We are concerned with the high-fidelity subsurface imaging of the soil, which commonly arises in geotechnical site characterization and geophysical explorations. Specifically, we attempt to image the spatial distribution of the Lame parameters in semi-infinite, three-dimensional, arbitrarily heterogeneous formations, using surficial measurements of the soil's response to probing elastic waves. We use the complete waveforms of the medium's response to drive the inverse problem. Specifically, we use a partial-differential-equation (PDE)-constrained optimization approach, directly in the time-domain, to minimize the misfit between the observed response of the medium at select measurement locations, and a computed response corresponding to a trial distribution of the Lame parameters. We discuss strategies that lend algorithmic robustness to the proposed inversion schemes. To limit the computational domain to the size of interest, we employ perfectly-matched-layers (PMLs). The PML is a buffer zone that surrounds the domain of interest, and enforces the decay of outgoing waves. In order to resolve the forward problem, we present a hybrid finite element approach, where a displacement-stress formulation for the PML is coupled to a standard displacement-only formulation for the interior domain, thus leading to a computationally cost-efficient scheme. We discuss several time-integration schemes, including an explicit Runge-Kutta scheme, which is well-suited for large-scale problems on parallel computers. We report numerical results demonstrating stability and efficacy of the forward wave solver, and also provide examples attesting to the successful reconstruction of the two Lame parameters for both smooth and sharp profiles, using synthetic records. We also report the details of two field experiments, whose records we subsequently used to drive the developed inversion algorithms in order to characterize the sites where the field experiments took place. We contrast the full-waveform-based inverted site profile against a profile obtained using the Spectral-Analysis-of-Surface-Waves (SASW) method, in an attempt to compare our methodology against a widely used concurrent inversion approach. We also compare the inverted profiles, at select locations, with the results of independently performed, invasive, Cone Penetrometer Tests (CPTs). Overall, whether exercised by synthetic or by physical data, the full-waveform inversion method we discuss herein appears quite promising for the robust subsurface imaging of near-surface deposits in support of geotechnical site characterization investigations.

[pt] É notório que a resistência global de uma estrutura composta por várias camadas coladas depende diretamente da qualidade da adesão entre as mesmas. Imperfeições ao longo das interfaces de adesão comprometem significativamente a performance da estrutura. A caracterização não destrutiva destas imperfeições é geralmente tarefa muito difícil. O principal objetivo deste trabalho é apresentar um método analítico-numérico que permite modelar o campo acústico resultante da interação entre ondas ultra-sônicas e interfaces imperfeitas, auxiliando na escolha de parâmetros para o emprego de métodos ultra-sônicos de inspeção. No trabalho, a aproximação quase-estática, proposta por Thompson em meados da década de oitenta,é combinada com o método das perturbações para a modelagem de defeitos localizados ao longo da camada de adesão. O método desenvolvido admite que as camadas sejam feitas de materiais anisotrópicos, e permite a modelagem em altas frequências. Três simulações são apresentadas para exemplificar a aplicação do método. Resultados destas simulações onde feixes ultra-sônicos são incidentes em placas imersas em água revelam frequências e ângulos de incidência para os quais efeitos de espalhamento, que permitem a caracterização de defeitos localizados, são mais significativos. / [en] It is well known that the global strength of multi-layered composite structures strongly depends on the quality of the adhesion between its constituent elements. Imperfections along interfaces of adhesion can strongly compromise structure s performance. The characterization of such defects is a very difficult task. The main goal of this study is the development of an analytic-numerical method to simulate the acoustic field resulting from the interaction between ultrasonic waves and imperfect interfaces, helping in selection of parameters for ultra-sonic inspecting methods. The Quasistatic-approximation (QSA), introduced by Thompson in 1982, is applied together with the perturbation method to allow modelling of interfacial localized aws. A solution algorithm for the problem is developed with the aid of the invariant embedding method. It is applicable to solve wave propagation problems in arbitrarily anisotropic layered plates and it is stable for high frequencies. Three simulations of multi-layered plates immersed in acoustic uid are presented as illustration of the application of the developed method. Results of those simulations indicate the frequencies and angles of incidence where the scattering effects, which allow the characterization of localized defects, are more significant.

Neste trabalho foram desenvolvidas as funções de interpolação com regularidades C1 e C2, utilizando o Método da Partição de Unidade, referentes ao elemento quadrilateral de quatro nós. Estes elementos quadrilaterais Hermitianos de regularidade C1 e C2 foram implementados em uma plataforma própria de elementos finitos, considerando uma estratégia do tipo sub-paramétrica. De forma comparativa com os elementos Lagrangeanos de regularidade C0 e diferentes ordens polinomiais, os elementos de regularidade C1 e C2 foram aplicados na solução de: problemas clássicos de elasticidade plana infinitesimal isotrópica; aproximação das frequências naturais de vibração livre de barras e viga; pro- pagação de onda elástica em barra devido à aplicação de força impulsiva. Os resultados obtidos mostraram que foi possível se obter um maior percentual de frequências naturais aproximadas do espectro discreto, dado um certo nível de erro máximo, com os elementos de regularidade C1 e C2 em comparação com os elementos Lagrangeanos de regularidade C0 de quatro, oito, dezesseis e vinte e cinco nós. Quanto ao problema de propagação de onda elástica devido à aplicação de força impulsiva, as soluções obtidas com os elementos de regularidade C1 e C2 também apresentaram-se satisfatórias em relação à solução ana- lítica e às soluções aproximadas obtidas com os elementos Lagrangeanos de regularidade C0 de quatro e oito nós. Por outro lado, nas simulações dos problemas de elasticidade plana infinitesimal isotrópica, os elementos de regularidade C1 e C2 não apresentaram um comportamento satisfatório. Os erros relativos em normas L2 e de energia da solução aproximada foram maiores do que aqueles obtidos com o elemento Lagrangeano de regularidade C0 de oito nós, por exemplo, e as taxas de convergência em norma de energia obtidas com tais elementos foram inferiores às preditas pelo estimador de erro a priori. / In this work the shape functions with regularity C1 e C2 were developed, by means of the Partition of Unity Method, concerning to the four-node quadrilateral element. These Hermitian quadrilateral elements with regularity C1 e C2 were implemented in an own platform of finite elements, considering the subparametric strategy. Comparatively with the C 0 regularity Lagrangian elements of different polynomial order, C1 and C2 regularity elements were applied in simulations of: classical isotropic infinitesimal plane elasticity problems; approximation of natural frequencies of free vibration for bars and beam; elastic wave propagation in bar caused by forced vibration with impulsive loading applied. The results obtained showed that was possible to get a major number of natural frequencies of free vibration for the discrete spectrum, given a certain level of error, for C1 and C2 regularity elements in comparison with C 0 regularity Lagrangian elements of four, eight, sixteen and twenty-five nodes. Regarding to the elastic wave propagation problem in bar due to the application of impulsive loading, the solution obtained with C1 and C2 regularity elements also presented satisfactory results with relation to the analytical solution and those obtained with C 0 regularity Lagrangian elements with four and eight nodes. On the other hand, for isotropic infinitesimal plane elasticity problems, C1 and C2 regularity elements did not present satisfactory results. Relative errors in L2 and energy norms of approximate solution were greater than those computed for the C 0 Lagrangian element of eight nodes, for example, and convergence rates obtained with the C1 and C2 regularity elements were lower than those predicted by the a priori error estimator.

Neste trabalho foram desenvolvidas as funções de interpolação com regularidades C1 e C2, utilizando o Método da Partição de Unidade, referentes ao elemento quadrilateral de quatro nós. Estes elementos quadrilaterais Hermitianos de regularidade C1 e C2 foram implementados em uma plataforma própria de elementos finitos, considerando uma estratégia do tipo sub-paramétrica. De forma comparativa com os elementos Lagrangeanos de regularidade C0 e diferentes ordens polinomiais, os elementos de regularidade C1 e C2 foram aplicados na solução de: problemas clássicos de elasticidade plana infinitesimal isotrópica; aproximação das frequências naturais de vibração livre de barras e viga; pro- pagação de onda elástica em barra devido à aplicação de força impulsiva. Os resultados obtidos mostraram que foi possível se obter um maior percentual de frequências naturais aproximadas do espectro discreto, dado um certo nível de erro máximo, com os elementos de regularidade C1 e C2 em comparação com os elementos Lagrangeanos de regularidade C0 de quatro, oito, dezesseis e vinte e cinco nós. Quanto ao problema de propagação de onda elástica devido à aplicação de força impulsiva, as soluções obtidas com os elementos de regularidade C1 e C2 também apresentaram-se satisfatórias em relação à solução ana- lítica e às soluções aproximadas obtidas com os elementos Lagrangeanos de regularidade C0 de quatro e oito nós. Por outro lado, nas simulações dos problemas de elasticidade plana infinitesimal isotrópica, os elementos de regularidade C1 e C2 não apresentaram um comportamento satisfatório. Os erros relativos em normas L2 e de energia da solução aproximada foram maiores do que aqueles obtidos com o elemento Lagrangeano de regularidade C0 de oito nós, por exemplo, e as taxas de convergência em norma de energia obtidas com tais elementos foram inferiores às preditas pelo estimador de erro a priori. / In this work the shape functions with regularity C1 e C2 were developed, by means of the Partition of Unity Method, concerning to the four-node quadrilateral element. These Hermitian quadrilateral elements with regularity C1 e C2 were implemented in an own platform of finite elements, considering the subparametric strategy. Comparatively with the C 0 regularity Lagrangian elements of different polynomial order, C1 and C2 regularity elements were applied in simulations of: classical isotropic infinitesimal plane elasticity problems; approximation of natural frequencies of free vibration for bars and beam; elastic wave propagation in bar caused by forced vibration with impulsive loading applied. The results obtained showed that was possible to get a major number of natural frequencies of free vibration for the discrete spectrum, given a certain level of error, for C1 and C2 regularity elements in comparison with C 0 regularity Lagrangian elements of four, eight, sixteen and twenty-five nodes. Regarding to the elastic wave propagation problem in bar due to the application of impulsive loading, the solution obtained with C1 and C2 regularity elements also presented satisfactory results with relation to the analytical solution and those obtained with C 0 regularity Lagrangian elements with four and eight nodes. On the other hand, for isotropic infinitesimal plane elasticity problems, C1 and C2 regularity elements did not present satisfactory results. Relative errors in L2 and energy norms of approximate solution were greater than those computed for the C 0 Lagrangian element of eight nodes, for example, and convergence rates obtained with the C1 and C2 regularity elements were lower than those predicted by the a priori error estimator.

Neste trabalho foram desenvolvidas as funções de interpolação com regularidades C1 e C2, utilizando o Método da Partição de Unidade, referentes ao elemento quadrilateral de quatro nós. Estes elementos quadrilaterais Hermitianos de regularidade C1 e C2 foram implementados em uma plataforma própria de elementos finitos, considerando uma estratégia do tipo sub-paramétrica. De forma comparativa com os elementos Lagrangeanos de regularidade C0 e diferentes ordens polinomiais, os elementos de regularidade C1 e C2 foram aplicados na solução de: problemas clássicos de elasticidade plana infinitesimal isotrópica; aproximação das frequências naturais de vibração livre de barras e viga; pro- pagação de onda elástica em barra devido à aplicação de força impulsiva. Os resultados obtidos mostraram que foi possível se obter um maior percentual de frequências naturais aproximadas do espectro discreto, dado um certo nível de erro máximo, com os elementos de regularidade C1 e C2 em comparação com os elementos Lagrangeanos de regularidade C0 de quatro, oito, dezesseis e vinte e cinco nós. Quanto ao problema de propagação de onda elástica devido à aplicação de força impulsiva, as soluções obtidas com os elementos de regularidade C1 e C2 também apresentaram-se satisfatórias em relação à solução ana- lítica e às soluções aproximadas obtidas com os elementos Lagrangeanos de regularidade C0 de quatro e oito nós. Por outro lado, nas simulações dos problemas de elasticidade plana infinitesimal isotrópica, os elementos de regularidade C1 e C2 não apresentaram um comportamento satisfatório. Os erros relativos em normas L2 e de energia da solução aproximada foram maiores do que aqueles obtidos com o elemento Lagrangeano de regularidade C0 de oito nós, por exemplo, e as taxas de convergência em norma de energia obtidas com tais elementos foram inferiores às preditas pelo estimador de erro a priori. / In this work the shape functions with regularity C1 e C2 were developed, by means of the Partition of Unity Method, concerning to the four-node quadrilateral element. These Hermitian quadrilateral elements with regularity C1 e C2 were implemented in an own platform of finite elements, considering the subparametric strategy. Comparatively with the C 0 regularity Lagrangian elements of different polynomial order, C1 and C2 regularity elements were applied in simulations of: classical isotropic infinitesimal plane elasticity problems; approximation of natural frequencies of free vibration for bars and beam; elastic wave propagation in bar caused by forced vibration with impulsive loading applied. The results obtained showed that was possible to get a major number of natural frequencies of free vibration for the discrete spectrum, given a certain level of error, for C1 and C2 regularity elements in comparison with C 0 regularity Lagrangian elements of four, eight, sixteen and twenty-five nodes. Regarding to the elastic wave propagation problem in bar due to the application of impulsive loading, the solution obtained with C1 and C2 regularity elements also presented satisfactory results with relation to the analytical solution and those obtained with C 0 regularity Lagrangian elements with four and eight nodes. On the other hand, for isotropic infinitesimal plane elasticity problems, C1 and C2 regularity elements did not present satisfactory results. Relative errors in L2 and energy norms of approximate solution were greater than those computed for the C 0 Lagrangian element of eight nodes, for example, and convergence rates obtained with the C1 and C2 regularity elements were lower than those predicted by the a priori error estimator.

Les objectifs de cette thèse sont la modélisation mathématique et la simulation numérique de l’élastographie impulsionnelle basée sur la force de radiation acoustique (FRA) dans un tissu mou précontraint, et en particulier le myocarde. La première partie du manuscript concerne la modélisation mathématique de la FRA, la propagation d’ondes de cisaillement qui en résulte et la caractérisation de la vitesse des ondes de cisaillement pour une loi de comportement générale du tissu myocardique. Nous montrons aussi des applications pour l’estimation de l’orientation des fibres cardiaques dans le myocarde et l’évaluation de “pathologies synthétiques ”. Une des contributions principales de ce travail est le développement d’un modèle mathématique original de la FRA. En particulier, à partir d’un modèle biomécanique tridimensionnel du coeur, nous obtenons, à travers une approche asymptotique, les équations qui régissent les champs de pression et de cisaillement induits par la FRA. De plus, nous calculons une expression analytique du terme source responsable de la génération des ondes de cisaillement à partir d’une impulsion acoustique en pression. Dans la deuxième partie de la thèse, nous proposons des outils numériques efficaces pour une simulation numérique réaliste d’une expérience d’élastographie impulsionnelle dans un tissu quasi-incompressible, précontraint et fibré. La discrétisation en espace se base sur des éléments finis spectraux d’ordre élevé. Pour la discrétisation en temps, nous proposons une nouvelle méthode adaptée à l’élasticité incompressible. En particulier, seuls les termes correspondant à des vitesses infinies, associés à la contrainte d’incompressibilité, sont traités implicitement, à travers la resolution d’un problème de Poisson à chaque pas de temps de l’algorithme. En outre, nous proposons une nouvelle méthode d’ordre élevé et efficace pour la résolution d’un problème de Poisson, qui se base sur la transformée de Fourier discrète. / This PhD thesis concerns the mathematical modelling and numerical simulation of impulsive Acoustic Radiation Force (ARF)-driven Shear Wave Elastography (SWE) imaging in a prestressed soft tissue, with a specific reference to the cardiac setting. The first part of the manuscript deals with the mathematical modelling of the ARF, the resulting shear wave propagation, and the characterisation of the shear wave velocity in a general constitutive law for the myocardial tissue. We also show some applications to the extraction of fibre orientation in the myocardium and the detection of “synthetic pathologies”. One of the main contributions of this work is the derivation of an original mathematical model of the ARF. In more detail, starting from an accurate biomechanical model of the heart, and based on asymptotic analysis, we infer the governing equation of the pressure and the shear wave field remotely induced by the ARF, and we compute an analytical expression of the source term responsible for the generation of shear waves from an acoustic pressure pulse. In the second part of the PhD thesis, we propose efficient numerical tools for a realistic numerical simulation of an SWE experiment in a nearly-incompressible, pre-stressed, fibered soft tissue. The spatial discretisation is based on high-order Spectral Finite Elements (HO-SEM). Concerning the time discretisation, we propose a novel method adapted to incompressible elasticity. In particular, only the terms travelling at infinite velocity, associated with the incompressibility constraint, are treated implicitly by solving a scalar Poisson problem at each time step of the algorithm. Furthermore, we provide a novel matrix-free, high-order, fast method to solve the Poisson problem, based on the use of the Discrete Fourier Transform.

[pt] Ondas acústicas, sônicas ou ultrassônicas, podem ser empregadas para a telemetria sem fio como alternativa a sistemas eletromagnéticos, transferindo dados e energia ao longo de um canal formado por uma ou mais camadas de sólidos elásticos ou fluidos acústicos. Um exemplo é a interrogação de sensores passivos através de uma parede metálica. Nesta configuração, pelo menos um transdutor acústico é fixado em um lado da parede (face externa), onde uma fonte de alimentação elétrica é disponível. No lado oposto (face interna), onde os sensores estão instalados, são fixados um ou mais transdutores. Na maioria das aplicações estes transdutores são cerâmicas piezelétricas que geram e recebem sinais ultrassônicos. Ondas acústicas ultrassônicas se propagam ao longo do sólido elástico, transferindo energia e dados entre as duas faces, possibilitando a alimentação e interrogação dos sensores. Este tipo de configuração pode ser empregado em aplicações onde o uso de penetradores elétricos ou ópticos não é recomendado. Entretanto, a resposta das piezocerâmicas pode sofrer influências de variações de temperatura e da própria deformação mecânica da parede metálica na qual são fixados. O presente trabalho procurou quantificar a influência da deformação mecânica e da variação de temperatura na comunicação entre dois transdutores piezocerâmicos ultrassônicos, aderidos à uma placa metálica por meio de adesivo epóxi. No estudo, tomou-se como parâmetro quantitativo o sinal S21, que é o logaritmo da razão entre a potência recebida pela saída do sistema (face interna da parede) pela potência transmitida pela entrada (face externa da parede). O trabalho apresenta comparações entre resultados experimentais e simulados através de um modelo numérico de elementos finitos desenvolvido no COMSOL Multiphysics. Os ensaios experimentais foram realizados com pastilhas piezocerâmicas circulares, do tipo PZT4, com diâmetro e espessura de 25 e 2 mm, respectivamente. Os transdutores foram fixados, de forma concentricamente alinhada e por meio de um adesivo epóxi, nas duas superfícies de uma placa de aço inoxidável AISI 316 L com 6 mm de espessura. O trabalho apresenta tabelas e funções para a amplitude do sinal S21 na frequência onde a transferência de potência é maximizada. Para os casos estudados, observou-se que a frequência ideal muda muito pouco com a temperatura ou a deformação da placa sobre a qual os transdutores são fixados, permanecendo com valores entre 0,988 e 0,995 MHz em todas as condições avaliadas. Em função da deformação da placa metálica, a amplitude do sinal S21 também variou muito pouco, de -3,70 para -3,14 dB, desde a condição indeformada da placa até a máxima deformação aplicada, que foi de 1250 (Micro)m/m. Quanto à variação com a temperatura, na faixa de 30 a 100 Graus C, mais uma vez observou-se apenas um pequeno aumento de 0,8 dB na amplitude do sinal S21. Entretanto, para temperaturas acima de 100 Graus C, o sinal passa a cair rapidamente. Em nenhuma das condições estudadas neste trabalho foi observado prejuízo na transferência de potência entre os transdutores, indicando que este tipo de comunicação pode ser uma alternativa robusta ao uso de penetradores elétricos. / [en] Acoustic, sonic or ultrasonic waves can be used for wireless telemetry as an alternative to electromagnetic systems, transferring data and energy along a channel formed by one or more layers of elastic solids or acoustic fluids. An example of this is the interrogation of passive sensors through a metallic wall. In this configuration, at least one acoustic transducer is attached to one face of the wall (external face) where electrical power supply is available. One or more transducers are also attached to its other side (internal face) where the sensors are installed. In most applications, these transducers are piezoelectric ceramics that generate and receive ultrasonic signals. Ultrasonic acoustic waves propagate along the elastic solid, transferring energy and data between both sides, which enables the power supply and interrogation of the sensors. This type of configuration can be used in applications where the use of an electrical or optical penetrator is not suitable. However, the response of piezoceramics may be affected by temperature variations and mechanical deformations of the metallic wall on which they are attached. The present work sought to quantify the influence of mechanical deformation and temperature changes on the communication between two ultrasonic piezo ceramic transducers, adhered to a metal plate by using an epoxy adhesive. The parameter used to quantify this influence was the S21 signal, which is the logarithm of the ratio between the power received from the output of the system (internal face of the wall) to the power transmitted by the input (external face of the wall). The work presents comparisons between experimental and simulated results obtained by using a finite element model developed through the commercial software COMSOL Multiphysics. In the configuration experimentally tested, two PZT-4 disks with diameter and thickness of, respectively, 25 and 2 mm were concentrically attached to both sides of a 6 mm thick, AISI 316 L stainless steel plate. Amplitudes of the S21 signal measured at the frequency where power transfer is maximized were obtained for different temperature and strain levels. Results for all of the evaluated conditions showed that the impedance matching frequency suffers little influence from temperature variations or strain in the plate on which the transducers are attached, having remained within a range from 0.988 to 0.995 MHz in all tests. As mechanical strains were applied to the metal plate, the amplitude of the S21 signal varied from -3.70 dB to -3.14 dB, from the undeformed condition to the maximum applied deformation (1250 (Micro)m/m). Regarding temperature changes, a small increase of 0.8 dB in the amplitude of the S21 signal was observed when increasing temperature from 30 C Degrees to 100 C Degrees. However, for temperatures above 100 C Degrees, the signal was found to quickly decay. None of the conditions studied in this work brought any impairment to the power transfer between the transducers, indicating that this type of communication can be a robust alternative to electrical penetrators.

Les travaux de la présente thèse portent sur l’estimation d'erreur a posteriori pour les solutions numériques par éléments finis de l'équation des ondes élastiques dans les milieux hétérogènes. Deux types d’estimation ont été développés. Le premier considère directement l’équation élastodynamique et conduit à un nouvel estimateur d'erreur a posteriori explicite en norme L∞ en temps. Les principales caractéristiques de cet estimateur explicite sont l'utilisation de la méthode de résidus et le développement de reconstructions en temps et en espace selon les différentes régularités exigées par les différents termes contribuant à l’obtention d’une borne supérieure. L’analyse numérique de cet estimateur dans le cas des maillages uniformes montre qu’il assure bien une borne supérieure mais avec une propriété asymptotique qui reste à améliorer. Le deuxième type d’estimateur d’erreur est développé dans le contexte de la propagation des ondes à haute fréquence dans des milieux hétérogènes à l’échelle mésoscopique. Il s’agit d’une nouvelle erreur en résidus basée sur l'équation de transfert radiatif, qui est obtenue par un développement asymptotique multi-échelle de l'équation d'onde en utilisant la transformation de Wigner en espace-temps. Les résidus sont exprimés en termes de densités énergétiques calculés dans l’espace des phases pour les solutions d’onde numériques transitoires par éléments finis. L’analyse numérique de cette erreur appliquée aux milieux homogènes et hétérogènes en 1D a permis de valider notre approche. Les champs d’application visés sont la propagation des ondes sismiques dans les milieux géophysiques ou la propagation des ondes ultrasonores dans les milieux polycristallins. / This thesis work deals with a posteriori error estimates for finite element solutions of the elastic wave equation in heterogeneous media. Two different a posteriori estimation approaches are developed. The first one, in a classical way, considers directly the elastodynamic equation and results in a new explicit error estimator in a non-natural L∞ norm in time. Its key features are the use of the residual method and the development of space and time reconstructions with respect to regularities required by different residual operators contributing to the proposed error bound. Numerical applications of the error bound with different mesh sizes show that it gives rise to a fully computable upper bound. However, its effectivity index and its asymptotic accuracy remain to be improved. The second error estimator is derived for high frequency wave propagation problem in heterogeneous media in the weak coupling regime. It is a new residual-type error based on the radiative transfer equation, which is derived by a multi-scale asymptotic expansion of the wave equation in terms of the spatio-temporal Wigner transforms of wave fields. The residual errors are in terms of angularly resolved energy quantities of numerical solutions of waves by finite element method. Numerical calculations of the defined errors in 1D homogeneous and heterogeneous media allow validating the proposed error estimation approach. The application field of this work is the numerical modelling of the seismic wave propagation in geophysical media or the ultrasonic wave propagation in polycrystalline materials.

La compréhension et l’amélioration de l’environnement vibratoire des charges utiles demande la mise au point de démarches prédictives maîtrisées qui permettent de comprendre les phénomènes de transmission des ondes de chocs d’origine pyrotechnique dans le lanceur Ariane5. Plus particulièrement, la maîtrise du comportement transitoire des coques sandwichs en nid d’abeilles, principaux constituants de l’Adaptateur de Charges Utiles – structure porteuse des satellites, est nécessaire pour prédire les vibrations au pied des équipements électroniques des satellites et des lanceurs. Cette problématique présente un caractère multi-échelle tant d’un point de vue temporel (charge mobile supersonique, temps d’analyse) que spatial (dimensions des structures du lanceur, taille des cellules en nid d’abeilles, longueurs d’ondes liées aux hautes fréquences). Celui-ci a été traité dans cette thèse en s’appuyant d’une part, sur une qualification à la fois analytique et numérique des modèles classiques homogénéisés des plaques sandwichs en nid d’abeilles pour la gamme de fréquence mise en jeu et d’autre part, sur une application des stratégies de remaillage adaptatif pour la propagation des ondes développées dans le cadre de la méthode de Galerkin espace-temps discontinue en temps. Deux catégories de modèles de plaques épaisses ont été ainsi construites dans le but d’enrichir la cinématique classique de plaques épaisses de Mindlin-Reissner qui s’est avérée être insuffisante pour correctement représenter le comportement dynamique hors-plan des plaques sandwich en nid d’abeilles. Ainsi ont été analysés les modèles dits monocouches basés sur un enrichissement de la cinématique par ajout de degrés de liberté dans l’épaisseur, et les modèles multicouches composés d’une superposition de trois plaques avec une homogénéisation séparée des matériaux. Il a été montré que ces deux sortes de modèles améliorent la description des phénomènes de hautes fréquences, notamment ceux de flexion et de cisaillement transverse qui sont plus délicats à retranscrire. Toutes les études numériques ont été effectuées avec un code éléments finis qui emploie des solveurs adaptatifs dynamiques basés sur la méthode de Galerkin espace-temps discontinue en temps. Cette méthode d’intégration en temps introduit un amortissement numérique dépendant du pas de temps et qui peut interférer avec un amortissement physique susceptible d’être introduit dans un modèle numérique et conduire au final à un amortissement total différent de celui qui est attendu. Cette interaction a été analysée et mise en évidence dans ce travail à travers l’introduction de l’amortissement de Rayleigh dans les modèles de propagation de chocs. Les outils et les modèles de propagation ainsi développés ont été validés sur plusieurs structures académiques et industrielles. Des comparaisons avec des données expérimentales sur des structures industrielles de grande taille, plus particulièrement sur un Adaptateur de Charges Utiles d’Ariane5, sont effectuées et soulignent la cohérence de notre approche ainsi que la fiabilité et l’efficacité des modèles de propagation proposés. / Reliable and efficient numerical models for the pyrotechnic shock wave propagation in structures of the Ariane5 launcher are necessary for a good understanding and a predictive analysis of the payload vibration environment. More precisely, the correct modeling of the dynamic behaviour of the honeycomb sandwich shells, the main material composing the payload adaptor, is essential to control the vibration environment of the payload and the embarked electronic equipments and so to prevent them from damages caused by the shock wave propagation. The topic is obviously a multi-scale problem from both temporal and spatial points of view : short time intervals imposed by supersonic moving loads vs. large total time interval that the slowest waves need to travel throughout the adaptor ; very short wavelengths of high frequency waves, and very small size of the honeycomb cells vs. large structure dimensions. To take into account all involved space-time scales in a reliable and efficient way, the herein study is based both on the analytical and numerical qualification of the classical homogenized models of honeycomb sandwich shells for the frequency range introduced by the pyrotechnic shock wave, and on a dynamic solver based on the well-known space-time discontinuous Galerkin method, allowing the use of adaptive remeshes for the wave propagation. The classical Mindlin-Reissner’s kinematics of thick plates being inefficient to correctly represent the dynamic out-of-plane behaviour of the honeycomb sandwich plates, two kinds of its enrichment are considered : One-layered models based on an enrichment of the kinematics by adding degrees of freedom in the thickness, and multi-layered models composed of a superposition of three plates with separated material homogenisations. It has been shown theoretically and numerically that, both types of enrichment allow more precise descriptions of flexure and transverse shear modes in the high frequency range. However, the multi-layered models give much more promising results, as the important role played by the honeycomb core for the transverse shear behaviour of the whole sandwich is not “smeared” in a one-layered homogenized model. All the numerical studies were conducted with a finite element code which uses a dynamic solverbased on the time discontinuous space-time Galerkin method. The built-in numerical damping of this solver can interfere with a physical damping potentially introduced by the numerical model and results in a global damping totally unexpected. This interaction has been analysed and underlined in this work thanks to the introduction of the Rayleigh damping in the shock wave propagation models. Theoretical and numerical tools and propagating models thus developed have been validated on several academic and industrial structures. Comparison with experimental data on large size industrial structures, especially a real size payload adaptor, is performed and emphasizes the coherence of our approach and the reliability and the efficiency of the proposed propagating models.