No description available.

Numerical methods based on rational Krylov spaces have become an indispensable tool of scientific computing. In this thesis we study rational Krylov spaces by considering rational Krylov decompositions; matrix relations which, under certain conditions, are associated with these spaces. We investigate the algebraic properties of such decompositions and present an implicit Q theorem for rational Krylov spaces. We derive standard and harmonic Ritz extraction strategies for approximating the eigenpairs of a matrix and for approximating the action of a matrix function onto a vector. While these topics have been considered previously, our approach does not require the last pole to be infinite, which makes the extraction procedure computationally more efficient. Typically, the computationally most expensive component of the rational Arnoldi algorithm for computing a rational Krylov basis is the solution of a large linear system of equations at each iteration. We explore the option of solving several linear systems simultaneously, thus constructing the rational Krylov basis in parallel. If this is not done carefully, the basis being orthogonalized may become poorly conditioned, leading to numerical instabilities in the orthogonalization process. We introduce the new concept of continuation pairs which gives rise to a near-optimal parallelization strategy that allows to control the growth of the condition number of this non orthogonal basis. As a consequence we obtain a more accurate and reliable parallel rational Arnoldi algorithm. The computational benefits are illustrated using our high performance C++ implementation. We develop an iterative algorithm for solving nonlinear rational least squares problems. The difficulty is in finding the poles of a rational function. For this purpose, at each iteration a rational Krylov decomposition is constructed and a modified linear problem is solved in order to relocate the poles to new ones. Our numerical results indicate that the algorithm, called RKFIT, is well suited for model order reduction of linear time invariant dynamical systems and for optimisation problems related to exponential integration. Furthermore, we derive a strategy for the degree reduction of the approximant obtained by RKFIT. The rational function obtained by RKFIT is represented with the aid of a scalar rational Krylov decomposition and an additional coefficient vector. A function represented in this form is called an RKFUN. We develop efficient methods for the evaluation, pole and root finding, and for performing basic arithmetic operations with RKFUNs. Lastly, we discuss RKToolbox, a rational Krylov toolbox for MATLAB, which implements all our algorithms and is freely available from http://rktoolbox.org. RKToolbox also features an extensive guide and a growing number of examples. In particular, most of our numerical experiments are easily reproducible by downloading the toolbox and running the corresponding example files in MATLAB.

[pt] Neste trabalho focamos sobre alguns modelos matemáticos na área do petróleo, com o objetivo de propor um modelo inicial de simulador numérico de reservatórios. Inicialmente apresentamos uma EDP do calor não-linear com um termo fonte de calor constante, estudada para o domínio sendo uma placa plana quadrada homogênea e heterogênea, onde aplicamos soluções numéricas utilizando o método das diferenças finitas implícito. Abordamos o problema de refinamento da malha no entorno dos poços utilizando o método JFNK (Jacobian-Free Newton-Krylov), que aumenta a eficiência computacional através de uma aproximação para a matriz Jacobiana. Por fim resolvemos um sistema de EDPs não-lineares que representam o escoamento bifásico de água e óleo, constituído por equações de transporte em termos da pressão e da saturação. Fizemos simulações numéricas de alguns casos conhecidos e os resultados mostraram uma boa qualidade no nosso método. / [en] In this work we focus on the numerical approximation of some mathematical models in the oil field. First, we present a non-linear heat equation with a constant heat source term, studied for the domain of a homogeneous and heterogeneous square domain, where we apply numerical solutions using an implicit finite difference method. We approach the problem of mesh refinement around the wells using the JFNK (Jacobian- Free Newton-Krylov) method, which improves the computational efficiency through an approximation to the Jacobian matrix. Finally, we solve a system of non-linear EDPs that represent the two-phase flow of water and oil, consisting of equations of transport in terms of pressure and saturation. Numerical simulations for some known cases showed accurate approximation of our method.

[pt] A representação matemática de comportamentos complexos em materiais exige formulações constitutivas sofisticada, como é o caso de modelos com múltiplas superfícies de plastificação. Assim, um modelo elastoplástico complexo demanda um procedimento robusto de integração das equações de evolução plástica. O desenvolvimento de esquemas de integração para modelos de plasticidade é um tópico de pesquisa importante, já que estes estão diretamente ligados à acurácia e eficiência de simulações numéricas de materiais como metais, concretos, solos e rochas. O desempenho da solução de elementos finitos é diretamente afetado pelas características de convergência do procedimento de atualização de estados. Dessa forma, este trabalho explora a implementação de modelos constitutivos complexos, focando em modelos genéricos com múltiplas superfícies de plastificação. Este estudo formula e avalia algoritmos de atualização de estado que formam uma estrutura robusta para a simulação de materiais regidos por múltiplas superfícies de plastificação. Algoritmos de integração implícita são desenvolvidos com ênfase na obtenção de robustez, abrangência e flexibilidade para lidar eficazmente com aplicações complexas de plasticidade. Os algoritmos de atualização de estado, baseados no método de Euler implícito e nos métodos de Newton-Raphson e Newton-Krylov, são formulados utilizando estratégias de busca unidimensional para melhorar suas características de convergência. Além disso, é implementado um esquema de subincrementação para proporcionar mais robustez ao procedimento de atualização de estado. A flexibilidade dos algoritmos é explorada, considerando várias condições de tensão, como os estados plano de tensões e plano de deformações, num esquema de integração único e versátil. Neste cenário, a robustez e o desempenho dos algoritmos são avaliados através de aplicações clássicas de elementos finitos. Além disso, o cenário desenvolvido no contexto de modelos com múltiplas superfícies de plastificação é aplicado para formular um modelo elastoplástico com dano acoplado, que é avaliado através de ensaios experimentais em estruturas de concreto. Os resultados obtidos evidenciam a eficácia dos algoritmos de atualização de estado propostos na integração de equações de modelos com múltiplas superfícies de plastificação e a sua capacidade para lidar com problemas desafiadores de elementos finitos. / [en] The mathematical representation of complex material behavior requires a sophisticated constitutive formulation, as it is the case of multisurface plasticity. Hence, a complex elastoplastic model demands a robust integration procedure for the plastic evolution equations. Developing integration schemes for plasticity models is an important research topic because these schemes are directly related to the accuracy and efficiency of numerical simulations for materials such as metals, concrete, soils and rocks. The performance of the finite element solution is directly influenced by the convergence characteristics of the state-update procedure. Therefore, this work explores the implementation of complex constitutive models, focusing on generic multisurface plasticity models. This study formulates and evaluates state-update algorithms that form a robust framework for simulating materials governed by multisurface plasticity. Implicit integration algorithms are developed with an emphasis on achieving robustness, comprehensiveness and flexibility to handle cumbersome plasticity applications effectively. The state-update algorithms, based on the backward Euler method and the Newton-Raphson and Newton-Krylov methods, are formulated using line search strategies to improve their convergence characteristics. Additionally, a substepping scheme is implemented to provide further robustness to the state-update procedure. The flexibility of the algorithms is explored, considering various stress conditions such as plane stress and plane strain states, within a single, versatile integration scheme. In this scenario, the robustness and performance of the algorithms are assessed through classical finite element applications. Furthermore, the developed multisurface plasticity background is applied to formulate a coupled elastoplastic-damage model, which is evaluated using experimental tests in concrete structures. The achieved results highlight the effectiveness of the proposed state-update algorithms in integrating multisurface plasticity equations and their ability to handle challenging finite element problems.

Nous nous intéressons, dans cette thèse, à l'étude des méthodes d'extrapolation polynômiales et à l'application de ces méthodes dans l'accélération de méthodes de points fixes pour des problèmes donnés. L'avantage de ces méthodes d'extrapolation est qu'elles utilisent uniquement une suite de vecteurs qui n'est pas forcément convergente, ou qui converge très lentement pour créer une nouvelle suite pouvant admettreune convergence quadratique. Le développement de méthodes cycliques permet, deplus, de limiter le coût de calculs et de stockage. Nous appliquons ces méthodes à la résolution des équations de Navier-Stokes stationnaires et incompressibles, à la résolution de la formulation Kohn-Sham de l'équation de Schrödinger et à la résolution d'équations elliptiques utilisant des méthodes multigrilles. Dans tous les cas, l'efficacité des méthodes d'extrapolation a été montrée.Nous montrons que lorsqu'elles sont appliquées à la résolution de systèmes linéaires, les méthodes d'extrapolation sont comparables aux méthodes de sous espaces de Krylov. En particulier, nous montrons l'équivalence entre la méthode MMPE et CMRH. Nous nous intéressons enfin, à la parallélisation de la méthode CMRH sur des processeurs à mémoire distribuée et à la recherche de préconditionneurs efficaces pour cette même méthode. / In this thesis, we study polynomial extrapolation methods. We discuss the design and implementation of these methods for computing solutions of fixed point methods. Extrapolation methods transform the original sequance into another sequence that converges to the same limit faster than the original one without having explicit knowledge of the sequence generator. Restarted methods permit to keep the storage requirement and the average of computational cost low. We apply these methods for computing steady state solutions of incompressible flow problems modelled by the Navier-Stokes equations, for solving the Schrödinger equation using the Kohn-Sham formulation and for solving elliptic equations using multigrid methods. In all cases, vector extrapolation methods have a useful role to play. We show that, when applied to linearly generated vector sequences, extrapolation methods are related to Krylov subspace methods. For example, we show that the MMPE approach is mathematically equivalent to CMRH method. We present an implementation of the CMRH iterative method suitable for parallel architectures with distributed memory. Finally, we present a preconditioned CMRH method.