in Fourier space and in particular the non linear energy transfers between turbulent eddies. Further we investigated the scale-by-scale anisotropy and compared various methods to address this quantity. Finally, the evolution of a passive scalar embedded in the flow was analyzed and it turned out that the characteristic anisotropy of the velocity field is reflected in the distribution of the scalar quantity. <p><p>In the second problem, the flow in a duct of square cross section subject to a transverse magnetic field has been considered. Here, unlike in the previous situation, the magnetic field has globally a destabilizing effect on the flow, because of the strong inhomogeneities it produces. For instance, high velocity regions develop along the walls that are parallel to the magnetic field. There, we are mostly interested in the possible development of persistent time-dependent fluctuations. It is observed that the transition between laminar and turbulent regimes occurs through at least two distinct bifurcations. The first one takes place at moderate Reynolds number and is characterized by highly organized fluctuations. The second is encountered at higher Reynolds number and presents very strong and localized disturbances.<p>/Il existe un grand nombre d'applications industrielles dans lesquelles un écoulement de métal liquide est soumis à un champ magnétique. La production d'acier par coulée continue, la fabrication de matériaux semi-conducteurs ou encore les échan-geurs de chaleur des futurs réacteurs à fusion nucléaire en sont de bons exemples. L'interaction du liquide conducteur avec le champ magnétique est à l'origine de nombreux phénomènes inhabituels en hydrodynamique classique et doit dès lors être décrite par la magnétohydrodynamique (ou MHD en abrégé). Le but de ce travail est d'étudier la physique de ces interactions, en se basant sur la résolution numérique des équations qui les gouvernent.<p><p>Plusieurs aspects du problème ont été considérés indépendamment. Tout d'abord, l'étude de la turbulence homogène a permis de mettre en evidence les comportements du fluide loin de toute paroi solide. Ceci est mis un oeuvre dans un domaine spatial périodique, où les variables sont représentées par leur série de Fourier. L'influence du champ magnétique dans ce cas consiste à dissiper les fluctuations turbulentes et à rendre le champ de vitesse anisotrope. Les principaux résultats obtenus dans ce cadre concernent la distribution ainsi que le transfert d'énergie dans l'espace spectral, l'anisotropie des différentes échelles turbulentes de l'écoulement ainsi que le transport d'un scalaire passif au sein du fluide. <p><p>Dans un deuxième temps, le travail a porté sur l'écoulement dans une conduite rectangulaire soumise à un champ magnétique et dont les parois sont conductrices d'électricité. La particularité de cet écoulement réside dans les zones de vitesse élevées qui se développent le long des parois parallèles au champ magnétique. Celles-ci donnent lieu à un intense cisaillement qui a généralement pour effet de rendre l'écoulement instable. La simulation numérique de ce problème a permis l'étude des instabilités au sein du fluide et de la transition du régime laminaire vers la turbulence. <p> / Doctorat en Sciences / info:eu-repo/semantics/nonPublished