We develop well-posedness theory and analytical and numerical solution techniques for Boussinesq-type equations. Firstly, we consider the Cauchy problem for a generalized Boussinesq equation. We show that under suitable conditions, a global solution for this problem exists. In addition, we derive sufficient conditions for solution blow-up in finite time. / Secondly, a generalized Jacobi/exponential expansion method for finding exact solutions of non-linear partial differential equations is discussed. We use the proposed expansion method to construct many new, previously undiscovered exact solutions for the Boussinesq and modified Korteweg-de Vries equations. We also apply it to the shallow water long wave approximate equations. New solutions are deduced for this system of partial differential equations. / Finally, we develop and validate a numerical procedure for solving a class of initial boundary value problems for the improved Boussinesq equation. The finite element method with linear B-spline basis functions is used to discretize the equation in space and derive a second order system involving only ordinary derivatives. It is shown that the coefficient matrix for the second order term in this system is invertible. Consequently, for the first time, the initial boundary value problem can be reduced to an explicit initial value problem, which can be solved using many accurate numerical methods. Various examples are presented to validate this technique and demonstrate its capacity to simulate wave splitting, wave interaction and blow-up behavior.

We consider linear and semilinear stochastic partial differential equations that in some sense can be viewed as being at the "endpoints" of the classical variational theory by Krylov and Rozovskii [25]. In terms of regularity of the coeffcients, the minimal assumption is boundedness and measurability, and a unique L2- valued solution is then readily available. We investigate its further properties, such as higher order integrability, boundedness, and continuity. The other class of equations considered here are the ones whose leading operators do not satisfy the strong coercivity condition, but only a degenerate version of it, and therefore are not covered by the classical theory. We derive solvability in Wmp spaces and also discuss their numerical approximation through finite different schemes.

First we consider implicit finite difference schemes on uniform grids in time and space for second order linear stochastic partial differential equations of parabolic type. Under sufficient regularity conditions, we prove the existence of an appropriate asymptotic expansion in powers of the the spatial mesh and hence we apply Richardson's method to accelerate the convergence with respect to the spatial approximation to an arbitrarily high order. Then we extend these results to equations where the parabolicity condition is allowed to degenerate. Finally, we consider implicit finite difference approximations for deterministic linear second order partial differential equations of parabolic type and give sufficient conditions under which the approximations in space and time can be simultaneously accelerated to an arbitrarily high order.

L'objet de cette thèse porte sur le problème de Cauchy singulier dans le domaine complexe. Il s'agit d'étudier les singularités de la solution du problème pour trois classes d'équations aux dérivées partielles. Cette thèse s'inscrit dans la continuité des travaux initiés par Jean Leray et son école. Pour décrire les singularités de la solution, on cherche la solution sous la forme d'un développement asymptotique de fonctions hypergéométriques de Gauss. Comme les singularités sont portées par les fonctions hypergéométriques, l'étude de la ramification de la solution se ramène à celle de ces fonctions. / This thesis deals with the singular Cauchy problem in the complex domain. We study the singularities of the solution of the problem for three classes of partial differential equations. This thesis is a continuation of the work initiated by Jean Leray and his school. To describe the singularities of the solution, we seek the solution in the form of asymptotic an expansion of Gauss hypergeometric functions. As the singularities are carried by the hypergeometric functions, the study of the ramification of the solution reduces to that of these functions.

We try to explain the mathematical theory of thin liquid film evolution. We start with introducing physical processes in which thin film evolution plays an important role. Derivation of the classical thin film equation and existing mathematical theory in the literature are also introduced. To explain the thin film evolution we derive a new family of degenerate parabolic equations. We prove results on existence, uniqueness, long time behavior, regularity and support properties of solutions for this equation. At the end of the thesis we consider the classical thin film Cauchy problem on the whole real line for which we use asymptotic equipartition to show H^1(R) convergence of solutions to the unique self-similar solution.

Orientador: Germán Jesus Lozada Cruz / Banca: Marcos Roberto Teixeira Primo / Banca: Andréa Cristina Prokopezyk Arita / Resumo: Neste trabalho vamos estudar a existência e unicidade de solução para equações da forma { u + Au = f(t,u) u(t0)= u0 ∈ X, (I) onde X é um espaço de Banach, A : D(A) ⊂ X → X é um operador linear, f é uma função não linear conhecida, u0 ∈ X é um dado inical conhecido e u : I ⊂ R → X é uma função desconhecida e t0 ∈ I. Faremos este estudo usando a Teoria dos Semigrupos de Operadores Lineares. Para melhor entendimento do estudo das equações (I), faremos duas aplicações. A primeira tratando de um modelo (linear) de divisão celular e a segunda, do modelo (não linear) de condução do calor. / Abstract: In this work we will study the existence and uniqueness of the solutions for the following equation { u + Au = f(t,u) u(t0)= u0 ∈ X, (I) where X is a Banach space, A : D(A) ⊂ X → X is a linear operator, f is a nonlinear function, u : I ⊂ R → X is unknown function. In this study we will use the theory of semigroup of linear operators. For a best understanding of the study of equations (I), we will do two applications. The first one, is a (linear) model of cellular division and the second one, is about the (nonlinear) model od conduction of the heat. / Mestre

Dans cette thèse nous étudions le problème de Cauchy en relativité générale. Motivés par la conjecture de censure cosmique faible formulée par Penrose, nous analysons le problème aux données initiales pour les équations d'Einstein dans le vide en faible régularité. Nous démontrons les deux résultats suivants. o Premièrement, nous nous intéressons aux équations de contrainte pour les données initiales et mettons en place une procédure de prolongement. Plus précisément, étant donné des données initiales pour les équations d'Einstein sur la boule unité dans R3, nous les prolongeons de manière continue en des données globales, asymptotiquement plates sur R3. Les équations de contrainte forment un système couplé d'équations non-lineaires sous-determinées géométriques. La preuve de notre procédure de prolongement repose sur un schéma iteratif où nous séparons ce système en deux problèmes de prolongement decouplés et solubles. Enfin, le résultat de prolongement pour les équations de contrainte est obtenu par un argument de point fixe. o Deuxièment, nous prouvons une version localisée du théorème de courbure L2 de Klainerman-Rodnianski-Szeftel. Nous montrons que, étant données des données initiales pour les équations d'Einstein sur une variété compacte avec bord, le temps d'existence de la solution des équations d'Einstein dans le domaine de dépendance de ces données initiales ne dépend que de normes de basse régularité des données initiales. En particulier, notre résultat est un critère localisé de continuité pour les équations d'Einstein. Notre preuve utilise un argument de localisation où, tout d'abord, nous généralisons la théorie de Cheeger-Gromov de convergence pour les variétés Riemanniennes à notre cas de régularité faible, et ensuite nous appliquons la procédure de prolongement pour les équations de contrainte mentionnée ci-dessus avec un argument de changement d’échelle. / In this thesis we study the Cauchy problem of general relativity. Motivated by the weak cosmic censorship conjecture formulated by Penrose, we analyse the initial value problem for the Einstein vacuum equations in low regularity. We prove the following two results. First, we consider the constraint equations of the initial data and demonstrate an extension procedure. More precisely, given small initial data for the Einstein equations on the unit ball in R3, we continuosly extend it to global, asymptotically flat initial data on R3. The constraint equations for the Einstein vacuum equations are a coupled system of non-linear under-determined geometric elliptic equations. The proof of our extension procedure is based on an iterative scheme where we split this system into two decoupled, solvable extension problems. The extension result for the constraint equations follows then by a fix point argument. Second, we prove a localised version of the bounded L2-curvature theorem by Klainerman-Rodnianski-Szeftel. We show that given low regularity initial data to the Einstein equations on a compact manifold with boundary, the time of existence of the solution to the Einstein equations in the domain of dependence of the initial data depends only on low regularity geometric data. In particular, this result is a localised continuation criterion for the Einstein vacuum equations. Our proof uses a localisation argument where we first generalise the known Cheeger-Gromov convergence theory for Riemannian manifolds to our low regularity setting, and then apply the above extension procedure for the constraint equations with a scaling argument.

Le but de cette thèse est d'étudier et de développer des méthodes permettant de résoudre deux types de problèmes d'identification portant sur des équations elliptiques. Ces problèmes étant connus pour leur caractère fortement instable, les méthodes proposées s'accompagnent de procédures de régularisation, qui permettent d'assurer que la solution obtenue conserve un sens physique.Dans un premier temps, on étudie la résolution du problème de Cauchy (identification de conditions aux limites) par la méthode de Steklov-Poincaré. On commence par proposer quelques améliorations basées sur le solveur de Krylov utilisé, en introduisant notamment une méthode de régularisation consistant à tronquer la décomposition de Ritz de l'opérateur concerné. Par la suite, on s'intéresse à l'estimation d'incertitude en utilisant des techniques issues de l'inversion Bayésienne. Enfin, on cherche à résoudre des problèmes plus exigeants, à savoir un problème transitoire en temps, un cas non-linéaire, et on donne des éléments pour effectuer des résolutions sur des géométries ayant un très grand nombre de degrés de liberté en s'aidant de la décomposition de domaine.Pour ce qui est du problème d'identification de fissures par la méthode de l'écart à la réciprocité, on commence par proposer et tester numériquement différents moyens de stabiliser la résolution (utilisation de fonctions-tests différentes, minimisation des gradients a posteriori ou régularisation de Tikhonov). Puis on présente une autre variante de la méthode de l'écart à la réciprocité, qui est applicable aux cas pour lesquels les mesures sont incomplètes. Cette méthode, basée sur une approche de Petrov-Galerkine, est confrontée entre autres à un cas expérimental. Enfin, on s'intéresse à certaines idées permettant d'étendre la méthode de l'écart à la réciprocité à l'identification de fissures non planes. / The goal of this thesis is to study and to develop some methods in order to solve two types of identification problems in the framework of elliptical equations. As those problems are known to be particularly unstable, the proposed methods are accompanied with regularization procedures, that ensure that the obtained solutions keep a physical meaning.Firstly, we study the resolution of the Cauchy problem (boundary conditions identification) by the Steklov-Poincaré method. We start by proposing some improvements based on the used Krylov solver, especially by introducing a regularization method that consists in truncating the Ritz values decomposition of the operator in question. We study afterwards the estimation of uncertainties by the mean of techniques stemming from Bayesian inversion. Finally, we aim at solving more demanding problems, namely a time-transient problem, a non-linear case, and we give some elements to carry out resolutions on geometries that have a very high number of degrees of freedom, with help of domain decomposition.As for the problem of crack identification by the reciprocity gap method, we firstly propose and numerically test some ways to stabilize the resolution (use of different test-functions, a posteriori minimization of the gradients or Tikhonov regularization). Then we present an other variant of the reciprocity gap method, that is applicable on cases for which the measurements are incomplete. This method, based on a Petrov-Galerkin approach, is confronted, among others, with an experimental case. Finally, we investigate some ideas that allow to extend the reciprocity gap method for the identification of non-plane cracks.

Elastic waves describe particles vibrating in materials holding the property of elasticity. Particularly, several kinds of resistance in elasticity lead to the models of elastic waves with different damping mechanisms. In the thesis, the influence from friction, structural damping, Kelvin-Voigt damping on the linear and semilinear elastic waves in two or three dimensions are studied. Concerning the Cauchy problem for linear elastic waves, some qualitative properties of solutions including well-posedness, smoothing effect, propagation of singularities, energy estimates and diffusion phenomena, are derived by using WKB analysis associated with diagonalization procedures or the spectral theory. By constructing suitable time-weighted Sobolev spaces and using Banach's fixed point theorem, global (in time) existence of small data solutions to the weakly coupled systems for semilinear elastic waves with different damping terms have been proved. The main tools to treat the nonlinear terms in Sobolev spaces are some fractional tools in Harmonic Analysis. Finally, well-posedness and Lp-Lq estimates for elastic waves without any damping terms in three dimensions are analyzed by employing Riesz transform theory and stationary phase methods.

The subject of the thesis is the investigation of asymptotic properties of solutions of the Cauchy problem for structurally damped sigma-evolution operators with time dependent, monotonous, dissipation term. An appropriate energy for solutions of the sigma-evolution equations is defined and some estimates for energies of higher order are proved. In the scale invariant case the optimality of these estimates is shown. Further, the influence of properties of the time dependent dissipation on L^p-L^q estimates for the energy with p and q bigger or equal to 2 and from the conjugate line is clarified. Also smoothing properties of the operators under consideration are investigated. The connection between the regularity of the data and the regularity of the solution in terms of L^2 based Gevrey spaces is considered. Finally, L^1-L^1-estimates in the special case delta = sigma/2 and decreasing dissipative coefficient. / Thema der vorliegenden Dissertation ist die Untersuchung asymptotischer Eigenschaften von Lösungen des Cauchy Problems für strukturell gedämpfte sigma-Evolutions-Operatoren mit zeitabhängigem, monotonen Dissipationskoeffizienten. Es wird eine geeignete Energie definiert und für diese Abschätzungen, auf für entsprechende Energien höherer Ordnung gezeigt. Darüber hinaus wird der Einfluss des Dissipationskoeffizienten auf L^p-L^q Abschätzungen auf und entfernt von der konjugierten Linie untersucht. Im skaleninvarianten Fall wird die Schärfe der Abschätzungen bewiesen. Weiterhin wird der Zusammenhang zwischen der Regularität der Daten und der der Lösung in Termen von L^2-basierten Gevrey-Räumen untersucht. Schließlich werden L^1-L^1-Abschätzungen für den Spezialfall delta = sigma/2 und monoton fallenden Dissipationskoeffizienten gezeigt.