Cette thèse porte sur l’étude d’un problème inverse de sources pour un modèle bidimensionnel d’intrusion saline. Dans un premier temps, nous nous intéressons à la modélisation du phénomène d’intrusion saline dans un aquifère côtier non confiné. En tenant compte des hypothèses particulières, nous obtenons dans le cas stationnaire une équation elliptique de la charge hydraulique dont le second membre est constitué des sources ponctuelles. L’étude du problème direct consiste à analyser le modèle dérivé et à établir un résultat d’existence et d’unicité d’une solution. Ensuite, dans la partie problème inverse, il s’agit de l’identification de termes sources à partir des mesures locales. Nous traitons les trois questions relatives à ce problème inverse, l’identifiabilité, l’identification et la stabilité. Concernant l’identification, nous formulons le problème inverse comme un problème de contrôle avec une fonctionnelle coût qui calcule l’écart quadratique entre les mesures expérimentales et celles obtenues par la résolution du problème direct. L’optimisation de cette fonction nécessite le calcul de son gradient que nous obtenons par la méthode de sensibilités et par la méthode de l’état adjoint. Quant à la stabilité, nous établissons deux types d’estimations, logarithmiques et lipschitziennes, pour les positions et les intensités de sources dans le cas de l’équation elliptique obtenue et en considérant toujours des mesures intérieures. De plus, nous avons généralisé les résultats des estimations lipschitziennes pour l’équation elliptique de la forme –Δu+k2u=F. La dernière partie de la thèse est destinée à montrer les résultats de l’identification numérique en fonction des paramètres intervenant dans le modèle principal. / This thesis deals with the study of an inverse source problem for a two dimensional seawater intrusion model. First, we focus on the modeling of the seawater intrusion phenomenon in a costal unconfined aquifer. Then considering some specific assumptions, we obtain, in the steady state, an elliptic equation of the hydraulic head with a left hand side formed by point wise sources. The study of the direct problem aims to analyze the derived model and to establish a result of existence and uniqueness of solution. The inverse problem concerns the identification of sources from local measurements. We are interested in the study of uniqueness, identification and stability.Concerning the identification, we transform the inverse problem to a control problem with a cost functional computing the quadratic error between the experimental measures and those obtained by solving the direct problem. To optimize this function, we need to compute its gradient and this can be done by the sensibility and the adjoint methods. Moreover, regarding the stability, we establish two types of estimates, logarithmic and lipschitz, for sources positions and intensities in the case of the elliptic equation assuming interior observations. Furthermore, we have generalized the results of Lipschitz estimates for the elliptic equation –Δu+k2u=F. The last part of the thesis is intended to show the results of the numerical identification based on parameters involved in the main model.

Cette thèse est consacrée aux quelques problèmes mathématiques concernant les extensions minimales de Lipschitz. Elle est organisée de manière suivante. Le chapitre 1 est dédié à l’introduction des extensions minimales de Lipschitz. Dans le chapitre 2, nous étudions la relation entre la constante de Lipschitz d’ 1-field et la constante de Lipschitz du gradient associée à ce 1-field. Nous proposons deux formules explicites Sup-Inf, qui sont des extensions extrêmes minimales de Lipschitz d’1-field. Nous expliquons comment les utiliser pour construire les extensions minimales de Lipschitz pour les applications Rmà Rn . Par ailleurs, nous montrons que les extensions de Wells d’1- fields sont les extensions absolument minimales de Lipschitz (AMLE) lorsque le domaine d’expansion d’1-field est infini. Un contreexemple est présenté afin de montrer que ce résultat n’est pas vrai en général. Dans le chapitre 3, nous étudions la version discrète de l’existence et l’unicité de l’AMLE. Nous montrons que la fonction tight introduite par Sheffield and Smart est l’extension de Kirszbraun. Dans le cas réel, nous pouvons montrer que cette extension est unique. De plus, nous proposons un algorithme qui permet de calculer efficacement la valeur de l’extension de Kirszbraun en complexité polynomiale. Pour conclure, nous décrivons quelques pistes pour la future recherche, qui sont liées au sujet présenté dans ce manuscrit. / The thesis is concerned to some mathematical problems on minimal Lipschitz extensions. Chapter 1: We introduce some basic background about minimal Lipschitz extension (MLE) problems. Chapter 2: We study the relationship between the Lipschitz constant of 1-field and the Lipschitz constant of the gradient associated with this 1-field. We produce two Sup-Inf explicit formulas which are two extremal minimal Lipschitz extensions for 1-fields. We explain how to use the Sup-Inf explicit minimal Lipschitz extensions for 1-fields to construct minimal Lipschitz extension of mappings from Rm to Rn. Moreover, we show that Wells’s extensions of 1-fields are absolutely minimal Lipschitz extensions (AMLE) when the domain of 1-field to expand is finite. We provide a counter-example showing that this result is false in general. Chapter 3: We study the discrete version of the existence and uniqueness of AMLE. We prove that the tight function introduced by Sheffield and Smart is a Kirszbraun extension. In the realvalued case, we prove that the Kirszbraun extension is unique. Moreover, we produce a simple algorithm which calculates efficiently the value of the Kirszbraun extension in polynomial time. Chapter 4: We describe some problems for future research, which are related to the subject represented in the thesis.

La présente thèse traite de deux questions de stabilité en mécanique des fluides. Les deux premiers résultats de la thèse sont consacrés au problème de la limite non-visqueuse pour les équations de Navier-Stokes. Il s'agit de déterminer si une famille de solutions de Navier-Stokes dans un demi-espace avec une condition de Navier au bord converge vers une solution du modèle non visqueux, l'équation d'Euler, lorsque les paramètres de viscosité tendent vers zéro. Dans un premier temps, on considère le modèle incompressible 2D. Nous obtenons la convergence dans L2 des solutions faibles de Navier-Stokes vers une solution forte d'Euler, et une instabilité dans L∞ en temps très court pour certaines données initiales qui sont des solutions stationnaires de l'équation d'Euler. Ces résultats ne sont pas contradictoires, et on construit un exemple de donnée initiale permettant de voir se réaliser les deux phénomènes simultanément dans le cadre périodique. Dans un second temps, on s'intéresse au modèle compressible isentropique (température constante) en 3D. On démontre l'existence de solutions dans des espaces de Sobolev conormaux sur un temps qui ne dépend pas de la viscosité lorsque celle-ci devient très petite, et on obtient la convergence forte de ces solutions vers une solution de l'équation d'Euler sur ce temps uniforme par des arguments de compacité. Le troisième résultat de cette thèse traite d'un problème de stabilité d'ondes solitaires. Précisément, on considère un fluide isentropique et non visqueux avec capillarité interne, régi par le modèle d'Euler-Korteweg, et on montre l'instabilité transverse non-linéaire de solitons, c'est-à-dire que des perturbations 2D initialement petites d'une solution sous forme d'onde progressive 1D peuvent s'éloigner de manière importante de celle-ci. / This thesis deals with a couple of stability problems in fluid mechanics. In the first two parts, we work on the inviscid limit problem for Navier-Stokes equations. We look to show whether or not a sequence of solutions to Navier-Stokes in a half-space with a Navier slip condition on the boundary converges towards a solution of the inviscid model, the Euler equation, when the viscosity parameters vanish. First, we consider the 2D incompressible model. We obtain convergence in L2 of weak solutions of Navier-Stokes towards a strong solution of Euler, as well as the instability in L∞ in a very short time of some initial data chosen as stationary solutions to the Euler equation. These results are not contradictory, and we construct initial data that allows both phenomena to occur simultaneously in the periodic setting. Second, we look at the 3D isentropic (constant temperature) compressible equations. We show that solutions exist in conormal Sobolev spaces for a time that does not depend on the viscosity when this is small, and we get strong convergence towards a solution of the Euler equation on this uniform time of existence by compactness arguments. In the third part of the thesis, we work on a solitary wave stability problem. To be precise, we consider an isentropic, compressible, inviscid fluid with internal capillarity, governed by the Euler-Korteweg equations, and we show the transverse nonlinear instability of solitons, that is that initially small 2D perturbations of a 1D travelling wave solution can end up far from it.

This thesis consists of four parts. In the first part we recall some background theory that will be used throughout the thesis. In the second part, we studied the absolute continuity of the laws of the solutions of some perturbed stochastic differential equaitons(SDEs) and perturbed reflected SDEs using Malliavin calculus. Because the extra terms in the perturbed SDEs involve the maximum of the solution itself, the Malliavin differentiability of the solutions becomes very delicate. In the third part, we studied the absolute continuity of the laws of the solutions of the parabolic stochastic partial differential equations(SPDEs) with two reflecting walls using Malliavin calculus. Our study is based on Yang and Zhang \cite{YZ1}, in which the existence and uniqueness of the solutions of such SPDEs was established. In the fourth part, we gave the existence and uniqueness of the solutions of the elliptic SPDEs with two reflecting walls and general diffusion coefficients.

Le travail de cette thèse a porté sur le développement de techniques numériques pour l'homogénéisation d'équations dont les coefficients présentent des hétérogénéités aléatoires à petite échelle. Les difficultés liées à la résolution de telles équations aux dérivées partielles peuvent être résolues grâce à la théorie de l'homogénéisation stochastique. On substitue alors la résolution d'une équation dont les coefficients sont aléatoires et oscillants à l'échelle la plus fine du problème par la résolution d'une équation à coefficients constants. Cependant, une difficulté subsiste : le calcul de ces coefficients dits homogénéisés sont définis par une moyenne ergodique, que l'on ne peut atteindre en pratique. Seuls des approximations aléatoires de ces quantités déterministes sont calculables, et l'erreur commise lors de l'approximation est importante. Ces questions sont développées en détail dans le Chapitre 1 qui tient lieu d'introduction. L'objet du Chapitre 2 de cette thèse est de réduire l'erreur de cette approximation dans un cas nonlinéaire, en réduisant la variance de l'estimateur par la méthode des variables antithétiques. Dans le Chapitre 3, on montre comment obtenir une meilleure réduction de variance par la méthode des vari- ables de contrôle. Cette approche repose sur un modèle approché, disponible dans le cas étudié. Elle est plus invasive et moins générique, on l'étudie dans un cas linéaire. Dans le Chapitre 4, à nouveau dans un cas linéaire, on introduit une méthode de sélection pour réduire l'erreur commise. Enfin, le Chapitre 5 porte sur l'analyse d'un problème in- verse, où l'on recherche des paramètres à l'échelle la plus fine, ne connaissant que quelques quantités macroscopiques, par exemple les coefficients homogénéisés du modèle / In this thesis, we design numerical techniques to address the homogenization of equations the coefficients of which exhibit small scale random heterogeneities. Solving such elliptic partial differential equations is prohibitively expensive. One may use stochastic homogenization theory to reduce the complexity of this task. We then substitute the random, fine scale oscillating coefficients of the equation with constant homogenized coefficients. These coefficients are defined through an ergodic average inaccessible to practical computation. Only random approximations thereof are available. The error committed in this approximation is significant. These issues are detailed in the introductory Chapter 1. In Chapter 2, we show how to reduce the error in this approximation, in a nonlinear case, by using an antithetic variable estimator that has a smaller variance than the standard Monte Carlo estimator. In Chapter 3, in a linear case, we show how to obtain an even better variance reduction with the control variate method. Such a method is based on a surrogate model. In Chapter 4, we use a selection method to reduce the global error. Chapter 5 is devoted to the analysis of an inverse problem, wherein we seek parameters at the fine scale whilst only being provided with a handful of macroscopic quantities, among which the homogenized coefficients

Orientadores: Maria Amélia Novais Schleicher, Joerg Dietrich Wilhelm Schleicher / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Matemática Estatística e Computação Científica / Made available in DSpace on 2018-08-24T12:48:22Z (GMT). No. of bitstreams: 1 Coimbra_TiagoAntonioAlves_D.pdf: 11322080 bytes, checksum: 57d63d91892a162c542c0a1d25e3c08b (MD5) Previous issue date: 2014 / Resumo: A operação para continuação de afastamento (Offset Continuation Operation - OCO) transforma um registro sísmico adquirido com um certo afastamento entre fonte e receptor, em um registro correspondente como se fosse adquirido com outro afastamento. O deslocamento de um evento sísmico sob esta operação pode ser descrito por uma equação diferencial parcial de segunda ordem. Baseado na aproximação WKBJ, deduzimos uma equação tipo iconal OCO que descreve os aspectos cinemáticos deste deslocamento em analogia a uma onda acústica, e uma equação de transporte que descreve a alteração das amplitudes. Baseado na teoria dos raios representamos uma forma de solução para a nova equação proposta. Notamos que operadores diferencias de transformação de configuração que corrigem o fator de espalhamento geométrico para qualquer afastamento, ao menos de modo assintótico, são novos na literatura. Baseados na cinemática da operação, propomos um operador de empilhamento multi-paramétrico no domínio não-migrado dos dados sísmicos. Esse empilhamento multi-paramétrico usa uma velocidade média, chamada de velocidade OCO, bem como outros parâmetros cinemáticos do campo de onda importantes. Por se basear na OCO, os tempos de trânsito usados neste empilhamento multi-paramétrico acompanham a trajetória OCO que aproxima à verdadeira trajetória do ponto de reflexão comum. Assim, os parâmetros extraídos servem para melhorar a correção do sobretempo convencional ou realizar correções correspondentes para afastamentos não nulos. Desta forma, é possível aumentar a qualidade das seções empilhadas convencionais de afastamento nulo ou até gerar seções empilhadas de outros afastamentos. Os parâmetros cinemáticos envolvidos ainda podem ser utilizado para construir um melhor modelo de velocidade. Exemplos numéricos mostram que o empilhamento usando trajetórias OCO aumenta, de forma significativa, a qualidade dos dados com uso de menos parâmetros que nos métodos clássicos / Abstract: The Offset Continuation Operation (OCO) transforms a seismic record with a certain offset between source and receiver in another record as if obtained with another offset. The displacement between a seismic event under this operation may be modeled by a second order partial differential equation. We base on the WKBJ approximation and deduce an OCO equation type-eikonal and a transport equation. The former decribes the kinematic features of this displacement, analogously to an acoustic wave, and the latter describes the change of the amplitudes. We present a solution for the proposed new equation, based on the ray theory. The differential configuration transformation operators that correct the geometric spreading for any common offset section (CO) in an asymptoptic way are a novelty in the literature. Based on the kinematics of the operation, we propose a multi-parametric stacking on the unmigrated data domain. This multi-parametric use stacking average velocity called OCO velocity and other kinematic parameters important field from waveform. Since it is based on OCO, travel times used in this multi-parametric stacking accompany OCO trajectory that approximates the true trajectory of the common reflection point (CRP). Thus, the extracted parameters are used to improve the precision of the moveout or to do corresponding corrections for nonzero offsets. Thus, it is possible to increase the quality of conventional sections stacked in zero offset or even generate stacked sections other common offsets. The kinematic parameters involved can also be used to build a velocity model better. Numerical examples show that the stacking using trajectories OCO increases, significantly, the quality of the data using fewer parameters than the classical methods / Doutorado / Matematica Aplicada / Doutor em Matemática Aplicada

Orientador: João Bosco Ribeiro do Val / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Eletrica e de Computação / Made available in DSpace on 2018-08-14T11:15:46Z (GMT). No. of bitstreams: 1 Pinheiro_NathalieCarvalho_M.pdf: 2872807 bytes, checksum: 929137a3cc76534c1832b041f3175ff3 (MD5) Previous issue date: 2009 / Resumo: Os elevados custos de produção na extração de petróleo marítimo são reduzidos com um sistema inovador que a Petrobras está desenvolvendo, denominado VASPS (do inglês, Vertical Annular Separation and Pumping System), capaz de separar gás e líquido ainda no assoalho oceânico. Esta dissertação trata do ajuste do nível de l'iquido no reservatório do VASPS, via controle impulsional estocástico, que se distancia bastante de um regulador convencional. A vazão de entrada flutua e o controle consiste em alterar a velocidade da bomba de saída do tanque, mantendo o nível numa faixa de operação, na qual ele varia livremente sem prejuízo. Todavia, é necessário de um lado observar o risco de operar próximo aos extremos de nível e de outro minimizar o número de intervenções na velocidade da bomba para prolongar sua vida útil. Optou-se por modelar o sistema por meio de um processo de difusão e formular por controle impulsional. Para a sua solução, o controle impulsional é convertido em uma sequência de problemas de parada ótima iterados, resolvidos utilizando-se a Discretização do Valor Médio, MVS (do inglês, Mean Value Scheme). Esta dissertação introduz o uso do controle impulsional nesta aplicação além da técnica citada para resolver problemas de parada ótima. / Abstract: The high costs in offshore oil production are reduced with the use of an innovative system which has been developed by Petrobras, the Brazilian oil company. Called VASPS (Vertical Annular Separation and Pumping System), it consists of an undersea gas/liquid separator. This work presents a strategy for the liquid level adjustment in the VASPS tank, which is subject to uncertain liquid inflow. This is far from a strict control regulator problem, since the liquid level may drifts freely inside an operation range. Although, in one hand, it is necessary to account for a risky operation near the limits, on the other hand, acting freely and continuously in the controlled pump may drastically shorten the lifetime of the equipment. To prevent premature worn with halt in the oil production, the control input variations should be meager. We propose a stochastic impulse control for varying the outflow pump speed. This formulation is transformed in a sequence of iterated optimal stopping problems, which results in a sequence of variational inequalities. We employ the numerical method called Mean Value Scheme (MVS) to solve this type of problem. This monograph introduces impulse control to the resevoir level adjustment of VASPS, together with the application of the MVS to its solution. / Mestrado / Automação / Mestre em Engenharia Elétrica

Orientador: João Frederico da Costa Azevedo Meyer / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Matemática Estatística e Computação Científica / Made available in DSpace on 2018-08-26T22:13:08Z (GMT). No. of bitstreams: 1 Miyaoka_TiagoYuzo_M.pdf: 9483350 bytes, checksum: 13a6ce526d2a0eca797c7b2c56f65600 (MD5) Previous issue date: 2015 / Resumo: Este trabalho trata da modelagem matemática e da simulação computacional de um problema de dinâmica populacional, mais precisamente a interação de um poluente tóxico a duas espécies que competem entre si por espaço e alimento. A modelagem é feita a partir de dispersão e advecção populacional juntamente com o modelo clássico de Lotka-Volterra e reprodução do tipo de Verhulst, mas com um termo inovador para a interação entre poluente e população. Este termo inovador visa a melhoria do modelo a médio e longo prazos, pois tem comportamento assintótico em relação ao tempo. Temos assim um sistema de equações diferenciais parciais não-linear, cuja solução analítica é impossível de ser obtida. Recorremos então a métodos numéricos e simulações computacionais para obter soluções aproximadas. Para isso, utilizamos os métodos de Elementos Finitos (com elementos triangulares de primeira ordem) nas variáveis espaciais e de Diferenças Finitas (mais especificamente, o método de Crank-Nicolson) na temporal, além do método preditor-corretor de Douglas e Dupont para tratar não linearidades, detalhando o procedimento de se obter um software capaz de gerar cenários qualitativamente realistas (os parâmetros utilizados foram estimados). Com o software obtido apresentamos gráficos das soluções aproximadas em cenários hipotéticos distintos, de forma a poder analisar possíveis impactos ambientais causados pela poluição despejada no meio ambiente / Abstract: This work treats the mathematical modeling and computational simulation of a populational dynamics problem, more precisely the interaction of a toxic pollutant in two species which compete with each other for space and food. The modeling is done from populational dispersion and advection together with the classical model of Lotka-Volterra and Verhulst type reproduction, but with a innovative term for the interaction of pollutant and population. This innovative term aims the improvement of the model in the medium and long time, because it has asymptotic behaviour in relation to time. Therefore we have a system of non linear partial differential equations, whose analytical solution is impossible to be obtained. We then appeal to numerical methods and computational simulations to obtain approximated solutions. For this, we use the Finite Elements method (with first order triangular elements) in spatial variables and Finite Differences method (more specifically the Crank-Nicolson method), in addition to the Douglas and Dupont predictor-corrector method to treat non linearities, detailing the process of obtaining a software capable of generating qualitatively realistic scenarios (the parameters used were estimated). With the obtained software we present plots of approximate solutions in different hypothetical scenarios, in order to analyze possible enviromental impacts caused by pollution released into the environment / Mestrado / Matematica Aplicada / Mestre em Matemática Aplicada

Orientador: Igor Leite Freire / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Matemática Estatística e Computação Científica / Made available in DSpace on 2018-08-27T16:11:39Z (GMT). No. of bitstreams: 1 Sampaio_JulioCesarSantos_D.pdf: 2598760 bytes, checksum: b1332349b50fc600ddddb7596fd4b5a4 (MD5) Previous issue date: 2015 / Resumo: Neste trabalho estudamos simetrias de Lie e a teoria de leis de conservação desenvolvida por Ibragimov nos últimos 10 anos. Leis de conservação para várias equações sem Lagrangeanas clássicas foram estabelecidas / Abstract: In this work we study Lie point symmetries and the theory on conservation laws developed by Ibragimov in the last 10 years. Conservation laws for several equations without classical Lagrangians were established / Doutorado / Matematica Aplicada / Doutor em Matemática Aplicada

Ce travail fait partie du projet Marie-Curie ITN STEEP, dans le domaine des faisceaux énergétiques. Nous étudions ici l'identification de paramètres pour un modèle générique d'usinage par jet d'eau abrasif. L'étude de ce problème trouve son origine dans les applications industrielles d'usinage, où la nécessité de modéliser et prédire la surface finale avec une très grande précision est essentielle en l'absence de connaissance des paramètres du modèle Nous proposons ici une méthode d'identification des paramètres du modèle basée sur la minimisation d'une fonction coût, mesurant la différence entre la solution numérique et les observations expérimentales. L'approche variationnelle, basée sur le Lagrangien, permet de considérer l'adjoint, et l'utilisation d'un logiciel de différentiation automatique (TAPENADE) conduit à une identification rapide et précise des paramètres, quelles que soient la complexité et la taille du problème étudié. La qualité de l'identification peut être fortement instable et dépendre largement des données expérimentales en cas de bruit. Nous introduisons alors des termes de régularisation permettant de gérer la présence d'erreurs de mesure. Plusieurs cas d'usinage par jet abrasif sont considérés: problème stationnaire, jet qui se déplace à vitesse constante, ou en accélérant, utilisation synthétiques ou réelles L'étude de sensibilité montre la robustesse de l'approche, qui permet d'obtenir de très bons résultats acceptables d'un point de vue industriel / This work is part of STEEP Marie-Curie ITN project, covering the research in field of energy beam processing. We focus on the identification of unknown parameters of the proposed generic Abrasive WaterJet Milling (AWJM) model. The necessity of studying this problem comes from the industrial milling applications where the possibility to predict and model the final surface with high accuracy is one of the primary tasks in the absence of any knowledge of the model parameters that should be used. We propose the method of the model parameters identification by minimizing a cost function, measuring the difference between experimental observation and numerical solution. The variational approach based on corresponding Lagrangian allows to obtain the adjoint state and the involvement of the automatic differentiation software tool (TAPENADE) leads to fast and efficient parameters identification. In fact the parameter identification problem is highly unstable and strictly depends on quality of input data. Regularization terms could be effectively used to deal with the presence of measurement errors. Various cases of the AWJM process such as a stationary problem and moving with constant feed speed or acceleration are studied based on both artificial and real experimental data. The sensitivity study related to these particular problems demonstrates the strong capability of the proposed approach to obtain acceptable