Rozprawy doktorskie na temat „Homogénéisation continue ou discrète”
Utwórz poprawne odniesienie w stylach APA, MLA, Chicago, Harvard i wielu innych
Sprawdź 30 najlepszych rozpraw doktorskich naukowych na temat „Homogénéisation continue ou discrète”.
Przycisk „Dodaj do bibliografii” jest dostępny obok każdej pracy w bibliografii. Użyj go – a my automatycznie utworzymy odniesienie bibliograficzne do wybranej pracy w stylu cytowania, którego potrzebujesz: APA, MLA, Harvard, Chicago, Vancouver itp.
Możesz również pobrać pełny tekst publikacji naukowej w formacie „.pdf” i przeczytać adnotację do pracy online, jeśli odpowiednie parametry są dostępne w metadanych.
Przeglądaj rozprawy doktorskie z różnych dziedzin i twórz odpowiednie bibliografie.
Verna, Patrice. "Modélisation continue des structures discrètes par homogénéisation : cas des treillis". Grenoble INPG, 1991. http://www.theses.fr/1991INPG0012.
Pełny tekst źródłaAlavi, Seyed Ehsan. "Homogénéisation de milieux architecturés périodiques et quasi-périodiques vers des milieux continus généralisés". Electronic Thesis or Diss., Université de Lorraine, 2021. http://www.theses.fr/2021LORR0305.
Pełny tekst źródłaThis thesis aims to revisit higher-order homogenization schemes towards higher-order or higher gradient continua, successively for periodic and quasi-periodic architected materials and composites, based on variational principles and an extension of Hill macrohomogeneity condition. Continuous homogenization methods are exposed in Part I for micropolar and micromorphic media, followed by an exposition of the alternative discrete homogenization method.We have extended these theoretical developments to the situation of quasi-periodic materials, which still have a regular microstructure. The common idea to the proposed periodic homogenization methods of continuous or discrete nature is to split the microscopic displacement into a homogeneous part representative of the kinematics of the adopted effective continuum and a fluctuation evaluated from a variational principle. In substance, the theoretical developments allow the elaboration of enriched continua (generalized continua) of micromorphic type and all sub continua obtained using suitable degeneration conditions. Numerical applications have been made for architected materials and inclusion-based composites prone to higher-order effects due to their inner architecture. On the theoretical framework, the performed developments remedy many existing limitations of existing higher-order homogenization schemes.In Part II, repetitive lattice materials' effective classical and higher-order mechanical properties have been evaluated based on discrete homogenization schemes. Following the idea of a phenomenological approach, consistent couple stress models of repetitive beam lattices have been elaborated. Enriched Cosserat media have been derived in the spirit of micromechanics, adopting Timoshenko beam models at a microlevel, and applying a continualization method towards a Cosserat effective substitution medium. The proposed continualization method proves to be accurate and computationally efficient compared to continuous homogenization schemes and fully resolved finite element simulations. One key outcome of the performed analyses is the quantification of edge effects in the response of lattice structures, relying on the surface formulation of the extended Hill macrohomogeneity condition.The theoretical background underlying quasi-periodic asymptotic homogenization in the framework of linearized anisotropic elasticity deserves the development of Part III. Different methodologies for evaluating the effective quasi-periodic properties have been elaborated, leading to the emergence of strain gradient effective media. Conformal transformations define a specific class of geometrical mappings, allowing for designing compatible architected materials with inner porosity gradient, making them suitable bone biomechanics candidates
Chesnais, Céline. "Dynamique de milieux réticulés non contreventés : application aux bâtiments". Phd thesis, Ecole Centrale de Lyon, 2010. http://tel.archives-ouvertes.fr/tel-00564489.
Pełny tekst źródłaFennan, Abdelhadi. "Homogénéisation élastoplastique discrète". Ecully, Ecole centrale de Lyon, 1995. http://www.theses.fr/1995ECDL0016.
Pełny tekst źródłaThe aim of this work is to understand and modelling the behaviour of heterogenous elastoplastic materials or more generally elasto-dissipative. To achieve his goal, a discret homogenization method is developped. Three main independent steps have to be implemented : 1- residual homogenization, a purelly elastic step, which allows the evaluation of localization operators. 2- formulation of a macroscopic model from the local behaviour. 3- the study of the macroscopic behaviour, independently of the elementary microscopic structure. This method can be used in many local rheological cases and leads to approximate, but complete and explicit, behaviour law. The macroscopic phenomenological model obtained is a standard generalized model with multiple kinematic hardening showing a discrete spectrum of internal variables. This model takes account the induced anisotropy resulting from microscopic heterogenity. Dealing with the applications and the link with micro-structure, two examples are treated : the cellular structures of dislocations developping in metals by cyclic loadings ; the metalic matrix composites with particulate reinforcement
Delorme, Matthieu. "Ecoulements en milieux fracturés : vers une intégration des approches discrètes et continues". Thesis, Toulouse, INPT, 2015. http://www.theses.fr/2015INPT0038/document.
Pełny tekst źródłaFluid flow simulation is used to optimize oil and gas production. Naturally or hydraulically fractured reservoirs hold a significant part of reserves, difficult to assess. Fractures may create preferential flow paths heavily impacting fluid flow. Accurate modeling of fractured media accounting for strong contrasts would allow operators to optimize resources exploitation while better controlling environmental risks. Integrating sparse available data, we aim at predicting fluid flow processes occurring in the earth’s subsurface accounting for multi-scale fractures with a simply parameterized model. Improving the computational time and results reliability, we propose a full integrated strategy suitable for fractured reservoir specificities by simulating the fractures complexity on large scales. The techniques developed in this thesis, whose interest is demonstrated in an unconventional field case study, can find other applications in geothermal engineering and water resources management
Goda, Ibrahim. "Micromechanical models of network materials presenting internal length scales : applications to trabecular bone under stable and evolutive conditions". Thesis, Université de Lorraine, 2015. http://www.theses.fr/2015LORR0055/document.
Pełny tekst źródłaA methodology based on micromechanics has been developed to determine the effective behavior of network materials endowed with a discrete architecture at the microscopic level. It relies on the discrete homogenization method, which has been applied to textile monolayers and trabecular bones. The initially discrete topology of the considered network materials results after homogenization at the mesoscopic level in anisotropic micropolar effective continuum, which proves able to capture the observed internal scale effects. Such micromechanical methods are useful to remedy the difficulty to measure the effective mechanical properties at the intermediate mesoscopic level scale. The bending and torsion responses of vertebral trabecular bone beam specimens are formulated in both static and dynamic situations, based on the Cosserat theory. 3D models have been developed for describing the multiaxial yield and brittle fracture behavior of trabecular bone, including the analysis of size-dependent non-classical plastic yield. We have constructed by FE analyses a homogeneous, orthotropic couple-stress continuum model as a substitute of the 3D periodic heterogeneous cellular solid model of vertebral trabecular bone, based on the equivalent strain energy approach. Bone tissues are able to adapt their local density and load bearing capacities as well as their size and shape to mechanical stimuli. We have developed models for combined internal and external bone remodeling in the framework of the thermodynamics of irreversible processes, at both the cellular and macroscopic levels. We lastly combined anisotropic internal remodeling with fatigue continuum damage
Landry, Jean-François. "Planification optimale discrète et continue : un joueur de billard autonome optimisé". Thèse, Université de Sherbrooke, 2012. http://hdl.handle.net/11143/6681.
Pełny tekst źródłaMoreau, Gabriel. "Homogénéisation de structures discrètes en élasticité et en incrémental : applications aux modélisations continues linéaires et non-linéaires de treillis quasi-périodiques". Grenoble INPG, 1996. https://tel.archives-ouvertes.fr/tel-01762234.
Pełny tekst źródłaYahya, Farouk. "Approche couplée discrète - continue adaptative pour la mise en forme des métaux". Electronic Thesis or Diss., Valenciennes, Université Polytechnique Hauts-de-France, 2023. http://www.theses.fr/2023UPHF0042.
Pełny tekst źródłaThe development of forming processes generally involves numerical simulation, particularly using the Finite Element Method, to detect, among other things, potential risks of cracking and their propagation, both in the forged piece and in the tooling.The classical Finite Element Method, as well as specific variants applied to crack modeling, have limitations when it comes to simulating multi-cracking problems due to material, geometric, and contact nonlinearities. The very nature of a crack – a discontinuity – is in opposition to the framework of Continuum Mechanics. On the other hand, the Discrete Element Method distinguishes itself by its ability to effectively handle discontinuities. It is used to model granular or brittle cohesive materials. However, this approach also has limitations, especially regarding the consideration of material nonlinearities and computation times, which may require significant computing resources to simulate complex problems.The objective of this thesis work was to develop a dynamic remeshing method that allowed transitioning between the two methods to leverage the advantages of both approaches. This objective was be achieved in three steps.Firstly, a non-overlapping coupling method based on the Lagrange multiplier method was developed. This method aimed at ensuring the compatibility of velocities between the discrete elements and finite elements subdomains to facilitate the communication of physical quantities between the two subdomains.Secondly, it was necessary to ensure the continuity of physical quantities within the same zone during its remeshing. This step was achieved by using polynomial interpolation of displacements. This approach allowed determining the fields within the discrete elements when the transition between a finite element subdomain and a discrete element subdomain took place. Test cases were set up to validate these two approaches.Finally, a method was developed to automate the management of coupling and field transfer operations. This approach involved the use of a remeshing technique that automatically generated discrete subdomains from the geometry of the entire finite elements to be replaced.The dynamic remeshing method implemented in this thesis work was applied to a Kalthoff test case and validated by comparing the crack propagation angle with experiments from the literature
Herr, Nathalie. "Contribution à l'ordonnancement post-pronostic de plateformes hétérogènes et distribuées : approches discrète et continue". Thesis, Besançon, 2015. http://www.theses.fr/2015BESA2049/document.
Pełny tekst źródłaThis thesis addresses the problem of maximizing the production horizon of a heterogeneous distributed platform composed of parallel machines and which has to provide a global production service. Each machine is supposed to be able to provide several throughputs corresponding to different operating conditions. It is assumed that using a machine with degraded performances compared to nominal ones allows to extend its useful life before maintenance. The study falls within the decisional step of PHM (Prognostics and Health Management), in which a prognostics phase allows to determine remaining useful lives of machines. The optimization problem consists in determining the set of machines to use at each time and a running profile for each of them so as to maximize the production horizon before maintenance. Machines running profiles are defined on the basis of two models. First one depicts the behavior of machines used with a discrete number of performances. For this case, the problem complexity is first studied considering many variants of the optimization problem. Several optimal and sub-optimal resolution methods are proposed to deal with the scheduling problem. Several sub-optimal resolution methods are then proposed for the second model, which applies to machines whose throughput rate can vary continuously between two bounds. These research works allow to determine the time before failure of a system on the basis of its components remaining useful lives
Floquet, Pascal. "Procédures discrète et continue d'optimisation et de CAO en génie chimique : étude de cas". Toulouse, INPT, 1986. http://www.theses.fr/1986INPT018G.
Pełny tekst źródłaRahmoun, Jamila. "Modélisation du comportement des matériaux granulaires par des approches discrètes et continues". Phd thesis, Université des Sciences et Technologie de Lille - Lille I, 2006. http://tel.archives-ouvertes.fr/tel-00348876.
Pełny tekst źródłaMadec, Yoann. "Modélisation markovienne discrète et continue d'un processus autocorrélé : application au suivi de patients VIH positifs". Paris 5, 2001. http://www.theses.fr/2001PA055005.
Pełny tekst źródłaLachihab, Adel. "Un modèle numérique pour les composites biphasés matrice - inclusions rigides : Application à la détermination des propriétés élastiques et en fatigue des enrobés bitumeux". Marne-la-vallée, ENPC, 2004. http://www.theses.fr/2004ENPC0025.
Pełny tekst źródłaBauer, Daniela. "Modélisation mécanique par approche continue et discrète des variations du flux sanguin dans la peau et validation expérimentale". Phd thesis, Université de Technologie de Compiègne, 2004. http://tel.archives-ouvertes.fr/tel-00007943.
Pełny tekst źródłaSiska, Filip. "Simulation numérique du comportement mécanique des films minces métalliques par la théorie continue et la dynamique discrète des dislocations". Phd thesis, École Nationale Supérieure des Mines de Paris, 2007. http://tel.archives-ouvertes.fr/tel-00204422.
Pełny tekst źródłaLachihab, Adel. "Un modèle numérique pour les composites biphasés matrice-inclusions rigides: Application à la détermination des propriétés élastiques et en fatigue des enrobés bitumineux". Phd thesis, Ecole des Ponts ParisTech, 2004. http://pastel.archives-ouvertes.fr/pastel-00001036.
Pełny tekst źródłaElNady, Khaled. "Modèles de comportement non linéaire des matériaux architecturés par des méthodes d'homogénéisation discrètes en grandes déformations. Application à des biomembranes et des textiles". Thesis, Université de Lorraine, 2015. http://www.theses.fr/2015LORR0032/document.
Pełny tekst źródłaThe present thesis deals with the development of micromechanical schemes for the computation of the homogenized response of architectured materials, focusing on periodical lattice materials. Architectured and micro-architectured materials cover a wide range of mechanical properties according to the nodal connectivity, geometrical arrangement of the structural elements, their moduli, and a possible structural hierarchy. The principal objective of the thesis is the consideration of geometrical nonlinearities accounting for the large changes of the initial lattice geometry, due to the small bending stiffness of the structural elements, in comparison to their tensile rigidity. The so-called discrete homogenization method is extended to the geometrically nonlinear setting for periodical lattices; incremental schemes are constructed based on a staggered localization-homogenization computation of the lattice response over a repetitive unit cell submitted to a controlled deformation loading. The obtained effective medium is a micropolar anisotropic continuum, the effective properties of which accounting for the geometrical arrangement of the structural elements within the lattice and their mechanical properties. The non affine response of the lattice leads to possible size effects which can be captured by an enrichment of the classical Cauchy continuum either by adding rotational degrees of freedom as for the micropolar effective continuum, or by considering second order gradients of the displacement field. Both strategies are followed in this work, the construction of second order grade continua by discrete homogenization being done in a small perturbations framework. We show that both strategies for the enrichment of the effective continuum are complementary due to the existing analogy in the construction of the micropolar and second order grade continua by homogenization. The combination of both schemes further delivers tension, bending and torsion internal lengths, which reflect the lattice topology and the mechanical properties of its structural elements. Applications to textiles and biological membranes described as quasi periodical networks of filaments are considered. The computed effective response is validated by comparison with FE simulations performed over a representative unit cell of the lattice. The homogenization schemes have been implemented in a dedicated code written in combined symbolic and numerical language, and using as an input the lattice geometry and microstructural mechanical properties. The developed predictive micromechanical schemes offer a design tool to conceive new architectured materials to expand the boundaries of the 'material-property' space
Mnara, Mondher. "Estimation discrète/continue du choix du lieu de pratique et de la durée de séjour des nouveaux médecins généralistes au Québec". Thesis, National Library of Canada = Bibliothèque nationale du Canada, 1997. http://www.collectionscanada.ca/obj/s4/f2/dsk3/ftp05/mq25341.pdf.
Pełny tekst źródłaGloria, Antoine. "Modélisation et méthodes numériques multi-échelles en élasticité non linéaire". Marne-la-vallée, ENPC, 2007. http://www.theses.fr/2007ENPC0710.
Pełny tekst źródłaEl-Bouanani, Hicham. "Vorticité dans des systèmes de spins à symétrie continue". Phd thesis, Université du Sud Toulon Var, 2008. http://tel.archives-ouvertes.fr/tel-00381684.
Pełny tekst źródłaDans le Chapître 1, on rappelle quelques propriétés de l'interaction entre plus proches voisins, pour le rotateur, ou sa version simplifiée appelée modèle de Villain. On introduit aussi le modèle du champ moyen.
Le modèle de Kac, qui partage certains aspects de ces deux modèles, est étudié au Chapître 2. Par un procédé d'homogénéisation, on ramène essentiellement l'étude de la mesure de Gibbs en volume fini à celle de la fonctionnelle énergie libre $\cal F$, généralisant des techniques utilisées dans le modèle d'Ising.
Les propriétés de vorticité du modèle de Kac sont analysées dans le Chapître 3, où l'on détermine les extrema de $\cal F$, avec conditions limite. On met ainsi en évidence des configurations très similaires à celles des solutions des équations de Ginzburg-Landau.
Dans le Chapître 4 on passe au cas quantique, en introduisant la notion de "matrice de vorticité" à température inverse $\beta$, dont on calcule le "degré non-commutatif". Il apparaît ainsi, pour le modèle XY de spin 1/2, des configurations de vorticité analogues à celles rencontrées dans le cas classique.
Gloria, Antoine. "Modélisation et méthodes numériques multiéchelles en élasticité non linéaire". Phd thesis, Ecole des Ponts ParisTech, 2007. http://tel.archives-ouvertes.fr/tel-00166171.
Pełny tekst źródłapour l'homogénéisation de fonctionnelles intégrales utilisées en élasticité non linéaire. Ces mé-
thodes couplent, au niveau mésoscopique, un matériau hyperélastique hétérogène ou un réseau de
liens en interaction, avec, au niveau macroscopique, un modèle d'élasticité non linéaire. La loi de
constitution macroscopique est obtenue par la résolution de problèmes mésoscopiques, continus ou
discrets. Aux chapitres 1, 2 et 3 on introduit les modèles mécaniques et les outils mathématiques et
numériques utilisés par la suite. Aux chapitres 5, 6 et 7, on présente une méthode directe de réso-
lution numérique du comportement homogénéisé d'un matériau composite périodique en grandes
déformations et un cadre général pour l'analyse des méthodes d'homogénéisation numérique. On
démontre notamment la convergence de méthodes numériques classiques sous des hypothèses gé-
nérales ainsi qu'un résultat de correcteur numérique. On étend enfin les résultats au couplage avec
des méthodes de sur-échantillonnage. Aux chapitres 8, 9 et 10, nous considérons une modélisation
mésoscopique par un système discret. Nous étudions d'abord un problème de G-fermeture pour un
réseau de résistances. Au chapitre suivant nous démontrons un résultat de représentation intégrale
pour l'énergie d'un système de spins en interaction. Enfin, nous dérivons un modèle hyperélastique
continu à partir d'un réseau stochastique de points en interaction, et l'appliquons pour démontrer
la convergence de modèles discrets développés en mécanique. Dans une dernière partie, chapitre 11,
nous présentons une nouvelle méthode numérique pour résoudre des problèmes d'interaction fluide
structure, où la structure est décrite par une coque tridimensionnelle.
Smerlak, Matteo. "Divergence des mousses de spins : Comptage de puissances et resommation dans le modèle plat". Thesis, Aix-Marseille 2, 2011. http://www.theses.fr/2011AIX22115/document.
Pełny tekst źródłaIn this thesis we study the flat model, the main buidling block for the spinfoam ap- proach to quantum gravity, with an emphasis on its divergences. Besides a personal introduction to the problem of quantum gravity, the manuscript consists in two part. In the first one, we establish an exact powercounting formula for the bubble divergences of the flat model, using tools from discrete gauge theory and twisted cohomology. In the second one, we address the issue of spinfoam continuum limit, both from the lattice field theory and the group field theory perspectives. In particular, we put forward a new proof of the Borel summability of the Boulatov-Freidel-Louapre model, with an improved control over the large-spin scaling behaviour. We conclude with an outlook of the renormalization program in spinfoam quantum gravity
Mourad, Ayman. "Description topologique de l'architecture fibreuse et modélisation mécanique du myocarde". Grenoble INPG, 2003. https://tel.archives-ouvertes.fr/tel-00004003.
Pełny tekst źródłaFaye, Alexandrine. "Troubles d'utilisation d'outils et de la cognition numérique après lésions vasculaires cérébrales : deux faces d'une même pièce ?" Thesis, Lyon, 2018. http://www.theses.fr/2018LYSE2135/document.
Pełny tekst źródłaTool use is a defining feature of the genus Homo. It is therefore fundamental to better understand the cognitive and cerebral bases that allow us to use tools. The current cognitivist models explain tool use through the hypothesis of an activation of gestural memories (i.e., gestural or visuo-kinetic engrams, or sensorimotor knowledge of manipulation; see Rothi, Ochipa, & Heilman, 1991; Buxbaum, 2001). This theory is unable to explain the use of novel tools. An alternative hypothesis suggests that any situation of tool use (familiar and new) requires technical reasoning (e.g., Osiurak & Badets, 2016). This reasoning, involving the left inferior parietal lobe, would enable to formulate the mechanical action and to evaluate the physical properties of tools and objects. One of the aims of this thesis was to better understand the tool use disorders in brain-damaged patients, within the framework of the technical reasoning hypothesis. This work has also focused on the investigation of numerical cognition. By this term we refer to mental arithmetic and math, but also to analogical code (see the Triple Code Model, Dehaene & Cohen, 1995). It corresponds to the representation of numerical quantities, stored in the parietal lobes. In other words, this code would contain the sense of number (Dehaene, 1997) to associate a symbolic label (e.g., Arabic digits) with the corresponding quantity. In everyday life, this representation would be critical to compare or estimate the numerosity of object sets.The main objective of this thesis was to explore, at cognitive and cerebral levels, whether links exist between both fields of interest that are tool use and numerical cognition. Indeed, we noticed that both capacities need a common process of magnitude estimation (i.e., physical properties and numerical quantity). In addition, at the cerebral level, they require the activation of common regions in the parietal lobe. We relied on the Theory Of Magnitude (ATOM) formulated by Walsh (2003). It postulates that all magnitudes, namely the dimensions described by “more than/less than” relationships (e.g., Is this stick long enough to reach a given place?), are processed within a common and unique system, in the right parietal lobe (Bueti & Walsh, 2009). We assumed that the magnitude of physical properties could be processed in this system as well as the discrete (e.g., numbers) and continuous (e.g., time, space) magnitudes. Our results highlighted a disorder of novel tool-use in LBD patients, who nevertheless had no difficulty in estimating physical properties. The RBD patients were impaired in all conditions assessing the numerical cognition, refuting the predictions derived from TCM. They were also impaired in the estimation of the length but not of the weight. As associations between estimation of length and of weight, and between estimation of length and numerical cognition have been observed in the different groups, we suggest that the magnitude system be divided into subsystems. Surprisingly, we found an association between tool use and approximate calculation in LBD patients assuming an attempt to compensate tool use by calculation. Finally, it seems that tool use and numerical cognition rely on distinct neurocognitive mechanisms since the different types of magnitudes might not be processed within a common and unique system of magnitude
Gazzo, Salvatore. "Characterisation of the mechanical behaviour of networks and woven fabrics with a discrete homogenization model". Thesis, Lyon, 2019. http://www.theses.fr/2019LYSET006/document.
Pełny tekst źródłaIn the past decades there has been an impressive progress in the development of new materials for mechanical related applications. New generations of composites have been developed, that can offer advantages over the unidirectional fibre-reinforced mats commonly used then materials take the name of woven fabrics. The behaviour of this material is strongly influenced by the micro-structure of the material. In the thesis mechanical models and a numerical scheme able to model the mechanical behaviour of woven fabrics and general network materials have been developed. The model takes in to account the micro-structure by means of a homogenization technique. The fibres in the network have been treated like microbeams, having both extensional and bending stiffness, with different types of connection, according to the pattern and detail of the network. The developed procedure was applied for obtaining the homogenized mechanical models for some types of biaxial and quadriaxial networks of fibres, simulating either fibre nets (in this case rigid connection were assumed among the fibres) or tissues with negligible interaction between the fibre bundles, and with relative sliding prevented (in this case the connections were simulated by means of pivots). Different geometries were analysed, including the cases in which the fibres are not orthogonal. A first gradient medium is usually obtained but, in some cases, the homogenization procedure itself indicates that a higher order continuum is better fit to represent the deformation of the micro-structure. Special results were obtained for the case of fibres connected by pivots. In this cases an orthotropic material with zero shear modulus was obtained. Such a material has a not elliptic constitutive tensor, thus it can lead to strain concentrations. However, it was shown that some considerations about the physical behaviour of such networks indicated that higher order terms had to be included in the expansion of the internal forces and deformations, so that a strain gradient material was obtained. The results obtained can be used for the design of specific materials requiring ad-hoc properties. Although the reference model is a network material, the results obtained can be applied to other similar kinds of microstructures, like pantographic materials, micro devices composed by microbeams etc. They have been limited at the range of linear elasticity, that is small deformation and linear elastic behaviour. Then, numerical simulations were focused on extension tests and bias tests. The obtained deformed configurations are consistent with the literature experimental tests, both for balanced and unbalanced tissues. Moreover, a comparison between first and second gradient numerical predictions was performed. It was observed that second gradient predictions better simulate the experimental evidences
Obin, Nicolas. "MeLos : analysis and modelling of speech prosody and speaking style". Phd thesis, Université Pierre et Marie Curie - Paris VI, 2011. http://tel.archives-ouvertes.fr/tel-00694687.
Pełny tekst źródłaObin, Nicolas. "MeLos : analysis and modelling of speech prosody and speaking style". Phd thesis, Paris 6, 2011. http://www.theses.fr/2011PA066172.
Pełny tekst źródłaCai, Jiatu. "Méthodes asymptotiques en contrôle stochastique et applications à la finance". Sorbonne Paris Cité, 2016. http://www.theses.fr/2016USPCC338.
Pełny tekst źródłaIn this thesis, we study several mathematical finance problems related to the presence of market imperfections. Our main approach for solving them is to establish a relevant asymptotic framework in which explicit approximate solutions can be obtained for the associated control problems. In the first part of this thesis, we are interested in the pricing and hedging of European options. We first consider the question of determining the optimal rebalancing dates for a replicating portfolio in the presence of a drift in the underlying dynamics. We show that in this situation, it is possible to generate positive returns while hedging the option and describe a rebalancing strategy which is asymptotically optimal for a mean-variance type criterion. Then we propose an asymptotic framework for options risk management under proportional transaction costs. Inspired by Leland’s approach, we develop an alternative way to build hedging portfolios enabling us to minimize hedging errors. The second part of this manuscript is devoted to the issue of tracking a stochastic target. The agent aims at staying close to the target while minimizing tracking efforts. In a small costs asymptotics, we establish a lower bound for the value function associated to this optimization problem. This bound is interpreted in term of ergodic control of Brownian motion. We also provide numerous examples for which the lower bound is explicit and attained by a strategy that we describe. In the last part of this thesis, we focus on the problem of consumption-investment with capital gains taxes. We first obtain an asymptotic expansion for the associated value function that we interpret in a probabilistic way. Then, in the case of a market with regime-switching and for an investor with recursive utility of Epstein-Zin type, we solve the problem explicitly by providing a closed-form consumption-investment strategy. Finally, we study the joint impact of transaction costs and capital gains taxes. We provide a system of corrector equations which enables us to unify the results in [ST13] and [CD13]
Debry, Edouard. "Modélisation et simulation numérique de la dynamique des aérosols atmosphériques". Phd thesis, Ecole des Ponts ParisTech, 2004. http://pastel.archives-ouvertes.fr/pastel-00001030.
Pełny tekst źródła