Pour améliorer les performances aérodynamiques notamment dans le domaine des transports (aéronautique, automobile, ..), le contrôle des écoulements constitue une solution de rupture prometteuse. Il présente à la fois un enjeu majeur pour l'industrie et un défi pour les scientifiques. Les retombées visées par le contrôle (notamment actif) sont d'ordre à la fois économique (réduction de la consommation) et environnemental (diminution des gaz à effet de serre). L'efficacité du contrôle est intimement liée à l'actionneur utilisé. Les actionneurs fluidiques et tout particulièrement le jet synthétique, semblent être une technique prometteuse de progrès. L'objectif de ce travail est le développement, la validation et la mise en œuvre de ce type d'actionneur. Nous avons ainsi conçu un actionneur de type jet synthétique basé sur un haut parleur. Les performances de cet actionneur ont été évaluées dans un milieu au repos en utilisant à la fois l'anémométrie à fil chaud et la PIV. Les effets des différents paramètres ont été analysés (fréquence, amplitude de forçage, …). Une attention particulière a été consacrée à l'étude de l'inclinaison de la fente d'éjection. Trois inclinaisons ont été utilisées (30°, 45° et 90°). Dans chaque cas l'évolution du jet synthétique ainsi que celle des tourbillons ont été caractérisées. Par les mêmes techniques de mesure dans une soufflerie, nous avons étudiés l'interaction du jet synthétique avec un écoulement transverse. Des visualisations par fumée du contrôle du décollement par cet actionneur montrent ses capacités à pouvoir recoller l'écoulement. / To improve the aerodynamic performance particularly in the field of transport (aerospace, automotive, ..), the flow control is a promising solution for rupture. It presents at the same time a major stake for industry and a challenge for the scientists. The fallout aimed by control (particularly active control) are of an at the same time economic (reduction of consumption) and environmental (decrease of greenhouse gases). The effectiveness of control is closely related to the actuator used. The fluidic actuators and particularly the synthetic jet appear to be a promising technique for progress. The objective of this work is the development, validation and implementation of this type of actuator. We thus designed an actuator of the synthetic jet type based on a loudspeaker. The performances of this actuator were estimated in a quiescent environment. Both hot wire anemometer and the Particle Image Velocimetry (PIV) are used. The effects of different parameters were analyzed (frequency, amplitude forcing, ...). Particular attention has been devoted to the study of the inclination of the ejection slot. Three angles were used (30 °, 45 ° and 90 °). In each case the development of synthetic jet and the vortices have been characterized. By the same techniques of measurement in a wind tunnel, we have studied the interaction of the synthetic jet with a transverse flow. Visualizations by smoke control detachment by the actuator to show his ability to pick up the flow.

The treatment of single ventricle anomalies is a formidable challenge for clinical teams caring for patients with congenital heart disease. Those patients fortunate to survive surgical palliation contend with lifelong physical limitations and late stage pathophysiology. A mechanical blood pump specifically designed to increase pressure in the great veins would augment flow through the lungs and provide hemodynamic stability until a donor heart is located. To support the development of such medical devices, this research characterized the fluid dynamics of mechanical assistance in the Fontan circulation by performing numerical analyses and particle image velocimetry (PIV) studies in a patient-specific in vitro model. This project investigated the performance of three pump prototype configurations. ANSYS-CFX was used to conduct the computational studies for a range of operating conditions and degrees of Fontan dysfunction. Pressure generation, blood trauma predictions, shear stresses, fluid streamlines, and velocity profiles were examined. Three-dimensional PIV studies were completed and compared to the numerical estimations. Computational findings and experimental data correlated to within literature expectations. Blood damage levels, shear stresses, and fluid residence times remained reasonable or below threshold limits. The blood pump configurations met expectations by achieving target design specifications for clinical application. The pumps enhanced the rate of hydraulic power gain in the cavopulmonary circuit, reduced inferior vena cava pressure, and minimally increased pulmonary arterial pressure. The blood pump with the twisted protective stent produced the most rapid increase in the rate of power gain and the highest pressure generation. The PIV measurements illustrated a strong dependency of the fluid dynamics on the patient-specific vessel geometry and the particular pump design. The pump having the twisted cage outperformed the other designs and had a dominating impact on the blood flow distribution in the cavopulmonary circuit. A strong rotational component in the flow was observed leaving the pumps. These results confirm that mechanical cavopulmonary assistance is a viable therapeutic option. Significant knowledge into a new class of blood pumps and how these pumps interact with a single ventricle physiology was gained, thus advancing the state-of-the-art in mechanical circulatory support and addressing a significant human health problem.

Ce travail s’inscrit dans le cadre de nos recherches sur les écoulements à bulles. C’est l’étude expérimentale du mélange d’un traceur très peu diffusif (fluoresceine dans l’eau) dans l’écoulement instationnaire engendré par le passage d’un écoulement uniforme au travers d’un réseau d’obstacles sphériques (2% de fraction volumique) répartis aléatoirement dans l’espace. Cet écoulement reproduit correctement les caractéristiques de l’agitation dans un essaim de bulles en ascension. La vitesse du fluide est mesurée par PIV de manière assez classique. Le traceur est injecté en amont du réseau et l’´evolution de sa concentration est mesurée par PLIF. L’utilisation de la PLIF pour mesurer des champs de grande extension (15 cm) et avec une grande dynamique d’intensité lumineuse fluorescée constitue une contribution importante de ce travail. Les résultats ainsi obtenus montrent que, à petit nombre de Reynolds, le mélange est régi par les forts gradients de vitesse qui existent dans le voisinage des sphères. A grand nombre de Reynolds, il est maîtrisé par la turbulence qui se développe alors. L’analyse des résultats comporte deux parties principales : (1) une analyse statistique des profils de concentration aboutissant à la détermination d’un coefficient de diffusion effectif et (2) une description de la distribution spectrale des fluctuations de vitesse et de concentration. / This research is a part of our research about bubbly flows. Experiments are performed about mixing of a high Schmidt scalar component (fluorescein in water) by the agitation generated by the flow through a random array of fixed spheres (at high Re and with a volume fraction of solid equal to 2%). This flow mimics for a great part the agitation in the liquid phase of a bubble swarm rising in a liquid otherwise at rest. The velocity of the liquid is estimated from PIV measurements. The scalar is injected through a point source in the array and the evolution of its concentration is estimated by PLIF method. An important part of this research is the measurement of large fields of concentration (15 cm) with a good precision by PLIF. The results show that, at moderate Reynolds number (100), mixing is controled by the steep velocity gradients located near the spheres, while, at large Reynolds number, it is controled by the turbulence that develops. The analysis of the results is composed of two parts : (1) the statistical analysis of the spatial distribution of scalar concentration, and the determination of an effective diffusion coefficient, (2) a spectral analysis of the velocity and concentration fields.

[pt] Estudos recentes mostram que a injeção de emulsões óleo-água pode levar a uma melhor varredura do reservatório e reduzir consideravelmente a saturação residual de óleo em processos de recuperação avançada. Estes efeitos estão diretamente ligados ao comportamento do escoamento de gotas de óleos suspensas em água através das gargantas de poros. Desta forma, a otimização do processo de injeção e da formulação da emulsão com o objetivo de aumentar o volume de óleo recuperado requer por um melhor entendimento do escoamento na escala de poros. Este trabalho apresenta um estudo do escoamento de gotas de óleos suspensa em água através de micro canais de seção reta constante e através de uma garganta, que são usados como modelos do espaço poroso. O campo de velocidade da fase contínua e a velocidade da gota de óleo foi determinado através da técnica de velocimetria por imagem de partículas em escala micrométrica (Micro-PIV) para diferentes tamanhos de gotas e geometria do capilar e garganta. Os resultados obtidos mostram a variação do padrão do escoamento devido a presença da gotas de óleo e fornecem importantes informações de como gotas de óleo mudam a mobilidade do fluido injetado quando o mesmo escoa através de poros com gargantas menores do que tamanho das gotas. / [en] Recent studies show that oil-water emulsion injection may lead to a better reservoir sweep and reduce residual oil saturation in enhanced oil recovery processes. These effects are directly linked to the flow behavior of oil drops suspended in water through the pore throats of a porous material. Therefore, the optimization of the injection process and of the emulsion properties with the goal to increase the volume of oil displaced requires a better understanding of the emulsion flow in the pore scale. This work presents an analysis of the flow of oil drops suspended in water through micro channels with constant cross-section area and with a throat, that are used as a model for the pore scale. The velocity field of the continuous phase and the drop velocity are obtained using the micro particle image velocimetry techniques (Micro-PIV) for different drop sizes and micro channel geometries. The results show the changes in the flow pattern due to the presence of oil drops and yield important information on how oil drops reduce the mobility of the injected liquid when it flows through pore throats smaller than the drop size.

<p>Microscale environments provide significant resolution and distortion challenges with respect to measurement techniques; however, with improvements to existing techniques, it is possible to gather relevant data to better understand the thermal and fluidic mechanisms at such small scales in evaporating droplets.</p> <p> </p> <p>Infrared thermography provides several unique challenges at small scales. A primary issue is that the low native resolution of traditional infrared cameras significantly hamper the collection of details of microscale features. Furthermore, surfaces exhibiting vastly different emissivities, results in inaccurate temperature measurements that can only be corrected with irradiance-based emissivity maps of the surface; however, due to the resolution limitations of infrared thermography, these emissivity maps can also display significant errors. These issues are overcome through the use of multi-frame super-resolution. The enhanced resolution allows for better capture of microscale features, therefore, enhancing the emissivity map. A quantitative error analysis of the system is conducted to quantify the feature size resolution improvement as well as the smoothing effect of super-resolution reconstruction. Furthermore, a sensitivity analysis is conducted to quantify the impact of registration uncertainty on the accuracy of the reconstruction. Finally, the improved emissivity map from super-resolution is demonstrated to show the increased accuracy over low-resolution mapping.</p> <p> </p> <p>When applied to water droplets, particularly on nonwetting surfaces, infrared thermography is confounded by the presence of nonuniform reflectivities due to the spherical curvature of the liquid-air interface. Thus, when measuring the temperature along the vertical axis of a water droplet, it is necessary to correct the reflection. Using a controlled background environment, in conjunction with the Fresnel equations, it is possible to correct the reflective effects on the interface and calculate the actual temperature profile. This allows for a better understanding of the governing mechanisms that determine the thermal transport within the droplet. While thermal conduction is the primary transport mechanism along the vertical axis of the droplet, it is determined that the temperature drop is partially dampened by the convective transport from the ambient air to the liquid interface. From this understanding revealed by the measurements, the vapor-diffusion-based model for evaporation was enhanced to better predict evaporation rates.</p> <p> </p> <p>Further exploration into the mechanisms behind droplet evaporation on nonwetting surfaces requires accurate knowledge of the internal flow behavior. In addition, the influence of the working fluid can have a significant impact on the governing mechanisms driving the flow and the magnitude of the flowrate. While water droplet evaporation has been shown to be governed by buoyancy-driven convection on nonwetting substrates, similar studies on organic liquid droplets are lacking. Particle image velocimetry is effective at generating a velocity flow field, but droplets introduce distortion due to the refraction from the spherical interface of the droplet. As such, velocity correction using a ray-tracing approach was conducted to correct the velocity magnitudes and direction. With the velocity measurements, the flow was determined to be surface-tension-driven and showed speeds that are an order of magnitude higher than those seen in buoyancy-driven flow in water droplets. This resulted in the discovery that advection plays a significant role in the transport within the droplet. As such, the vapor-diffusion-governed evaporation model was adjusted to show a dramatic improvement at predicting the temperature gradient along the vertical axis of the droplet.</p> <p> </p> <p>Armed with the knowledge of flow behavior inside droplets, it is expected that droplets with aqueous solutions should exhibit buoyancy-driven convection. The final part of this work, therefore, leverages this phenomenon to enhance mixing during reactions. Colorimetry is a technique that is widely utilized to measure the concentration of a desired sample within some liquid; the sample reacts with a reagent dye the color change is measured, usually through absorbance measurements. In particular, the Bradford assay is used to measure protein concentration by reacting the protein to a Coomassie^TM Brilliant Blue G-250. The absorbance of the dye increases, most significantly at the 590 nm wavelength, allowing for precise quantitation of the amount of protein in the solution. A droplet-based reaction chamber with buoyancy-enhanced mixing has the potential to speed up the measurement process by removing the need for a separate pre-mixing step. Furthermore, the reduced volume makes the process more efficient in terms of reactant usage. Experimental results of premixed solutions of protein sample and reagent dye show that the absorbance measurement through a droplet tracks strongly with the protein concentration. When the protein sample and dye reagent are mixed <i>in situ</i>, the complex interaction between the reactants, the mixing, and the adsorption of protein onto the substrate creates a unique temporal evolution in the measured absorbance of the droplet. The characteristic peaks and valleys of this evolution track strongly with concentration and provide the framework for measurement of concentration in a droplet-based system.</p> <p> </p> <p>This thesis extends knowledge about droplet thermal and fluidic behavior through enhanced measurement techniques. This knowledge is then leveraged in a novel application to create a simple, buoyancy-driven colorimetric reaction setup. Overall, this study contributes to the field of miniaturized, efficient reaction and measurement devices.</p>

Le présent travail de thèse s’inscrit dans le contexte fondamental de la propulsion marine instationnaire. Il analyse les performances propulsives des profils de différents rapports d’aspects (envergure par rapport à la corde), en oscillation harmonique de tangage. On modélise la principale partie active d’un système propulsif naturel, tel une nageoire caudale de poisson, par des profils simples, minces et rigides quasiment identiques, à l’exception du rapport d’aspect très important du premier profil (4.8), dit bidimensionnel, et très faible du second (1/6), dit tridimensionnel. Le mouvement de tangage est imposé au tiers de corde, en partant du bord d’attaque des profils immergés dans un écoulement. L’objectif étant de faire varier le nombre de Strouhal, basé sur l’amplitude d’excursion totale du bord de fuite du profil, dans une gamme relativement faible et étroite, où évoluent plusieurs espèces aquatiques. Cette gamme est donc intéressante pour l’étude des systèmes propulsifs artificiels. L’aptitude à la propulsion a été déduite en analysant la structure de l’écoulement induit dans le sillage, et en déterminant l’effort réel exercé par le profil sur l’écoulement. En raison des effets importants de la traînée induite associée aux tourbillons marginaux et de ceux de la traînée visqueuse, inversement proportionnelle au rapport d’aspect, il apparait que les performances propulsives du profil tridimensionnel sont bien moindres que celles du profil bidimensionnel. L’influence des configurations du battement sur les performances propulsives a également été étudiée. / This thesis work is part of the fundamental context of unsteady marine propulsion. The propulsive performances of different aspect ratios (span to corde) profiles, in harmonic pitching motion are analyzed. The main active part of a natural propulsion system, such as a fish caudal fin, is modelled by simple, thin and rigid profiles that are almost identical, with the exception of the very high aspect ratio of the first profile (4.8), called two-dimensional, and very low aspect ratio of the second (1/6), called three-dimensional. The pitching motion is imposed on the third corde, starting from the leading edge of the profiles immersed in a flow. The objective is to vary the Strouhal number, based on the total excursion amplitude of the trailing edge of the profile, over a relatively small and narrow range, where several aquatic species evolve. This range is therefore interesting for the study of artificial propulsion systems. The propulsion capability was inferred by analyzing the structure of the induced flow in the wake, and by determining the actual force exerted by the profile on the flow. Due to the significant effects of the induced drag associated with marginal vortices and those of the viscous drag, inversely proportional to the aspect ratio, it appears that the propulsive performance of the three-dimensional profile is much lower than that of the two-dimensional profile. The influence of flapping parameters on propulsive performance has also been studied.

Em virtude da sua facilidade de operação e elevada eficiência, sedimentadores são bastante utilizados pelas indústrias na separação de componentes com diferentes densidades. Porém o desempenho do tanque sedimentador depende do campo de fluxo dentro do equipamento que, por sua vez, é influenciado pelas características dos sólidos em suspensão, da geometria e dimensões do tanque separador. As características hidrodinâmicas em um sedimentador circular vertical foram investigadas no presente trabalho, tanto por abordagens experimentais como numéricas. As experiências foram realizadas por velocimetria de imagem de partículas em um protótipo de sedimentador, construído em acrílico cristal transparente. As simulações foram realizadas, utilizando o pacote ANSYS CFX na versão 16.0, considerando escoamento 3D, turbulento, isotérmico e em regime permanente. O modelo em fluidodinâmica computacional foi criado em forma progressiva. Inicialmente, foram testados modelos para o sistema monofásico (água) e verificados através da comparação aos dados experimentais. Seis modelos de turbulência do tipo RANS foram analisados: epsilon, Renormalization Group (RNG) k-epsilon, k-omega, Shear Stress Transport (SST) e os de tensão de Reynolds (BSL RSM e SSG RSM). Para o estudo bifásico, empregou-se o modelo multifásico Euler-Euler, juntamente com o modelo de partícula e o de Gidaspow. Nesta situação, diversas configurações de condições de contorno e modificações no projeto do tanque original foram analisadas por meio de campo vetorial, linhas de fluxo, planos de concentração volumétrica e de energia cinética turbulenta. Os resultados mostraram que o modelo de turbulência BSL RSM foi o mais adequado para descrever o escoamento no sedimentador contínuo, sendo capaz de prever as recirculações e as variações de velocidade ao longo do equipamento. As zonas de recirculações (posição e intensidade) mostraram-se influenciadas pelas condições de contorno e forma geométrica do tanque de separação. Foi verificado que o aumento do diâmetro do tanque sedimentador (em 40,0%) provocou um favorecimento na separação água e sólido, elevando a eficiência em 17,3%, enquanto o aumento da altura (em 40,0%) do tanque reduziu a eficiência de separação em 27,5% em comparação ao projeto original. / Due to their operation facility and high efficiency, settlers are widely used by industries for separating components of different densities. The performance of sedimentation, however, depends on the flow field is inside the equipment, which in turn is influenced by the characteristics of the solids in suspension and by the geometry and size of the separation tank. The hydrodynamic characteristics of a vertical circular settler were investigated in the present work by means of both experimental and numeric approaches. The experimental analyses were carried out by particle image velocimetry in a settler prototype built in transparent crystal acrylic. The simulations were carried out using the software ANSYS CFX version 16.0, considering a 3-D, turbulent, isothermal and steady-state flow. The computational fluid dynamics model was used in the progressive approach. The models were initially tested for the one-phase process (water) and checked against the experimental data. Six RANS turbulence models were analyzed: k-epsilon, Renormalization Group (RNG) k- epsilon, k-omega, Shear Stress Transport (SST) and Reynolds Stress models (BSL RSM and SSG RSM). For the two-phase flow, the Euler-Euler multiphase model was used together with the particle model and the Gidaspow model. In this context, several configurations of boundary conditions and modifications in the original tank design were analyzed using vector field, flow lines, volumetric concentration and turbulent kinetic energy. The results showed that the BSL RSM turbulence model was more appropriate to explain the flow in the continuous settler, being able to predict the recirculations and velocity variations along the equipment. The recirculation zones (position and magnitude) were influenced by the boundary conditions and by the geometric shape of the separation tank. It was found that an increase in the diameter of the settler tank (in 40.0%) favored the water and solid separation, increasing its efficiency by 17.3%, while an increase in the height of the tank (in 40.0%) reduced the efficiency of separation by 27.5% when compared to the original design.

Um regime ou fluxo turbulento pode ser definido simplificadamente como uma forma de escoamento no qual o movimento de um fluido ocorre de forma aparentemente aleatória. A natureza caótica desse tipo de regime não possibilita que o problema seja abordado de modo determinístico, consequentemente, não pode ser descrito através de expressões analíticas para descrever seu campo de velocidades. Apesar de ser impossível estudar o movimento de cada partícula, pode-se visualizar o problema de forma macroscópica. Utilizando-se de simplificações e generalizações pode-se fazer uso de estudos numéricos em uma tentativa de encontrar uma face determinística do fenômeno, como por exemplo, verificação da isotropia e homogeneidade. A turbulência com estas características é um regime em que as componentes da velocidade no espaço e as velocidades em diferentes pontos do escoamento apresentam parâmetros estatísticos semelhantes. Tais condições não existem na prática, mas pode-se construir equipamentos que produzem escoamentos que se aproximam desta condição. Este trabalho consistiu na verificação das condições de turbulência isotrópica e homogênea em um tanque de grade oscilante. Foi utilizado o método de trilhagem de partículas por fotografia estereoscópica, PIV (particle imaging velocimeter), para observar e quantificar a turbulência gerada. Observou-se a turbulência gerada por 8 velocidades diferentes de oscilação da grade. Os resultados mostraram que o tanque em estudo aproximou-se de condições homogêneas, porém sua isotropia não pôde ser afirmada. O tanque com grade oscilante foi eficiente na geração de turbulência, reproduzindo o fenômeno real e colaborando para o seu estudo. Também foi realizado uma análise sobre a eficácia do método de velocimetria com o uso de PIV, que mostrou ser adequado para a observação de campos de velocidades de um escoamento, com limitações quanto à região periférica da área alvo. / Turbulence or turbulent flow can be defined in a simplified way as a flow with random moviment. The chaotic behavior of this regime does not allow a deterministic approach, therefore, it can not be described by analytical formulas for the velocity field. Although it is impossible to study the movement of each particle, the problem can be studied in a macroscopic way. Simplifications and generalizations can be used in numerical studies in an attempt to find a deterministic view of the phenomenon, for example, checking the isotropy and homogeneity. The turbulence with these characteristics is a regime in which the components of velocity in space and velocity at different points of the flow have similar statistical parameters. Such conditions do not exist in practice, but equipment can be built in order to produces flow approaching the condition. This work consisted of verify the conditions of homogeneous isotropic turbulence in a tank with oscillating grid. It was used the method of particule tracking by stereoscopic photography, PIV (particle imaging velocimetry), to observe and quantifying the generated turbulence. It was observed turbulence generated by 8 different speeds of oscillation of the grid. The results showed that the tank approached homogeneous conditions, but its isotropy can not be affirmed. The tank with an oscillating grid was efficient in the generation of turbulence, reproducing a real phenomenon and contribute to its study. Analysis was also conducted on the effectiveness of velocimetry method using PIV, shown to be suitable for observation of a flow velocity field with limitations on the peripheral region of the target area.

Une des pistes majeures pour obtenir des turbines à gaz moins polluantes est l'utilisation de chambres de combustion fonctionnant en régime prémélangé pauvre. Cependant, dans de tels régimes de fonctionnement, la flamme obtenue est cependant plus facilement sujette aux instabilités thermo-acoustiques.Dans cette thèse, un brûleur étagé swirlé fonctionnant au propane a été employé pour illustrer une méthode d'analyse de la combustion prémélangée. Cette installation a été étudiée à l'aide de diagnostics classiques (micros, PLIF, film à haute cadence de la flamme).La stratégie de post-traitement développée dans cette thèse utilise par contre des outils eux aussi classiques (PSD, moyenne de phase) mais aussi une méthode d'analyse avancée, la Décomposition en Modes Dynamiques. Celle-ci est notamment étudiée em détail afin d'en cerner les avantages et les limitations dans le cadre de l'analye des instabilités de combustion. / One of the major ways of reducing pollutant emissions in gas turbines is using combustion chambers with lean premixed mixtures. However, in such operating conditions the flame is more susceptible to thermo-acoustic instabilities. In the present work, a propane-fed swirled burner using two injection stages is used to illustrate a post-processing strategy for premixed combustion analysis. This model burner has been studied using classic diagnostics (microphones, PLIF, highspeed imaging).The strategy developed in the present work uses both classic postprocessing tools (PSD, phase averaging) and an advanced method, the Dynamic Mode Decomposition. In particular, this method is studied in detail and compared to more lassic ones in order to clearly point its advantages as well as its shortcomings when used to study combustion instabilities.

Ce travail est consacré à l'étude expérimentale des écoulements diphasiques et triphasiques d'inclusions (bulles, gouttes, sphères) en milieux tant newtoniens que non newtoniens à l'échelle microscopique et mésoscopique, en utilisant la visualisation par une caméra rapide, la vélocimétrie par images des particules (PIV) ainsi que la micro-vélocimétrie par images des particules.<br /><br />Des bulles et des gouttes ont été étudiées expérimentalement depuis leur formation, en passant par leur déformation jusqu'à leur coalescence. La formation de bulles dans des micro-mélangeurs a été étudiée et caractérisée par l'obtention de champs de vitesses. Différents paramètres, tels que le cisaillement, la géométrie de la zone de formation, les débits ou encore les propriétés physiques ont été testés afin de développer des lois d'échelles. <br /><br />La traversée d'une interface liquide-liquide par une inclusion a été abordée par des expériences originales, permettant de décrire la dynamique du phénomène, de définir des nombres adimensionnels et de mettre en évidence des instabilités interfaciales. L'effet Weissenberg a également été étudié aux différentes échelles afin de comprendre les phénomènes conduisant à son amplification.<br /> <br />Enfin, en milieu viscoélastique et rhéofluidifiant, nous avons caractérisé l'écoulement autour d'une inclusion isolée solide par l'obtention de champs de vitesses. Ces résultats ont permis de confirmer l'origine viscoélastique du sillage négatif et de prédire ses caractéristiques.