La thèse s'inscrit dans un axe de recherche visant à améliorer les connaissances et prédictions des chargements hydrodynamiques subis par les réservoirs de carburant lors d'impacts balistiques (coup de bélier hydrodynamique) pour améliorer la survivabilité de la structure. Les modèles numériques les plus avancés ne permettent toujours pas de simuler le phénomène complet. De plus les modèles développés sont trop coûteux pour être utilisés lors d'optimisations de réservoirs durant la phase de conception. L'étude proposée consiste à développer un modèle analytique capable de simuler la séquence d'expansion et d'effondrement de la bulle de cavitation et d'utiliser ce modèle pour déterminer les paramètres influant sur les chargements hydrodynamiques lors de coups de bélier hydrodynamiques. La problématique du phénomène de coup de bélier hydrodynamique et l'état de l'art sont présentés dans le premier chapitre du mémoire. Dans le deuxième chapitre, le modèle de Rayleigh-Plesset est modifié pour prendre en compte l'effet de confinement d'un réservoir sphérique sur la dynamique d'expansion de la bulle en négligeant la présence du gaz dans la cavité dans les prédictions des chargements hydrodynamiques. Ce modèle a été appliqué à des cas expérimentaux de coup de bélier, en calibrant un paramètre de confinement lié à la rigidité de la structure. L'étape suivante, présentée dans le troisième chapitre, consiste à évaluer la capacité d'un modèle analytique linéaire élastique (de type plaque) à estimer la valeur du paramètre de confinement. Un bon accord est trouvé entre cette valeur et celle calibrée précédemment, ce qui valide la méthode d'estimation de ce paramètre et supprime la nécessite de la calibration expérimentale. Dans le quatrième chapitre le modèle incompressible de Rayleigh-Plesset est comparé à des simulations explicites éléments finis pour déterminer les effets de l'inertie de la structure et de la compressibilité du liquide sur la dynamique des bulles confinées. Enfin dans le cinquième chapitre un modèle analytique de dynamique de bulles confinées, dans un liquide compressible, est développé et validé. Pour cela une formulation basée sur l'équation de Keller-Miksis est proposée. La pertinence de ce modèle est vérifiée puis il est validé par rapport à des simulations éléments finis, ce qui permet d'estimer l'amélioration apportée par le modèle de Keller-Miksis dans la prédiction des chargements hydrodynamiques par comparaison au modèle de Rayleigh-Plesset. Finalement une analyse critique du travail de thèse et des perspectives sont données. / The context of the thesis consists of improving the knowledge and the predictions of hydrodynamic loads applied on fuel tanks during ballistic impacts (Hydrodynamic Ram) to improve the survivability of aircraft structures. The most advanced numerical models still cannot simulate the entire phenomenon. Moreover, these models are too expensive to be used for optimisation or for design purposes during the tank development stage. The proposed study consists of developing an analytical model capable of simulating the expansion and collapse of the cavitation bubble and to use it to determine the parameters that influence the consecutive hydrodynamic loads. The state of the art hydrodynamic ram problem is first presented. In the second chapter, the Rayleigh-Plesset model is modified to include the stiffness effect of a spherical container on the described bubble dynamics, when neglecting the presence of the bubble gas on the hydrodynamic loads. This model is applied to hydrodynamic ram test cases, by calibrating a confinement parameter which is related to the structure rigidity. The next step, presented in the third chapter evaluates the capacity of an elastic linear analytical model (plate formulae) to provide the value of the confinement parameter. A good agreement is found with the calibrated value in the previous chapter, which validates the method to get this parameter and suppress the need for experimental calibration. In the fourth chapter, the incompressible Rayleigh-Plesset model is compared to explicit finite element simulations to determine the influence of the liquid compressibility and structural inertia on the dynamics of confined bubbles. Then, in the fifth chapter an analytical model for confined bubbles in a compressible liquid is developed. A formulation based on the Keller-Miksis model is proposed. The relevance of this model is verified and it is validated again with respect to finite element simulations. Then the improvement of hydrodynamic loads predictions using this model is estimated by comparison with the Rayleigh-Plesset model. Finality a critical analysis of the thesis work and some outlooks are given.

A Lagrangian formalism presented by Hay, Ilinskii, Zabolotskaya, and Hamilton [J. Acoust. Soc. Am. 132, 124--137 (2012)] to calculate the pulsation of a spherical bubble, immersed in liquid and near one or two viscoelastic layers, is extended here to include bubble translation. The method presented here is simplified from that given by Hay et al. in that only a single interface between a liquid and a viscoelastic half-space is considered. In the present approach the force on the bubble due to the presence of the liquid-solid interface is calculated using a Green's function that takes into account elastic waves and viscosity in the layer, and the viscous boundary layer within the liquid near the interface. Previous models and experiments have shown that the direction of bubble translation near a viscoelastic layer is correlated with the direction of a liquid jet often produced by the bubble during collapse. In this dissertation an attempt is made to model the pulsation and translation of a spherical bubble near a liquid-solid interface to infer the direction of bubble translation in reference to material parameters of the liquid and viscoelastic medium, and the standoff distance of the bubble from the interface. The analysis is simplified by demonstrating that the direction of bubble translation can be inferred from the phase of the component of the Green's function associated with the reverberant pressure gradient. For linear bubble pulsation it is shown that the domain of material properties of the viscoelastic medium which generally corresponds to bubble translation away from the interface occurs when the effective stiffness of the viscoelastic medium is greater than the effective damping for both itself and the liquid. The analysis is performed assuming the viscoelastic medium is similar to soft tissue, and its dynamics are described by a Voigt, Kelvin, or Maxwell model. The simulations are compared with existing experimental data. Effects of high-amplitude bubble pulsation are explored in terms of how the simulations differ as the pulsation amplitude increases. At higher pulsation amplitudes, it is shown that bubble translation is still described qualitatively by analyzing the phase of the reverberant pressure gradient. / text

Presented work consists of research and experimental part. Main focus of the research part is on mutual interaction between magnetic field and water and between magnetic field and cavitating flow. Prediction of the result is almost impossible, because the opinions on the effects differ, sometimes even go against each other. Experimental part consisted of three trials with permanent magnets. Electromagnetic probe was employed for measuring of the voltage. It appears, that magnetic field affects the cavitation, because in all of the trials the measured voltage changed after the switch from non-cavitating to cavitating flow. The change of the voltage was not great enough, thus application of this phenomenon in practice cannot be recommended.

This thesis deals with the numerical solution of cavitation bubble dynamics and with cyanobacteria gas vesicle behaviour. A program for the numerical calculation of bubble dynamics is created using the Rayleigh-Plesset equation and its modifications. Subsequently, bubbles of different sizes are investigated during acoustic cavitation with various driving frequencies. Furthermore, a model for hydrodynamic cavitation is created. The model combines CFD computation of flow in the Venturi nozzle with the cavitation bubble dynamics calculation. The last part of the work is dedicated to cyanobacteria gas vesicle behaviour in a variable pressure field and during passage through the Venturi nozzle.

In this thesis, the dynamics of an isolated cavitation bubble submerged in a steady flow is studied numerically. A Lagrangian-Eulerian approach is considered, in which properties of the fluid are computed first by means of Eulerian methods (in this study the commercial CFD software Ansys Fluent 19 was used) and the trajectory of the bubble is then computed in a Lagrangian fashion, i.e. the bubble is considered as a small particle moving relative to the fluid, due to the effect of several forces depending on fluid's pressure field, fluid's velocity field and bubble's radius. Bubble's radius dynamics, modeled by Rayleigh-Plesset equation, has a big influence on its kinetics, so a special attention is given to it. Two study cases are considered. The first one, motivated by acoustic cavitation is concerned with the response of the bubble's radius in a static flow under the influence of an oscillatory pressure field, the second one studies the trajectory of the bubble submerged in a fluid passing by a Venturi tube and a sharp-edged orifice plate.

This thesis deals with the issue of cavitation and its effects. In this context, it describes the mechanism of origin and implosion of cavities and cavitation regimes. It lists various types of hydrodynamic cavitation. It presents the Rayleight-Plesset equation and describes micro jet. It also highlights cavitation erosion and the effects of cavitation on some types of materials. It deals with three types of cavitation resistance testing, namely cavitation tunnels, a vibrating cavitation system, supported by the ASTM G32 standard, and last but not least, cavitation nozzles, which follow the ASTM G134-17 standard. In correlation with cavitation nozzles, it frames its four basic parameters, which are stand of distance, the cavitation number, the speed of sound and the geometry of the nozzle. At the end of the theoretical part it characterizes the construction of test bench. The practical part is focused on performing the experiment. It first presents the procedure for carrying out the experiment and then evaluates this experiment. Part of the evaluation is the visual observation of selected samples of AlCu4Mg1Mn1 material and the monitoring of cavitation erosion on specific samples. First, these data are processed in the form of graphs and tables. It uses a microscope as a tool for detailed observation of samples. The conclusion of the practical part is devoted to the evaluation of the experiment.

No description available.

Dans cette thèse nous avons proposé un modèle robuste permettant d’expliquer la propagation de bulles dans des systèmes composés de plusieurs cavités de taille micrométrique. Notre modèle s’applique d’abord à deux cavités placées côte à côte, où l’une d’entre elle contient une bulle vibrante provoquant le mouvement de la paroi séparant les deux cavités. Un système d’équations modélisant ce système a été établi. Nous montrons que le choix des conditions initiales contrôle le déclenchement de l’onde de cavitation. En effet, le déphasage entre la vibration de la bulle et de la membrane vibrante donne les conditions nécessaires et suffisantes pour expliquer les observations expérimentales. En s’appuyant sur les résultats théoriques, nous avons développé un modèle macroscopique numérique, très phénoménologique, permettant la propagation de bulles de cavitation et la formation de clusters de bulle. / In this thesis, a robust model is proposed towards explaining bubbles propagation within systems composed by several micrometric cavities. Our first model is based on two cavities placed side by side, one of which contains a vibrating bubble which sets into motion the wall separating the two cavities. A system of equations modeling this system is proposed. It has been proved that the choice of the initial conditions controls the cavitation wave initiation. Indeed, the phase shifts between the bubble vibration and the vibrating membrane gives the necessary and sufficient conditions to explain the experimental observations. In this framework, we have developed a phenomenological and numerical macroscopic model, allowing the propagation of cavitation bubbles and the formation of bubble clusters.