This work is dedicated to the control of the accuracy of computational simulations of sound propagation and scattering. Assuming time-harmonic behaviour, the mathematical models are given as boundary value problems for the Helmholtz equation <i>Delta u+k2u=0 </i> in <i>Oméga</i>. A distinction is made between interior, exterior and coupled problems and this work focuses mainly on interior uncoupled problems for which the Helmholtz equation becomes singular at eigenfrequencies. As in other application fields, error control is an important issue in acoustic computations. It is clear that the numerical parameters (mesh size h and degree of approximation p) must be adapted to the physical parameter k. The well known ‘rule of the thumb’ for the h version with linear elements is to resolve the wavelength <i>lambda=2 pi k-1</i> by six elements characterising the approximability of the finite element mesh. If the numerical model is stable, the quality of the numerical solution is entirely controlled by the approximability of the finite element mesh. The situation is quite different in the presence of singularities. In that case, <i>stability</i> (or the lack thereof) is equally (sometimes more) important. In our application, the solutions are ‘rough’, i.e., highly oscillatory if the wavenumber is large. This is a singularity inherent to the differential operator rather than to the domain or the boundary conditions. This effect is called the <i>k-singularity</i>. Similarly, the discrete operator (“stiffness” matrix) becomes singular at eigenvalues of the discretised interior problem (or nearly singular at damped eigenvalues in solid-fluid interaction). This type of singularities is called the <i>lambda-singularities</i>. Both singularities are of global character. Without adaptive correction, their destabilizing effect generally leads to large error of the finite element results, even if the finite element mesh satisfies the ‘rule of the thumb’. The k- and lambda-singularities are first extensively demonstrated by numerical examples. Then, two <i>a posteriori</i> error estimators are developed and the numerical tests show that, due to these specific phenomena of dynamo-acoustic computations, <i>error control cannot, in general, be accomplished by just ‘transplanting’ methods that worked well in static computations</i>. However, for low wavenumbers, it is necessary to also control the influence of the geometric (reentrants corners) or physical (discontinuities of the boundary conditions) singularities. An <i>h</i>-adaptive version with refinements has been implemented. These tools have been applied to two industrial examples : the GLT, a bi-mode bus from Bombardier Eurorail, and the Vertigo, a sport car from Gillet Automobiles. As a conclusion, it is recommanded to replace the rule of the thumb by a criterion based on the control of the influence of the specific singularities of the Helmholtz operator. As this aim cannot be achieved by the <i>a posteriori</i> error estimators, it is suggested to minimize the influence of the singularities by modifying the formulation of the finite element method or by formulating a “meshless” method.

La plupart des simulations interactives ont besoin d'un modèle de détection de collisions. Cette détection nécessite d'une part d'effectuer des requêtes de proximité entre les entités concernées et d'autre part de calculer un comportement à appliquer. Afin d'effectuer ces requêtes, les entités présentes dans une scène sont soit hiérarchisées dans un arbre ou dans un graphe de proximité, soit plongées dans une grille d'enregistrement. Nous présentons un nouveau modèle de détection de collisions s'appuyant sur deux piliers : une représentation de l'environnement par des cartes combinatoires multirésolutions et un suivi en temps réel de particules plongées dans ces cartes. Ce modèle nous permet de représenter des environnements complexes tout en suivant en temps réel les entités évoluant dans cet environnement. Nous présentons des outils d'enregistrement et de maintien de l'enregistrement de particules, d'arêtes et de surfaces dans des cartes combinatoires volumiques multirésolutions. / Most interactive simulations need a collision detection system. First, this system requires the querying of the proximity between the objects and then the computing of the behaviour to be applied. In order to perform these queries, the objects present in a scene are either classified in a tree, in a proximity graph, or embedded inside a registration grid.Our work present a new collision detection model based on two main concepts: representing the environment with a combinatorial multiresolution map, and tracking in real-time particles embedded inside this map. This model allows us to simulate complex environments while following in real-time the entities that are evolving within it.We present our framework used to register and update the registration of particles, edges and surfaces in volumetric combinatorial multiresolution maps. Results have been validated first in 2D with a crowd simulation application and then in 3D, in the medical field, with a percutaneous surgery simulation.

Ce document synthétise un ensemble de travaux portant sur le développement et l'analyse de méthodes de volumes finis utilisées pour l'approximation numérique d'équations aux dérivées partielles issues de la physique. Le mémoire aborde dans sa première partie des schémas colocalisés de type Godunov d'une part pour les équations de l'électromagnétisme, et d'autre part pour l'équation des ondes acoustiques, avec une étude portant sur la perte de précision de ce schéma à bas nombre de Mach. La deuxième partie est consacrée à la construction d'opérateurs différentiels discrets sur des maillages bidimensionnels relativement quelconques, en particulier très déformés ou encore non-conformes, et à leur utilisation pour la discrétisation d'équations aux dérivées partielles modélisant des phénomènes de diffusion, d'électrostatique et de magnétostatique et d'électromagnétisme par des schémas de type volumes finis en dualité discrète (DDFV) sur maillages décalés. La troisième partie aborde ensuite l'analyse numérique et les estimations d'erreur a priori et a posteriori associées à la discrétisation par le schéma DDFV de l'équation de Laplace. La quatrième et dernière partie est consacrée à la question de l'ordre de convergence en norme $L^2$ de la solution numérique du problème de Laplace, issue d'une discrétisation volumes finis en dimension un et en dimension deux sur des maillages présentant des propriétés d'orthogonalité. L'étude met en évidence des conditions nécessaires et suffisantes relatives à la géométrie des maillages et à la régularité des données du problème afin d'obtenir la convergence à l'ordre deux de la méthode.

This work is dedicated to the control of the accuracy of computational simulations of sound propagation and scattering. Assuming time-harmonic behaviour, the mathematical models are given as boundary value problems for the Helmholtz equation <i>Delta u+k2u=0 </i> in <i>Oméga</i>. A distinction is made between interior, exterior and coupled problems and this work focuses mainly on interior uncoupled problems for which the Helmholtz equation becomes singular at eigenfrequencies. <p><p>As in other application fields, error control is an important issue in acoustic computations. It is clear that the numerical parameters (mesh size h and degree of approximation p) must be adapted to the physical parameter k. The well known ‘rule of the thumb’ for the h version with linear elements is to resolve the wavelength <i>lambda=2 pi k-1</i> by six elements characterising the approximability of the finite element mesh. If the numerical model is stable, the quality of the numerical solution is entirely controlled by the approximability of the finite element mesh. The situation is quite different in the presence of singularities. In that case, <i>stability</i> (or the lack thereof) is equally (sometimes more) important. In our application, the solutions are ‘rough’, i.e. highly oscillatory if the wavenumber is large. This is a singularity inherent to the differential operator rather than to the domain or the boundary conditions. This effect is called the <i>k-singularity</i>. Similarly, the discrete operator (“stiffness” matrix) becomes singular at eigenvalues of the discretised interior problem (or nearly singular at damped eigenvalues in solid-fluid interaction). This type of singularities is called the <i>lambda-singularities</i>. Both singularities are of global character. Without adaptive correction, their destabilizing effect generally leads to large error of the finite element results, even if the finite element mesh satisfies the ‘rule of the thumb’. <p><p>The k- and lambda-singularities are first extensively demonstrated by numerical examples. Then, two <i>a posteriori</i> error estimators are developed and the numerical tests show that, due to these specific phenomena of dynamo-acoustic computations, <i>error control cannot, in general, be accomplished by just ‘transplanting’ methods that worked well in static computations</i>. However, for low wavenumbers, it is necessary to also control the influence of the geometric (reentrants corners) or physical (discontinuities of the boundary conditions) singularities. An <i>h</i>-adaptive version with refinements has been implemented. These tools have been applied to two industrial examples :the GLT, a bi-mode bus from Bombardier Eurorail, and the Vertigo, a sport car from Gillet Automobiles.<p><p>As a conclusion, it is recommanded to replace the rule of the thumb by a criterion based on the control of the influence of the specific singularities of the Helmholtz operator. As this aim cannot be achieved by the <i>a posteriori</i> error estimators, it is suggested to minimize the influence of the singularities by modifying the formulation of the finite element method or by formulating a “meshless” method.<p> / Doctorat en sciences appliquées / info:eu-repo/semantics/nonPublished