L'objectif de cette thèse consiste en l'étude d'une stratégie de contrôle actif à faible autorité avec comme application le contrôle actif du rayonnement acoustique d'une plaque. Depuis l'essor du contrôle actif, son application aux problèmes acoustiques et vibracoustiques a été investiguée par de nombreux chercheurs, exploitant soit la théorie du contrôle optimal, soit des approches originales basées plus particulièrement sur la physique. Des notions spécifiques au contrôle vibroacoustique ont été développées comme, par exemple, les modes radiatifs pouvant caractériser le rayonnement acoustique d'une plaque d'une manière adaptée au contrôle. Le contrôle actif à faible autorité, pour lequel le Laboratoire de Structures Actives a développé une expertise dans le domaine de l'amortissement et du contrôle actif des vibrations, est une solution attractive par sa simplicité de mise en oeuvre. Le plus souvent implémenté sous la forme d'un contrôle décentralisé constitué de boucles indépendantes, le contrôle à faible autorité bénéficie de certaines garanties de stabilité et de robustesse. Bien que notre stratégie de contrôle puisse s'appliquer à n'importe quel type de plaque, l'application considérée dans ce travail a été motivée par le contexte socio-économique actuel en rapport avec les nuisances acoustiques. Il était en effet intéressant d'évaluer la stratégie de contrôle pour le problème de la transmission acoustique d'un vitrage. La stratégie de contrôle se divise en deux étapes. Tout d'abord le développement d'un capteur unique destiné à fournir une mesure représentative du bruit rayonné par une plaque en basse fréquence. Deux capteurs de vitesse volumétrique (l'un discret, l'autre distribué) ont ainsi été développés et évalués expérimentalement. Ensuite, une procédure d'optimisation de l'emplacement d'un ensemble d'actionneurs pilotés en parallèle est proposée. L'objectif de cette phase d'optimisation est de forcer la réponse fréquentielle du système à posséder les propriétés d'un système colocalisé. La stratégie de contrôle est ensuite évaluée sur deux structures expérimentales. / This thesis is concerned with a low authority active control strategy applied to the sound radiation control of a baffled plate. Since the development of active control , numerous researchers have studied its application to acoustical or vibroacoustical problems using either the modern control theory or other methods based rather on the understanding of the physics of the problem. Vibroacoustical active control has lead to the definition of radiation modes allowing to describe the radiated sound of a plate in an appropriate manner for active control purposes. Low autorithy control (LAC), for which the Active Structures Laboratory has gained an expertise for active vibration control applications is an interesting solution for its implementation simplicity. Most of the time it consists of several decentralized control loops, and benefits from guaranteed stability and robustness properties. Although our control strategy can be applied to any kind of plates, the application considered here has been motivated by the present socio-economical context related to noise annoyances. The active control strategy has been applied the problem of the sound transmission loss of glass plates (windows). This strategy is in two steps : first a volume velocity sensor is developed as to give a measure representative of the radiated sound at low frequencies. Two sensors have been developed (one discrete and one distributed) and experimentally tested. Next, an optimisation strategy is proposed which allow to locate on the plate a set of several actuators driven in parallel. The goal of this optimisation task is to obtain an open-loop frequency response which behave like a collocated system. The control strategy is finally evaluated on two plate structures.

The aim of this study is to create CFD tools and models capable of simulating pulsatile blood flow in abdominal aortic aneurysm (AAA) and stent graft. It helps to increase the current physiological understanding of rupture risk of AAA and stent graft fixation or migration. Firstly, in order to build a general solver for the AAA modeling with reasonable accuracy, a third/fourth order modified OCI scheme is originally developed for general numerical simulation. The modified OCI scheme has a wider cell Reynolds number limitation. This high order scheme performs well with general rectangular mesh for incompressible fluid. Second, a velocity based finite volume method is originally developed to calculate the stress field for solid in order to capture the transient changes of the blood vessel since the artery is a rubber like material. All one, two and three dimensional classical cases for solid are tested and good results are obtained. The velocity based finite volume method show good potential to calculate the stress field for solid and easy to blend with the finite volume fluid solver. It has been recognized that fluid structure interaction (FSI) is very crucial in biomechanics. In this regard, the velocity based finite volume method is then further developed for FSI application. A well known one dimensional piston problem is studied to understand the feasibility of the fluid structure coupling. The numerical prediction matches the analytical solution very well. The velocity based method introduces less numerical damping compared with a stagger method and a monolithic method. Finally, the work focuses on practical pulsatile boundary conditions, non-Newtonian blood viscous properties and bifurcating geometry, and provides an overview of the hemodynamic within the AAA model. A modified Womersley inlet and imbalance pressure outlet boundary conditions are originally used in this study. The Womersley inlet boundary represents better approximation for pulsatile flow compared with the parabolic inlet condition. Numerical results are presented providing comparison between different boundary conditions using different viscous models in both 2D and 3D aneurysms. Good agreement between the numerical predictions and the experimental data is achieved for 2D case. 3D stent models with different bifurcation angles are also tested. The Womersley inlet boundary condition improves the existing inlet conditions significantly and it can reduce the Aneurysm neck computation domain. The influence of the non-Newtonian model to the wall shear stress (WSS) and strain-rate is also studied. The non-Newtonian model tends to produce higher WSS at both proximal and distal end of the aneurysm as compared with the Newtonian model (both 2D and 3D cases). The computed strain-rate distribution at the centre of the aneurysm is different between these two models. The influence of imbalance outlet pressure at the iliac arteries to the blood flow is originally investigated. The imbalance outlet pressure boundary conditions affect the computed wall shear stress significantly near the bifurcation point. All the pulsatile Womersley inlet, non-Newtonian viscosity properties and the imbalance pressure outlet need to be considered in blood flow simulation of AAA.

Recently, 3D printing has been increasingly used to create physical models of the vocal tract with geometries obtained from magnetic resonance imaging. These printed models allow measuring the vocal tract transfer function, which is not reliably possible in vivo for the vocal tract of living humans. The transfer functions enable the detailed examination of the acoustic effects of specific articulatory strategies in speaking and singing, and the validation of acoustic plane-wave models for realistic vocal tract geometries in articulatory speech synthesis. To measure the acoustic transfer function of 3D-printed models, two techniques have been described: (1) excitation of the models with a broadband sound source at the glottis and measurement of the sound pressure radiated from the lips, and (2) excitation of the models with an external source in front of the lips and measurement of the sound pressure inside the models at the glottal end. The former method is more frequently used and more intuitive due to its similarity to speech production. However, the latter method avoids the intricate problem of constructing a suitable broadband glottal source and is therefore more effective. It has been shown to yield a transfer function similar, but not exactly equal to the volume velocity transfer function between the glottis and the lips, which is usually used to characterize vocal tract acoustics. Here, we revisit this method and show both, theoretically and experimentally, how it can be extended to yield the precise volume velocity transfer function of the vocal tract.

L'objectif de cette thèse consiste en l'étude d'une stratégie de contrôle actif à faible autorité avec comme application le contrôle actif du rayonnement acoustique d'une plaque. Depuis l'essor du contrôle actif, son application aux problèmes acoustiques et vibracoustiques a été investiguée par de nombreux chercheurs, exploitant soit la théorie du contrôle optimal, soit des approches originales basées plus particulièrement sur la physique. Des notions spécifiques au contrôle vibroacoustique ont été développées comme, par exemple, les modes radiatifs pouvant caractériser le rayonnement acoustique d'une plaque d'une manière adaptée au contrôle. <p>Le contrôle actif à faible autorité, pour lequel le Laboratoire de Structures Actives a développé une expertise dans le domaine de l'amortissement et du contrôle actif des vibrations, est une solution attractive par sa simplicité de mise en oeuvre. Le plus souvent implémenté sous la forme d'un contrôle décentralisé constitué de boucles indépendantes, le contrôle à faible autorité bénéficie de certaines garanties de stabilité et de robustesse. <p>Bien que notre stratégie de contrôle puisse s'appliquer à n'importe quel type de plaque, l'application considérée dans ce travail a été motivée par le contexte socio-économique actuel en rapport avec les nuisances acoustiques. Il était en effet intéressant d'évaluer la stratégie de contrôle pour le problème de la transmission acoustique d'un vitrage. La stratégie de contrôle se divise en deux étapes. Tout d'abord le développement d'un capteur unique destiné à fournir une mesure représentative du bruit rayonné par une plaque en basse fréquence. Deux capteurs de vitesse volumétrique (l'un discret, l'autre distribué) ont ainsi été développés et évalués expérimentalement. <p>Ensuite, une procédure d'optimisation de l'emplacement d'un ensemble d'actionneurs pilotés en parallèle est proposée. L'objectif de cette phase d'optimisation est de forcer la réponse fréquentielle du système à posséder les propriétés d'un système colocalisé. La stratégie de contrôle est ensuite évaluée sur deux structures expérimentales. <p><p>/ This thesis is concerned with a low authority active control strategy applied to the sound radiation control of a baffled plate. Since the development of active control ,numerous researchers have studied its application to acoustical or vibroacoustical problems using either the modern control theory or other methods based rather on the understanding of the physics of the problem. Vibroacoustical active control has lead to the definition of radiation modes allowing to describe the radiated sound of a plate in an appropriate manner for active control purposes. <p>Low autorithy control (LAC), for which the Active Structures Laboratory has gained an expertise for active vibration control applications is an interesting solution for its implementation simplicity. Most of the time it consists of several decentralized control loops, and benefits from guaranteed stability and robustness properties. Although our control strategy can be applied to any kind of plates, the application considered here has been motivated by the present socio-economical context related to noise annoyances. The active control strategy has been applied the problem of the sound transmission loss of glass plates (windows). This strategy is in two steps :first a volume velocity sensor is developed as to give a measure representative of the radiated sound at low frequencies. <p>Two sensors have been developed (one discrete and one distributed) and experimentally tested. Next, an optimisation strategy is proposed which allow to locate on the plate a set of several actuators driven in parallel. The goal of this optimisation task is to obtain an open-loop frequency response which behave like a collocated system. The control strategy is finally evaluated on two plate structures. / Doctorat en sciences appliquées / info:eu-repo/semantics/nonPublished

Recently, 3D printing has been increasingly used to create physical models of the vocal tract with geometries obtained from magnetic resonance imaging. These printed models allow measuring the vocal tract transfer function, which is not reliably possible in vivo for the vocal tract of living humans. The transfer functions enable the detailed examination of the acoustic effects of specific articulatory strategies in speaking and singing, and the validation of acoustic plane-wave models for realistic vocal tract geometries in articulatory speech synthesis. To measure the acoustic transfer function of 3D-printed models, two techniques have been described: (1) excitation of the models with a broadband sound source at the glottis and measurement of the sound pressure radiated from the lips, and (2) excitation of the models with an external source in front of the lips and measurement of the sound pressure inside the models at the glottal end. The former method is more frequently used and more intuitive due to its similarity to speech production. However, the latter method avoids the intricate problem of constructing a suitable broadband glottal source and is therefore more effective. It has been shown to yield a transfer function similar, but not exactly equal to the volume velocity transfer function between the glottis and the lips, which is usually used to characterize vocal tract acoustics. Here, we revisit this method and show both, theoretically and experimentally, how it can be extended to yield the precise volume velocity transfer function of the vocal tract.