Les travaux réalisés dans cette thèse concernent l’étude de problèmes quasi-linéaires paraboliques et elliptiques singuliers. Par singularité, nous signifions que le problème fait intervenir une non linéarité qui explose au bord du domaine où l’équation est posée. La présence du terme singulier entraine un manque de régularité des solutions. Ce défaut de régularité génère en conséquence un manque de compacité qui ne permet pas d’appliquer directement les méthodes classiques d’analyse non linéaires pour démontrer l’existence de solutions et discuter les propriétés de régularité et de comportement asymptotique des solutions. Pour contourner cette difficulté dans le contexte des problèmes que nous avons étudiés, nous sommes amenés à établir des estimations a priori très fines au voisinage du bord en combinant diverses méthodes : méthodes de monotonie (reliées au principe du maximum), méthodes variationnelles, argument de convexité, méthodes d’interpolation dans les espaces de Sobolev, méthodes de point fixe. / In this thesis I have studied the Evolution p-laplacian equation with singular nonlinearity. We start by studying the corresponding elliptic problem and then by defining a proper cone in a suitable Sobolev space find the uniqueness of the solution. Taking that into account and using the semi discretization in time we arrive at the uniqueness and existence result. Next we prove some regularity theorem using tools from Nonlinear Semigroup theory and Interpolation spaces. We also establish some related result for the laplacian case where we improve our result on the existence and regularity, due to the non degeneracy of the laplacian. In another related work we work with a semilinear equation with singular nonlinearity and using the moving plane method prove the symmetry properties of any classical solution. We also give some related apriori estimates which together with the symmetry provide us the existence of solution using the bifurcation result.

The work in this thesis is inspired by the fabrication of Zn₄₅Au₃₀Cu₂₅. This is the first alloy undergoing ultra-reversible martensitic transformations and closely satisfying the cofactor conditions, particular conditions of geometric compatibility between phases, which were conjectured to influence reversibility. With the aim of better understanding reversibility, in this thesis we study the martensitic microstructures arising during thermal cycling in Zn₄₅Au₃₀Cu₂₅, which are complex and different in every phase transformation cycle. Our study is developed in the context of continuum mechanics and nonlinear elasticity, and we use tools from nonlinear analysis. The first aim of this thesis is to advance our understanding of conditions of geometric compatibility between phases. To this end, first, we further investigate cofactor conditions and introduce a physically-based metric to measure how closely these are satisfied in real materials. Secondly, we introduce further conditions of compatibility and show that these are nearly satisfied by some twins in Zn₄₅Au₃₀Cu₂₅. These might influence reversibility as they improve compatibility between high and low temperature phases. Martensitic phase transitions in Zn₄₅Au₃₀Cu₂₅ are a complex phenomenon, especially because the crystalline structure of the material changes from a cubic to a monoclinic symmetry, and hence the energy of the system has twelve wells. There exist infinitely many energy-minimising microstructures, limiting our understanding of the phenomenon as well as our ability to predict it. Therefore, the second aim of this thesis is to find criteria to select physically-relevant energy minimisers. We introduce two criteria or selection mechanisms. The first involves a moving mask approximation, which allows one to describe some experimental observations on the dynamics, while the second is based on using vanishing interface energy. The moving mask approximation reflects the idea of a moving curtain covering and uncovering microstructures during the phase transition, as appears to be the case for Zn₄₅Au₃₀Cu₂₅, and many other materials during thermally induced transformations. We show that the moving mask approximation can be framed in the context of a model for the dynamics of nonlinear elastic bodies. We prove that every macroscopic deformation gradient satisfying the moving mask approximation must be of the form 1 + a(x) ⊗ n(x), for a.e. x. With regards to vanishing interface energy, we consider a one-dimensional energy functional with three wells, which simplifies the physically relevant model for martensitic transformations, but at the same time highlights some key issues. Our energy functional admits infinitely many minimising gradient Young measures, representing energy-minimising microstructures. In order to select the physically relevant ones, we show that minimisers of a regularised energy, where the second derivatives are penalised, generate a unique minimising gradient Young measure as the perturbation vanishes. The results developed in this thesis are motivated by the study of Zn₄₅Au₃₀Cu₂₅, but their relevance is not limited to this material. The results on the cofactor conditions developed here can help for the understanding of new alloys undergoing ultra-reversible transformations, and as a guideline for the fabrication of future materials. Furthermore, the selection mechanisms studied in this work can be useful in selecting physically relevant microstructures not only in Zn₄₅Au₃₀Cu₂₅, but also in other materials undergoing martensitic transformations, and other phenomena where pattern formation is observed.

We approximate the solution of a quasilinear stochastic partial differential equa- tion driven by fractional Brownian motion B_H(t); H in (0,1), which was calculated via fractional White Noise calculus, see [5].

[pt] Esse trabalho tem como objetivo ser uma introdução ao estudo da existência e unicidade de soluções fracas para equações diferenciais parciais elípticas. Começamos definindo o espaço de Sobolev para, a partir da definição, provarmos algumas propriedades básicas que nos ajudarão no estudo das equações diferenciais parciais elípticas. Finalizamos com o desenvolvimento do Teorema de Lax-Milgram e de Stampacchia que permitirão o uso de técnicas de Análise Funcional para estudarmos alguns exemplos de equações elípticas. / [en] This dissertation aims to be an introduction to the study of the existence and uniqueness of weak solutions for elliptic partial differential equations. We begin by defining the Sobolev spaces and proving some basics properties that will assist in the study of the elliptical equations. Lastly, we develop the Theorems of Lax-Milgram and Stampacchia that allow the use of Functional Analysis for the studying of some examples of elliptic equations.

abstract: In order to verify the dispersive nature of transverse displacement in a beam, a deep understanding of the governing partial differential equation is developed. Using the finite element method and Newmark’s method, along with Fourier transforms and other methods, the aim is to obtain consistent results across each numerical technique. An analytical solution is also analyzed for the Euler-Bernoulli beam in order to gain confidence in the numerical techniques when used for more advance beam theories that do not have a known analytical solution. Three different beam theories are analyzed in this report: The Euler-Bernoulli beam theory, Rayleigh beam theory and Timoshenko beam theory. A comparison of the results show the difference between each theory and the advantages of using a more advanced beam theory for higher frequency vibrations. / Dissertation/Thesis / Masters Thesis Civil Engineering 2016

Submitted by Izabel Franco (izabel-franco@ufscar.br) on 2016-09-23T14:42:46Z No. of bitstreams: 1 DissPLS.pdf: 887262 bytes, checksum: e54f2438d019bad9fa31a2f0e8b98d66 (MD5) / Approved for entry into archive by Marina Freitas (marinapf@ufscar.br) on 2016-09-26T20:40:36Z (GMT) No. of bitstreams: 1 DissPLS.pdf: 887262 bytes, checksum: e54f2438d019bad9fa31a2f0e8b98d66 (MD5) / Approved for entry into archive by Marina Freitas (marinapf@ufscar.br) on 2016-09-26T20:40:46Z (GMT) No. of bitstreams: 1 DissPLS.pdf: 887262 bytes, checksum: e54f2438d019bad9fa31a2f0e8b98d66 (MD5) / Made available in DSpace on 2016-09-26T20:40:52Z (GMT). No. of bitstreams: 1 DissPLS.pdf: 887262 bytes, checksum: e54f2438d019bad9fa31a2f0e8b98d66 (MD5) Previous issue date: 2015-04-24 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / The Rosenau-Hyman, or K(m, n), equations are a generalized version of the Korteweg-de Vries (KdV) equation where the dipersive term is non-linear. Such partial differential equations not always have a specific method by which can be solved, besides the solutions are not always analytical. The Lie symmetry method was applied to look for solutions of these equations. This method consists in finding the most general symmetry group of the equation, wherewith the solution can be found. It was found an expression to the solution and to some particular cases. It was shown that in the case K(2, 2) a new kind of solution, called compacton, with peculiar properties is found. / Equações Rosenau-Hyman, ou K(m, n), são uma versão generalizada da equação Kortewegde Vries (KdV) em que o termo dispersivo é não-linear. Essas equações diferencias nãolineares nem sempre possuem um método específico pelo qual podem ser resolvidas, além de que as soluções nem sempre são analíticas. O método de simetria de Lie foi aplicado para buscar por soluções dessas equações. Esse método consiste em encontrar o grupo de simetria mais geral da equação, por meio do qual a solução pode ser encontrada. Obteve-se uma expressão para a solução e alguns casos particulares. Foi mostrado que para K(2, 2) um novo tipo de solução, chamada compacton, com propriedades peculiares é encontrado.

Submitted by Maike Costa (maiksebas@gmail.com) on 2016-03-29T13:33:49Z No. of bitstreams: 1 arquivo total.pdf: 5320246 bytes, checksum: 158dd460a20ce46c96d4a34623612264 (MD5) / Made available in DSpace on 2016-03-29T13:33:49Z (GMT). No. of bitstreams: 1 arquivo total.pdf: 5320246 bytes, checksum: 158dd460a20ce46c96d4a34623612264 (MD5) Previous issue date: 2011-08-25 / This work is mainly concerned with the existence and nonexistence of classical solution to the nonhomogeneous semilinear equation Δu + up + f(x) = 0 in Rn, u > 0 in Rn, when n 3, where f 0 is a Hölder continuous function. The nonexistence of classical solution is established when 1 < p n=(n 􀀀 2). For p > n=(n 􀀀 2) there may be both existence and nonexistence results depending on the asymptotic behavior of f at infinity. The existence results were obtained by employed sub and supersolutions techniques and fixed point theorem. For the nonexistence of classical solution we used a priori integral estimates obtained via averaging. / Neste trabalho, estamos interessados na existência e não existência de solução clássica para a equação não homogênea semilinear Δu + up + f(x) = 0 em Rn; u > 0 em Rn, n 3 onde f 0 é uma função Hölder contínua. A não existência de solução clássica é estabelecida quando 1 < p n=(n 􀀀 2). Para p > n=(n 􀀀 2), temos resultados de existência e não existência de solução clássica, dependendo do comportamento assin- tótico de f no infinito. Os resultados de existência foram obtidos usando o método de sub e supersolução e teoremas de ponto fixo. A não existência de solução clássica é obtida usando-se estimativas integrais a priori via média esférica.

Orientador: Heloisa Helena Marino Silva / Banca: Antonio Castelo Filho / Banca: Maurílio Boaventura / Resumo: O objetivo deste trabalho þe apresentar um estudo sobre a aplicação de métodos numéricos para a resolução do modelo proposto por Barcelos, Boaventura e Silva Jr. [7], para a eliminação de ruídos em imagens digitais por meio de uma equação diferencial parcial, e propor uma anþalise da estabilidade do mþetodo iterativo comumente aplicado a este modelo. Uma anþalise comparativa entre os vários mþetodos abordados þe realizada atravþes de resultados experimentais em imagens sintéticas e imagens da vida real. / Abstract: The purpose of this work is to present a study on the application of numerical methods for the resolution of model considered by Barcelos, Boaventura and Silva Jr [7], for image denoising through a partial di erential equation, and to consider a stability analysis of an iterative method usually applied to this model. A comparative analysis among various considered methods is carried out through experimental results for synthetic and real images. / Mestre

Orientadores: Jose Mario Martinez, Sandra Augusta Santos / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Matematica, Estatistica e Computação Cientifica / Made available in DSpace on 2018-08-12T03:24:09Z (GMT). No. of bitstreams: 1 Pisnitchenko_Fedor_D.pdf: 2087061 bytes, checksum: a87c208fa0681c4ecec62408f70f29ae (MD5) Previous issue date: 2008 / Resumo: Em ciência e engenharia existe uma vasta classe de problemas que consistem em resolver um sistema de equações diferenciais parciais para encontrar as variáveis (como velocidade, temperatura, deslocamento, etc), dada a informação de decisão necessária (como domínio, condições iniciais e de contorno, etc). Entretanto, para os problemas reais são muito comuns situações em que a informação de decisão seja incompleta e contenha erros, e, por outro lado, exista alguma informação sobre as variáveis de estado, obtida de uma outra simulação ou de algum tipo de observação (dados observados). Uma forma natural de resolver esse tipo de problema, utilizando toda a informação de decisão, é interpretá-lo como um problema de otimização. Ou seja, minimizar alguma função objetivo escolhida como a distância entre os dados observados e as variáveis de estado, sujeito à discretização do sistema. Neste trabalho propomos um método Quase-Newton para resolver o problema EDP restrito utilizando como modelos a equação unidimensional de Rossby-Obukhov e a equação de Kortewegde Vries. Um aspecto muito importante do método é não ter restrição de estabilidade para escolha dos passos na discretização das equações diferenciais. Um outro é poder utilizar passos maiores, em comparação com os métodos tradicionais evolutivos como diferenças finitas. Foi realizado um grande número de testes computacionais. Os resultados obtidos foram muito promissores, mostrando a robustez do método e a possibilidade de resolver problemas de grande porte. / Abstract: In science and engeneering there is a wide class of problems that consist in solving a system of partial differential equations to find variables (such as velocity, temperature, displacement, etc.), given the necessary decision information (such as domain, initial and boundary conditions, etc.). However,it is very common for real problems that the decision information is incomplete and contains errors. On the other hand, there is some additional information about state variables, which come from other simulation or some kind of observations (observed data). A natural way to solve this kind of problem, using all the decision information, is to interpret it as an optimization problem. That is, minimize an objective function chosen such as distance between the observed data and the state variables, subject to the system discretization. In this work, we propose a Quasi-Newton method to solve the PDE-constrained problem using as models the unidimensional Rossby-Obukhov and Korteweg-de Vries equations. A very importante aspect of the method is that there is no stability restriction for the stepsize in the differential equations discretization. Another aspect is to be able to use stepsizes larger than the ones used in traditional evolutive methods such as finite differences. A large number of computational test was performed. The results were promising and showed the robustness of the method and its ability to solve large scale problems. / Doutorado / Otimização / Doutor em Matemática Aplicada

Cette thèse porte sur l'étude des dynamiques spatiales et évolutives d'une population à l'aide d'outils probabilistes et déterministes. Dans la première partie, nous cherchons à comprendre l'effet de l'hétérogénéité de l'environnement sur l'évolution des espèces. La population considérée est modélisée par un processus individu-centré avec interactions qui décrit les événements de naissances, morts, mutations et diffusions spatiales de chaque individu. Les taux des événements dépendent des caractéristiques des individus : traits phénotypes et positions spatiales. Dans un premier temps, nous étudions le système d'équations aux dérivées partielles qui décrit la dynamique spatiale et démographique d'une population composée de deux traits dans une limite grande population. Nous caractérisons précisément les conditions d'extinction et de survie en temps long de cette population. Dans un deuxième temps, nous étudions le modèle individuel initial sous deux asymptotiques : grande population et mutations rares de telle sorte que les échelles de temps démographiques et mutationnelles sont séparées. Ainsi, lorsqu'un mutant apparaît, la population résidente est à l'équilibre démographique. Nous cherchons alors à caractériser la probabilité de survie de la population issue de ce mutant. Puis, en étudiantle processus dans l'échelle des mutations, nous prouvons que le processus individu-centré converge vers un processus de sauts qui décrit les fixations successives des traits les plus avantagés ainsi que la répartition spatiale des populations portant ces traits. Nous généralisons ensuite le modèle pour introduire des interactions de type mutualiste entre deux espèces. Nous étudions ce modèle dans une limite de grande population. Nous donnons par ailleurs des résultats numériques et une analyse biologique détaillée des comportements obtenus autour de deux problématiques : la coévolution de niches spatiales et phénotypiques d'espèces en interaction mutualiste et les dynamiques d'invasions d'un espace homogène par des espèces mutualistes. Dans la deuxième partie de cette thèse, nous développons un modèle probabiliste pour étudier finement l'effet d'une préférence sexuelle sur la spéciation. La population est ici structurée sur deux patchs et les individus, caractérisés par un trait, sont écologiquement et démographiquement équivalents et se distinguent uniquement par leur préférence sexuelle: deux individus de même trait ont plus de chance de se reproduire que deux individus de traits distincts. Nous montrons qu'en l'absence de toute autre différence écologique, la préférence sexuelle mène à un isolement reproductif entre les deux patchs. / We study the spatial and evolutionary dynamics of a population by using probabilistic and deterministic tools. In the first part of this thesis, we are concerned with the influence of a heterogeneous environment on the evolution of species. The population is modeled by an individual-based process with some interactions and which describes the birth, the death, the mutation and the spatial diffusion of each individual. The rates of those events depend on the characteristics of the individuals : their phenotypic trait and their spatial location. First, we study the system of partial differential equations that describes the spatial and demographic dynamics of a population composed of two traits in a large population limit. We characterize precisely the conditions of extinction and long time survival for this population. Secondly, we study the initial individual-based model under two asymptotic: large population and rare mutations such as demographic and mutational timescales are separated. Thus, when a mutant appears, the resident population has reached its demographic balance. We characterize the survival probability of the population descended from this mutant. Then, by studyingthe process in the mutational scale, we prove that the microscopic process converges to a jump process which describes the successive fixations of the most advantaged traits and the spatial distribution of populations carrying these traits. We then extend the model to introduce mutualistic interactions between two species. We study this model in a limit of large population. We also give numerical results and a detailed biological behavior analysis around two issues: the co-evolution of phenotypic and spatial niches of mutualistic species and the invasion dynamics of a homogeneous space by these species. In the second part of this thesis, we develop a probabilistic model to study the effect of the sexual preference on the speciation. Here, the population is structured on two patches and the individuals, characterized by a trait, are ecologically and demographically similar and differ only in their sexual preferences: two individuals of the same trait are more likely to reproduce than two individuals of distinct traits. We show that in the absence of any other ecological differences, the sexual preferences lead to reproductive isolation between the two patches.