No description available.

Turbulent flow inside a street canyon was investigated in an open circuit wind tunnel and in a blow-down wind channel. Two geometries were used for comparison purposes: buildings with pitched roofs and with flat roofs. Both generate the flow of a different category, so the induced ventilation regimes are fundamentally different. Quadrant, Fourier and Wavelet analysis, Proper Orthogonal Decomposition (POD) and vortex detection methods are used to identify coherent structures in the flow and establish their impact on the ventilation of pollution. Two types of the organised motions are detected: the compact areas of sweep and ejection with the scale comparable to the size of building and the small vortices generated in the shear layer behind the building roof. POD identifies the most dominant modes with high coherency in the flow and evaluates the relative contributions of each mode to the overall kinetic energy of turbulence. Rigorous analysis of the correctness of the physical interpretation for such a decomposition is carried out. Wavelet analysis is applied to the time-series of the POD expansion coefficients in order to reveal control mechanism of the dynamics of the modes. Vorticity, calculated from the original velocity data, is decomposed by POD as well. Finally, the correlation between the vorticity...

Cette thèse vise à démêler les principaux mécanismes impliqués dans les écoulements transitoires et turbulents. L’idée centrale est d'utiliser une technique d’optimisation non linéaire pour étudier l’origine et le rôle des structures cohérentes habituellement observées dans ces écoulements. Cette méthode a été utilisée dans trois contextes différents. Tout d’abord, un écoulement laminaire linéairement stable a été considéré et l'optimisation a été utilisée pour calculer les perturbations les plus amplifiées parmi toutes les perturbations capables de déclencher une transition vers la turbulence. Une fois que la turbulence est bien établie, une optimisation non linéaire entièrement 3D maximisant l'énergie cinétique turbulente est utilisée pour étudier les structures cohérentes qui peuplent l’écoulement turbulent et les mécanismes responsables de la croissance et de l’échange d’énergie (optimale) sont étudiés. Ensuite, une approche de type système dynamique est appliquée aux équations du mouvement. La géométrie de l’espace des phases est étudiée en utilisant la théorie de la croissance transitoire pour évaluer l’importance des variétés stable et instable dans la dynamique. Dans le même cadre, un algorithme de minimisation non linéaire est utilisé pour calculer les connexions hétérocliniques parmi les solutions invariantes des équations de Navier-Stokes. / This thesis aims at unraveling the main mechanisms involved in transitional and turbulent flows. The central idea is that of using a nonlinear optimization technique to investigate the origin and role of coherent structures usually observed in these flows. This method has been used in three different contexts. First, a linearly stable laminar flow has been considered and the optimization has been used to compute the most amplified perturbations among all disturbances able to trigger transition to turbulence. Once turbulence is well established, a fully 3D nonlinear optimization maximizing the turbulent kinetic energy is used to study coherent structures populating turbulent shear flow as well as investigate the mechanisms responsible for the energy (optimally) growth and exchange. Then, a dynamical system approach is applied to fluid flow equations. The geometry of the state space is investigated by using transient growth theory to reveal the importance of the stable and unstable manifold. In the same framework, a nonlinear minimization algorithm is used to compute heteroclinic connections among invariant solutions of the Navier-Stokes equations.

Le but de ce travail est d'étudier une couche limite soumise à un gradient de pression et de la comparer avec une couche limite de plaque plane à grands nombres de Reynolds. Dans ce cadre, l’accent est mis sur le comportement des structures cohérentes à grande échelle. En raison de leur grande longueur, ces structures ne sont pas faciles à extraire et à caractériser en utilisant des techniques de mesure standard. Pour cette raison, des dispositifs expérimentaux spécifiques utilisant la PIV dans les plans longitudinaux et parallèles à la paroi ont été conçus pour capturer les structures à grande échelle et pour mieux comprendre les mécanismes régissant la dynamique de ces écoulements. La première partie revisite les résultats obtenus sur une couche limite plaque plane en sondant l'origine d'une décroissance spectrale en dans la couche limite turbulente. Dans cette perspective, un modèle simpliste basé sur le modèle de Townsend-Perry est proposé. Ce modèle peut, en principe, être appliqué à n'importe quel écoulement turbulent de paroi. La deuxième partie se concentre sur l'amélioration de la compréhension de la turbulence en gradient de pression adverse en effectuant une caractérisation complète d’un écoulement académique au dessus d’une géométrie. L’accent est mis sur les caractéristiques des structures (longueur, scaling, contribution énergétique et distribution selon la normale à la paroi) ainsi que sur l'influence du gradient de pression adverse sur les structures des grandes échelles. L'analyse permet de comparer le comportement d'une couche limite en présence de gradient de pression adverse avec le cas d’une plaque plane à grands nombres de Reynolds / The aim of the present work is to study a boundary layer subjected to a pressure gradient and to compare it with a zero pressure gradient (ZPG) boundary layer at high Reynolds numbers. Within this framework, focus is laid on the behaviour of large-scale coherent structures. Due to their large streamwise extent, these structures are not easy to extract and characterize using standard measurement techniques. For this reason, specific experimental set-ups using PIV in the streamwise/wall-normal planes was designed to capture the large-scale structures and to gain more insight into the mechanisms governing the dynamics of these flows. The achievements of the present investigation can be divided into two parts. The first part revisits the results obtained on a ZPG turbulent boundary layer by probing the origin of a spectral range in a turbulent boundary layer. To this end, a simple model which can in principle be applied to various wall-bounded turbulent flows is proposed from a new perspective based on the work of Townsend-Perry. The second part focuses on improving the understanding of turbulence under an adverse pressure gradient (APG) by performing a complete flow characterisation of an academic test case on a large scale geometry. Emphasis is laid on the characteristics of the structures (length, scaling, energetic contribution and their wall normal distribution) along with the influence of the APG on the large-scale structures. The analysis is also extended to compare the behaviour of APG with the ZPG case at high Reynolds numbers

Esta tese trata o escoamento ao redor de um cilindro logo após a sua primeira instabilidade, dentro do seu regime bidimensional periódico. A abordagem é principalmente teórica, passa por experimentos e culmina em uma importante parte numérica que complementa a teoria com evidências e ilustrações. As principais contribuições são a análise sobre a composição modal da solução dentro do regime periódico e o método desenvolvido para identificar autovetores de uma linearização da equação de Navier-Stokes presentes em uma dada solução. As bases compostas pelos autovetores identificados servem para a projeção da equação de Navier-Stokes e dão a essência dos modelos reduzidos deste estudo. A aplicação numérica apresentada para Re = 60 traz duas iterações do processo, com duas bases de autovetores de dimensões 12 e 24. Os modelos reduzidos são numericamente estáveis e a sua integração apresenta custo várias ordens mais baixo que o da simulação numérica completa. As séries temporais das coordenadas e as bases de autovetores possibilitam a recomposição do escoamento e a sua comparação com a simulação numérica de referência. A análise de aderência foi baseada nas médias temporais, nos valores de Strouhal e na estrutura dos harmônicos. Ambos modelos reduzidos têm correspondência próxima com o comportamento assintótico do escoamento e a tendência convergente das iterações é clara. As simetrias espaciais e temporais dos harmônicos são facilmente identificadas na estrutura dos modelos, de forma que as bases construídas podem ser entendidas como conjuntos de estruturas coerentes do fenômeno. / This thesis concerns the flow past a cylinder just after its first bifurcation, within its two-dimensional periodic regime. The approach is mainly theoretical, goes through experiments and is concluded by an important numerical part which complements the theory with evidences and illustrations. The main contributions are the analysis concerning the modal composition of the solution within the periodic regime and a method to identify eigenvectors of some linearizaton of the Navier-Stokes equation participating on a given solution. The bases spanned by the identified eigenvectors are employed in the projection of the Navier-Stokes equation and are central to the reduced models of this study. The numerical results for Re = 60 present two iterations of the process, with two bases of dimensions 12 e 24. The reduced models are numerically stable and their integration is many orders less costly than that of the full simulation. The time series of the modal coordinates and the eigenvectors bases allow the recomposition of the flow and its comparison with the full simulation results. The convergence analysis was based on the time averages, the Strouhal number values and the harmonic structure. Both reduced models have close correspondence with the asymptotic behavior of the flow and the convergent trend of the iterations is clear. The space and time symmetries of the harmonics have a simple representation within the structure of the models, therefore the identified bases can be understood as sets of coherent structures of the phenomenon.

L'écoulement autour des véhicules produit une couche limite turbulente très proche de la paroi. Le caractère turbulent induit des fluctuations de pression pariétale qui font vibrer les panneaux du véhicule. Ces vibrations sont alors transmises à travers la structure et rayonnent du bruit dans l’habitacle. Les niveaux sonores dus à l'écoulement augmentent avec la vitesse du véhicule. Pour cette raison, cette problématique connaît un intérêt croissant dans le secteur aéronautique.Le but de cette thèse est double : comprendre les mécanismes à l’origine de ces fluctuations de pression à la paroi et prédire l’excitation de la structure avion due à l’écoulement turbulent.Pour ce faire, des calculs Large Eddy Simulation (LES) sont disponibles. Il s’agit de bases de données numériques d’écoulements de couches limites turbulentes en présence de gradients de pression favorable, adverse et nul. Ceci permet de caractériser l’écoulement sur des géométries courbes telles que la pointe avant d’un avion. L’effet du gradient de pression sur des structures cohérentes de type « hairpins » et paquets de « hairpins » a pu être identifié et quantifié à travers des méthodes de visualisation et d’analyse statistique. Une méthode d’estimation stochastique du champ de vitesse a révélé la présence de tourbillons contra-rotatifs au-dessus des paquets de hairpins. Ces tourbillons ont une vorticité opposée à celle des hairpins et un modèle de « hairpins inversé » a été proposé.En s’inspirant du travail de Ahn et al. (2010), un modèle stochastique de spectre de la pression pariétale a été développé. Il s’agit de reconstruire un champ stochastique de vitesse instantanée et d’en déduire le champ de pression à la paroi en résolvant une équation de Poisson sur la pression. Le champ de vitesse est obtenu en soumettant des structures de type paquets de hairpins à un écoulement moyen. Les caractéristiques des paquets de hairpins en fonction du gradient de pression sont basées sur l’analyse des bases LES. Les résultats de ce modèle sont comparés à ceux issues de bases de données numériques et expérimentales. Enfin, ce modèle est utilisé pour caractériser l’écoulement de couche limite turbulente dans une simulation de Statistical Energy Analysis (SEA) afin de prédire les niveaux de vibration des panneaux d’une portion de fuselage d’avion. / The flow around vehicles creates a turbulent boundary layer in the vicinity of the wall. The turbulent behavior induces pressure fluctuations that make the panels vibrate. These vibrations are then transmitted though the structure of the vehicle and radiate noise inside the cabin. The flow-induced noise levels increase with the speed of the vehicle. For this reason, aircraft manufacturers show a great interest in this topic.There are two objectives for this thesis: understand the mechanisms responsible for the wall-pressure fluctuations and predict this source of aircraft panel excitation.A study of available Large Eddy Simulation (LES) computations was performed. The database consists in simulations of turbulent boundary layer flows submitted to favorable, adverse and zero pressure gradients. This is necessary to understand the nature of the flow over curved geometries such as the aircraft flight deck. The effect of pressure gradients on coherent hairpin structures and hairpin packets could be identified and quantified based on visualization and statistical analysis methods. Linear stochastic estimation of the velocity fields revealed a pair of counter-rotating streamwise vortices above hairpin packets. These vortices have a vorticity opposite to that of the hairpins and an “inverse hairpin” model was proposed.Following the work of Ahn et al. (2010), a stochastic model for wall-pressure spectrum was developed. The idea is to build a stochastic turbulent velocity field using hairpin packets which are subjected to a mean flow. The characteristics of the packets depending on the pressure gradient are based on the analyses of the LES database. The pressure field at the wall is obtained by solving a Poisson equation. The results of the hairpin packet model are compared to numerical and experimental data. Finally, the model is used as input for a Statistical Energy Analysis (SEA) simulation in order to predict the levels of vibrations of panels submitted to a turbulent boundary layer flow over a portion of an aircraft cabin.

Ce travail porte sur l’étude des structures cohérentes dans une couche limite de plaque plane à hauts nombres de Reynolds (Reθ=9830 et Reθ=19660). L'estimation Stochastique Linéaire est utilisée pour reconstruire un champ de vitesses résolu en espace et en temps à partir des mesures aux fils chauds à 30 KHz et des mesures PIV à 4 Hz. Une base de données DNS d’un écoulement de canal turbulent a été utilisée pour valider la procédure de reconstruction. Le champ de vitesse reconstruit est comparé à un champ de référence extrait de la DNS avec la visualisation des contours du champ de vitesse, l’analyse spectrale, les corrélations, etc. Par ailleurs, une analyse statistique est réalisée sur le champ de PIV originale, les données issues des fils chauds et celles reconstruites. Les résultats obtenus sont comparés à ceux de Carlier et Stanislas (2005) qui servent de référence. Des algorithmes ont été développés pour extraire les structures cohérentes du champ reconstruit. Les régions de vitesse uniforme sont caractérisées avec leur diamètre hydraulique, leur durée de vie et leur contribution aux tenseurs de Reynolds. Pour les tourbillons, on leur associe un rayon, une circulation et une vorticité en plus de leur durée de vie et leur nombre calculé à une distance fixe de la paroi. L’organisation spatiale des structures est étudiée avec la corrélation de leurs fonctions indicatives. Le modèle simplifié à grandes échelles qui en découle est comparé à ceux qui existent dans la littérature / This study lies in the context of large scale coherent structures (uniform momentum regions and vortical motions) investigation in a near wall turbulent boundary layer at high Reynolds numbers (Reθ=9830 and Reθ=19660).   With a Linear Stochastic Estimation procedure based on correlations computation, a full time-resolved 3 component field is reconstructed at high frequency from stereo-PIV at 4 Hz and hot wire data at 30 kHz . A DNS database of turbulent channel flow was used to validate the reconstruction method. To evaluate the quality of the method, reconstructed velocity fluctuations are compared to refence ones extracted directly from the DNS database.  Furthermore,  a statistical analysis is performed on HWA, PIV and LSE velocity fields. The results obtained are compared to those from Carlier and Stanislas (2005). Algorithms were developed to extract coherent structures from the reconstructed field. Uniform momentum regions are characterized with their mean hydraulic diameter in the YZ plane, their life time and their contribution to Reynolds stresses. The vortical motions are characterized by their position, radius, circulation and vorticity in addition to their life time and their number computed at a fixed position from the wall. The spatial organization of the structures was investigated through a correlation of their respective indicative functions in the spanwise direction . The simplified large scale model that arised is compared to the ones in the literature

This thesis focused on investigating flow, coherent structures, and mechanisms leading to scour around a series of three submerged groynes in a high curvature (radius of curvature (R)/channel width (B)=1.5) channel bend using a Large Eddy Simulation Numerical (LES) model. Flow was investigated during both an initial and a later stage of scour. The results showed that the groynes appeared effective in keeping the main core of high streamwise velocity away from the outer bank wall in the region where they were installed, although high potential still existed for local scour around the groynes. During the initial stage of scour, horseshoe vortices (HVs) showed the greatest propensity to induce scour immediately upstream of the groyne tips. During the later stage of scour, the HV in front of the first upstream groyne (G1) induced very high mean pressure fluctuations on the outer bank wall. Scour was also of very great concern around the tip of G1 due to severe mean bed pressure fluctuations. Downstream of the groyne field, the presence of a counter-rotating outer bank cell was capable of endangering the stability of the outer bank. The second focus of this thesis was to investigate flow in a 135° channel bend using both Reynolds Averaged Navier Stokes (RANS) and LES numerical models. The RANS study examined the effects of curvature ratio (R/B), and aspect ratio (B/H, where H is the inlet flow depth), on secondary circulation strength, and bed shear stresses. The study revealed that a decrease in R/B was associated with an increase in secondary circulation strength and peak bed shear stress. A change in B/H also substantially affected cross stream circulation strength. The LES study was conducted in a 135° (R/B = 1.5) bend flume with a fixed bed corresponding to near equilibrium scour conditions, and the results were compared to a similar high curvature 193° bend numerical study. Inner bank vortices and shear layers were present in both cases although their characteristics were substantially different. Distributions of boundary friction velocities, and turbulence were also quite different for each case.

No description available.

Cette thèse présente une étude numérique des régimes turbulents au sein d'un écoulement de Poiseuille plan forcé par un gradient de pression constant. L'effort numérique a porté principalement sur le concept d'Unité Minimale. Dans la première partie, des simulations en régime turbulent ont été conduites en géométrie périodique. Les DNS en Unité Minimale montrent que, l'activité turbulente se trouve localisée à proximité d'une des parois, et que la dynamique aux temps longs s'organise autour de renversements abrupts. Dans la seconde partie, on recherche par le calcul les états cohérents exactes en particulier les états dits frontière. Ces états frontière, obtenus par dichotomie, sont caractérisés par tourbillons longitudinaux et une paire unique de stries toujours localisées à proximité d'une seule paroi. Des représentations de la dynamique dans l'espace des phases sont reconstruites à l'aide de divers observables. La dynamique d'un renversement s'articule autour de visites transitoires vers un espace de solutions quasi-symétriques. Une onde progressive exacte, instable et quasi-symetrique a ainsi été identifiée. L'analyse de stabilité révèle que ses vecteurs propres séparent l'espace des phases en deux basins distincts. La dernière partie remet en question l'auto-similarité des différents régimes d'équilibre d'écoulement. Contrairement aux études récentes qui se concentrent sur les solutions à structure symétrique imposée, nos résultats suggèrent que les unités de parois sont également pertinentes pour les états frontière lorsqu'ils sont localisés près d'une paroi, meme si l'auto-similarité n'est pas aussi flagrante que pour les régimes turbulents. / This thesis numerically investigates the dynamics of turbulence in plane Poiseuille flow driven by a fixed pressure gradient. The focus is especially on computations carried out within the minimal flow unit (M.F.U.). In the first part, turbulent simulations are carried out in spatially periodic channels. In the M.F.U. simulations, the turbulent activity appears to be localised near one wall and the long term dynamics features abrupt reversals. In the next part, we look numerically for exact coherent states in the M.F.U. system. Edge states, which are computed using bisection exhibit streamwise vortices and a single pair of streaks localised near only wall at all times. Different state space representations and phase portraits were constructed using appropriately chosen variables. The dynamics along a turbulent reversal is organised around transient visits to a subspace of (almost) symmetric flow fields. A nearly-symmetric exact travelling wave (TW) solution was found in this subspace. Stability analysis of the TW revealed that its unstable eigenvectors separate the state space into two symmetric basins. In the last part of this thesis, the self-similarity of the different non-trivial equilibrium flow regimes computed in this work, is addressed. Contrarily to most studies focusing on symmetric solutions, the present study suggests that inner scaling is relevant for the description of edge regimes as well although the self-similarity is not as satisfactory as for the turbulent regimes.