No description available.

The work presented in this thesis studies acoustic perturbations in slowly varying pipes. The slow variation is introduced in the form of a small parameter ${\epsilon}$ and through this in turn gives rise to a slow axial scale $X$ such that $X = {\epsilon}x$ where $x$ is the normal axial coordinate. This allows an asymptotic approach and the WKB method is used to solve the subsequent mathematical problems. The first deals with the existence of a trapped mode in a hard-walled pipe of varying radius conveying fluid. For the derived leading order propagating mode solution, its amplitude becomes singular at transition points $X_{t}$ and $X_{t'}$ where $X_{t} > 0$ and $X_{t'} < 0$ and thus is unable to propagate past these points. Because of the break down in the solution, this leads to the theory that in the neighbourhood of these points there exists a boundary layer in which the original assumption about having slow variation does not hold. By first seeking the thickness of the layer, valid solutions can then be derived and then matched to the outer solutions in order to produce a uniform solution which holds for the entire axial domain. Once this is achieved, it is then used to derive trapped mode solutions. In this case, the theory used is that of two single turning points which are then combined to obtain the full solution. It is illustrated through consideration of examples and the dependence on ${\epsilon}$ is also shown through various plots. This problem will be considered for a symmetric and asymmetric duct and for differing duct parameters. Problems may arise when the two turning points lie close together and so we seek to improve on the method used by deriving a solution to trapped modes encompassing both turning points, which will be proposed together with some illustrations in order to justify its use and reliability. Next, the case of mode propagations on a thin elastic shell of varying radius conveying fluid is studied. The acoustic solutions of a straight shell in vacuo are first briefly reviewed and then built up by the addition of radius variation and the presence of a stationary fluid. The work presented first outlines the analysis for wave propagation in a slowly-varying thin elastic shell in vacuo. It is found that the shell and the fluid terms are coupled through the fluid pressure term, which is added to the equation governing the radial shell displacements since the pressure is assumed to affect radial motion only. Once the newly corrected equation for the radial shell displacements has been obtained, together with the axial and azimuthal displacements equations, this new system of governing equations is then separated into leading order ${\epsilon}^{0}$ and first order ${\epsilon}^{1}$ systems. In order to simplify the calculations, only the zeroth azimuthal order $m = 0$ will be studied here. With this simplification, a notable result is that the solutions of the torsional motion is decoupled from the axial and radial solutions. Once the dispersion equation is extracted from the leading order system, it can be seen that the axial and radial solutions are in fact coupled. The solution to the in vacuo with varying radius problem is first briefly presented and it is then followed by the solution to the fluid inclusion problem with varying radius, which makes up the main part of this section. The solution is studied for various frequencies and at various points along the shell. In addition, the axial and radial components of the first three modes are examined along with their amplitudes and energy distributions. Finally, mean flow is added and the same analysis is carried out, paying particular attention to the differences which arise in comparison to the stationary flow case.

Il est intéressant d'observer qu'une grande partie des objets déformables qui nous entourent sont caractérisés par une forme élancée : soit filiforme, comme les cheveux, les plantes, les fils ; soit surfacique, comme le papier, les feuilles d'arbres, les vêtements ou la plupart des emballages. Simuler (numériquement) la mécanique de telles structures présente alors un intérêt certain : cela permet de prédire leur comportement dynamique, leur forme statique ou encore les efforts qu'elles subissent. Cependant, pour pouvoir réaliser correctement ces simulations, plusieurs problèmes se posent. Les modèles (mécaniques, numériques) utilisés doivent être adaptés aux phénomènes que l'on souhaite reproduire ; le modèle mécanique choisi doit pouvoir être traité numériquement ; enfin, il est nécessaire de connaître les paramètres du modèle qui permettront de reproduire l'instance du phénomène souhaitée. Dans cette thèse nous abordons ces trois points, dans le cadre de la simulation de structures élancées.Dans la première partie, nous proposons un modèle discret de tiges de Kirchhoff dynamiques, de haut degré, basé sur des éléments en courbures et torsion affines par morceaux : les Super-Clothoïdes 3D. Cette discrétisation spatiale est calculée de manière précise grâce à une méthode dédiée, adaptée à l'arithmétique flottante, utilisant des développements en séries entières. L'utilisation des courbures et de la torsion comme degrés de liberté permet d'aboutir à un schéma d'intégration stable grâce à une implicitation, à moindres frais, des forces élastiques. Le modèle a été utilisé avec succès pour simuler la croissance de plantes grimpantes ou le mouvement d'une chevelure. Nos comparaisons avec deux modèles de référence de la littérature ont montré que pour des tiges bouclées, notre approche offre un meilleur compromis en termes de précision spatiale, de richesse de mouvements générés et d'efficacité en temps de calcul.Dans la seconde partie, nous nous intéressons à l'élaboration d'un algorithme capable de retrouver la géométrie au repos (non déformée) d'une coque en contact frottant, connaissant sa forme à l'équilibre et les paramètres physiques du matériau qui la compose. Un tel algorithme trouve son intérêt lorsque l'on souhaite simuler un objet pour lequel on dispose d'une géométrie (numérisée) « à l'équilibre » mais dont on ne connaît pas la forme au repos. En informatique graphique, un exemple d'application est la modélisation de vêtements virtuels sous la gravité et en contact avec d'autres objets : simplement à partir de la forme objectif et d'un simulateur de vêtement, le but consiste à identifier automatiquement les paramètres du simulateur tels que la forme d'entrée corresponde à un équilibre mécanique stable. La formulation d'un tel problème inverse comme un problème aux moindres carrés nous permet de l'attaquer avec la méthode de l'adjoint. Cependant, la multiplicité des équilibres, donnant au problème direct son caractère mal posé, nous conduit à « guider » la méthode en pénalisant les équilibres éloignés de la forme objectif. On montre enfin qu'il est possible de considérer du contact et du frottement solide dans l'inversion, en reformulant le calcul d'équilibres en un problème d'optimisation sous contraintes coniques, et en adaptant la méthode de l'adjoint à ce cas non-régulier. Les résultats que nous avons obtenus sont très encourageants et nous ont permis de résoudre des cas complexes où l'algorithme se comportait de manière intuitive. / It is interesting to observe that many of the deformable objects around us are characterized by a slender structure: either in one dimension, like hair, plants, strands, or in two dimensions, such as paper, the leaves of trees or clothes. Simulating the mechanical behavior of such structures numerically is useful to predict their static shape, their dynamics, or the stress they undergo. However, to perform these simulations, several problems need to be addressed. First, the model (mechanical, numerical) should be adapted to the phenomena which it is aimed at reproducing. Then, the chosen mechanical model should be discretized consistently. Finally, it is necessary to identify the parameters of the model in order to reproduce a specific instance of the phenomenon. In this thesis we shall discuss these three points, in the context of the simulation of slender structures.In the first part, we propose a discrete dynamic Kirchhoff rod model of high degree, based on elements with piecewise affine curvature and twist: the Super-Space-Clothoids. This spatial discretization is computed accurately through a dedicated method, adapted to floating-point arithmetic, using power series expansions. The use of curvature and twist as degrees of freedom allows us to make elastic forces implicit in the integration scheme. The model has been used successfully to simulate the growth of climbing plants or hair motion. Our comparisons with two reference models have shown that in the case of curly rods, our approach offers the best trade-off in terms of spatial accuracy, richness of motion and computational efficiency.In the second part, we focus on identifying the undeformed configuration of a shell in the presence of frictional contact forces, knowing its shape at equilibrium and the physical parameters of the material. Such a method is of utmost interest in Computer Graphics when, for example, a user often wishes to model a virtual garment under gravity and contact with other objects regardless of physics. The goal is then to interpret the shape and provide the right ingredients to the cloth simulator, so that the cloth is actually at equilibrium when matching the input shape. To tackle such an inverse problem, we propose a least squares formulation which can be optimized using the adjoint method. However, the multiplicity of equilibria, which makes our problem ill-posed, leads us to "guide" the optimization by penalizing shapes that are far from the target shape. Finally, we show how it is possible to consider frictional contact in the inversion process by reformulating the computation of equilibrium as an optimization problem subject to conical constraints. The adjoint method is also adjusted to this non-regular case. The results we obtain are very encouraging andhave allowed us to solve complex cases where the algorithm behaves intuitively.