Neste trabalho buscou-se consolidar aspectos referentes à otimização de problemas envolvidos na mecânica dos meios contínuos, envolvendo diferentes áreas do conhecimento, tais como: otimização matemática, diferenciação automática, análise estrutural, análise aerodinâmica, parametrização de curvas, superfícies e sólidos do tipo B-spline racionais não-uniformes (NURBS, acrônimo do inglês), análise IsoGeométrica (IGA, acrônimo do inglês) e análise por Elementos Finitos (FEA, acrônimo do inglês). Como objetivo final busca-se otimizar formas de cascas estruturais e formas de corpos aerodinâmicos imersos em escoamentos compressíveis. No que concerne à análise estrutural, esta é realizada via análise IsoGeométrica utilizando elementos sólidos para modelar cascas. Uma cinemática co-rotacional abrangente e precisa baseada na exata decomposição polar é desenvolvida, para lidar com problemas estáticos e dinâmicos altamente não lineares. Na análise estática foram implementados o método de Newton-Raphson e controle de deslocamentos generalizado, para problemas dinâmicos foram implementados o método -generalizado (G) e o método energia momento generalizado (GEMM+). A análise aerodinâmica é realizada via análise por Elementos Finitos para modelar escoamentos compressíveis viscosos e não viscosos em regimes transônicos e supersônicos. Um esquema característico baseado na separação da equação de momento (CBS, acrônimo do inglês) é utilizado para obter uma adequada integração temporal. No que concerne à otimização matemática, é utilizado um método baseado em gradientes, conhecido por programação quadrática sequencial (SQP, acrônimo do inglês), onde a avaliação as derivadas de Fréchet são levadas a cabo via diferenciação automática (AD, acrônimo do inglês). No que concerne aos resultados finais é realizada a otimização estrutural de forma de cascas modeladas como sólidos são apresentados, evidenciando um desempenho ótimo com respeito à energia de deformação interna. Os resultados de otimização aerodinâmica bidimensionais apresentam perfis aerodinâmicos ótimos com respeito à relação arrasto/sustentação para uma ampla gama de número de Mach, enquanto um resultado tridimensional é apresentado evidenciando a robustez e eficiência da implementação proposta. Pretendese estabelecer com este trabalho as bases para pesquisas em problemas de otimização aeroelástica. / Consolidation of the link among optimization problems in continuum mechanics, involving different fields, such as mathematical optimization, automatic differentiation, structural analysis, aerodynamic analysis, curves, surfaces and solids parameterization using Non Uniform Rational B-spline (NURBS), IsoGeometric Analysis (IGA), Finite Element Analysis (FEA) is looked for. Structural shape optimization of shell structures and aerodynamic shape optimization of immersed bodies in compressible flows are the main goals of this work. Concerning structural analysis, the so-called IsoGeometric analysis is employed. An accurate and comprehensive corotational kinematic based on the exact polar decomposition is developed in order to study highly nonlinear static and dynamic problems. Static analysis is carried out with Newton-Raphson and Generalized Displacement Control Method, while dynamic analysis is carried out with Generalized- (G) and Generalized Energy-Momentum Method (GEMM+). Aerodynamic analysis is carried out via Finite Element Analysis (FEA) in order to solve compressible flows in transonic and supersonic regimes. A Characteristic Based Split (CBS) method is employed to obtain an accurate time integration, which is based on the splitting of the momentum equation. Concerning mathematical optimization, the so-called Sequential Quadratic Programming (SQP) is employed, which is a gradient-based method, where the Fréchet derivatives are evaluated using Automatic Differentiation (AD). Final results consisting in structural optimization shown an optimal behaviour with respect to internal strain energy. While, results concerning aerodynamic bi-dimensional shape optimization exhibit a optimal behaviour with respect drag/lift ratio, for a large range of Mach number, and a simple result for tri-dimensional case is presented in order to show the efficiency and robustness of the implementation. Bases for future research in aeroelastic optimization problems are established in this work.

O presente trabalho tem por objetivo desenvolver uma formulação numérica baseada em Análise Isogeométrica para o estudo de escoamentos incompressíveis isotérmicos de fluidos newtonianos. Com o emprego desta metodologia, os procedimentos de pré-processamento e análise são unificados, melhorando as condições de continuidade das funções de base empregadas tanto na discretização espacial do problema como na aproximação das variáveis do sistema de equações. O sistema de equações fundamentais do escoamento é formado pelas equações de Navier-Stokes e pela equação de conservação de massa, descrita segundo a hipótese de pseudo-compressibilidade, além de uma equação constitutiva para fluidos viscosos de acordo com a hipótese de Stokes. Para problemas com escoamentos turbulentos emprega-se a Simulação de Grandes Escalas - LES (Large Eddy Simulation), na qual o modelo clássico de Smagorinsky é utilizado para a representação das escalas inferiores à resolução da malha. O esquema explícito de dois passos de Taylor-Galerkin é aplicado no contexto da Análise Isogeométrica para a discretização das equações governantes, sendo que a discretização espacial é realizada empregando-se funções NURBS (Non Uniform Rational Basis B-Splines). Essas funções base apresentam vantagens em relação às tradicionais funções utilizadas no MEF (Método dos Elementos Finitos), principalmente no que diz respeito à facilidade de obtenção de continuidade superior a C0 entre os elementos e representação precisa das geometrias. Propõe-se também o desenvolvimento de ferramentas de pré e pós-processamento baseadas na estrutura de dados da Análise Isogeométrica para a geração de malhas e visualização de resultados. Alguns problemas clássicos da Dinâmica dos Fluidos Computacional são analisados para a validação da metodologia apresentada. Os resultados apresentados demonstram boa aproximação da formulação em relação a dados de referência, além de maior versatilidade quanto à discretização espacial dos problemas em comparação com as tradicionais formulações baseadas em elementos finitos. / This work aims to develop a numerical formulation based on Isogeometric Analysis for the study of incompressible flows of Newtonian fluids under isothermal conditions. By using this methodology, pre-processing and analysis procedures are unified, improving the conditions of continuity of the basis functions utilized in the approximations of the equation variables and spatial discretization of the problem. The system of fundamental equations of the fluid flow is constituted by the Navier-Stokes equations and the mass conservation equation, which is described according to the pseudo-compressibility hypothesis. In addition, a constitutive equation for viscous fluids according to Stokes' hypothesis is also provided. Turbulent flows are analyzed using LES (Large Eddy Simulation), where the Smagorinsky’s model is adopted for sub-grid scales. The explicit two-step Taylor-Galerkin method is applied into the context of Isogeometric Analysis for the discretization of the flow equations, where spatial discretization is carried out taking into account Non Uniform Rational Basis B-Splines (NURBS) basis functions. These basis functions have advantages over traditional functions employed in the FEM (Finite Element Method). Particularly, it is easier to obtain continuity order higher than C0 between adjacent elements and geometry representation is more accurate. Pre and post-processing tools for mesh generation and results visualization are also proposed considering the data structure inherent to Isogeometric Analysis. Some classic problems of Computational Fluid Dynamics are analyzed in order to validate the proposed methodology. Results obtained here show that the present formulation has good approximation when compared with predictions obtained by reference authors. Moreover, Isogeometric Analysis is more versatile than traditional finite element formulations when spatial discretization procedures are considered.

CoordenaÃÃo de AperfeÃoamento de Pessoal de NÃvel Superior / The laminate structures are made using a set of layers of a composite material stacked in a particular sequence in order to obtain a good structural performance. Currently, the analysis of laminated structures is mainly performed using the Finite Element Method (FEM). However, this method is not able to accurately represent complex geometries. An alternative to the FEM is the Isogeometric Analysis (IGA). IGA uses in the numerical analysis the same functions used by Geometric Modeling in CAD systems, as B-splines and NURBS, allowing an exact representation of the geometry regardless of model discretization level. This study used the isogeometric formulation based on NURBS for performing geometric nonlinear analysis of laminated structures. This formulation was implemented in an academic finite element software. Using an appropriate formulation of the method and the Object Oriented Programming (OOP), it was possible to minimize the changes made in the structure of the program for implementing the Isogeometric Analysis, including in laminated structures problems. The verification of the implementation is carried out based on available examples in literature. Several examples of linear and non-linear analyzes of structures with isotropic and laminated composite material were performed and they obtained excellent results. In laminated structures project, it is necessary to determine the number of layers of composite material and the characteristics of each layer (material, thickness, and fiber orientation). Because there are numerous possible combinations, the standard procedure based on trial and error is not appropriate, requiring the use of optimization techniques. Bio-inspired optimization algorithms, such as Genetic Algorithms and Particle Swarm Optimization, perform well in combinatorial optimization problems. Considering these aspects, the present study was developed a hybrid algorithm, based on the Particle Swarm Optimization and Genetic Algorithm methods for optimization of laminated structures. Some variants of the proposed algorithm were compared considering several optimization examples. A calibration process of the algorithm parameters was conducted in order to avoid biased results. These variants were used in the optimization of laminated plates and shells. In the case of shells, the isogeometric analysis was used as a structural analysis tool. The results showed that the proposed optimization method presents comparable performance with the genetic algorithms in traditional laminates optimization, where the orientation of the fibers is limited to a few angles. Moreover, the proposed method outperforms genetic algorithm in the optimization of dispersed laminates. / As estruturas laminadas sÃo fabricadas utilizando um conjunto de camadas de material compÃsito empilhadas em uma sequÃncia determinada de forma a se obter um desempenho estrutural adequado. Atualmente, a anÃlise de estruturas laminadas à realizada principalmente utilizando o MÃtodo dos Elementos Finitos (MEF). Contudo, este mÃtodo nÃo à capaz de representar exatamente geometrias complexas. Uma alternativa ao MEF à a AnÃlise IsogeomÃtrica (AIG). A AIG utiliza na anÃlise numÃrica as mesmas funÃÃes utilizadas pelo sistemas CAD para Modelagem GeomÃtrica, como as B-Splines e NURBS, permitindo que a geometria dos modelos seja representada de forma exata para qualquer nÃvel de discretizaÃÃo adotado. O presente trabalho utilizou a formulaÃÃo isogeomÃtrica baseada em NURBS para realizar a anÃlise nÃo linear geomÃtrica de estruturas laminadas. Esta formulaÃÃo foi implementada em um software acadÃmico de anÃlise por elementos finitos. Utilizando uma formulaÃÃo apropriada do mÃtodo e o paradigma de ProgramaÃÃo Orientada a Objetos (POO), foi possÃvel minimizar as alteraÃÃes realizadas na estrutura deste programa para a implementaÃÃo da AnÃlise IsogeomÃtrica, inclusive em problemas de estruturas laminadas. A verificaÃÃo da implementaÃÃo foi realizada com base em exemplos disponÃveis na literatura. Exemplos de anÃlises lineares e nÃo-lineares de estruturas com material isotrÃpico e compÃsito laminado foram realizados, tendo obtido excelentes resultados. No projeto de estruturas laminadas à necessÃrio determinar o nÃmero de camadas de material compÃsito e as caracterÃsticas de cada camada (material, espessura e orientaÃÃo das fibras). Pelo fato de existirem um grande nÃmero de combinaÃÃes possÃveis, o procedimento padrÃo de tentativa e erro nÃo à apropriado, sendo necessÃrio a utilizaÃÃo de tÃcnicas de otimizaÃÃo. Algoritmos de otimizaÃÃo bio-inspirados, como Algoritmos GenÃticos e Nuvem de PartÃculas, apresentam bom desempenho em problemas de otimizaÃÃo combinatÃria. Considerando estes aspectos, no presente trabalho foi desenvolvido um algoritmo hÃbrido, baseado nos mÃtodos da Nuvem de PartÃculas e Algoritmo GenÃtico, para otimizaÃÃo de estruturas laminadas. Algumas variantes do algoritmo proposto foram comparadas considerando vÃrios exemplos de otimizaÃÃo. Um processo de calibraÃÃo dos parÃmetros numÃricos do algoritmo de otimizaÃÃo foi realizado, de modo a permitir uma comparaÃÃo isenta entre as variantes. Estas variantes foram utilizadas na otimizaÃÃo de placas e cascas laminadas. No caso de cascas, a anÃlise isogeomÃtrica foi utilizada como ferramenta de anÃlise estrutural. Os resultados obtidos mostraram que o mÃtodo de otimizaÃao proposto apresentou desempenho comparÃvel com Algoritmos GenÃticos na otimizaÃÃo de laminados tradicionais, onde a orientaÃÃo das fibras à limitada a poucos Ãngulos. Por outro lado, o mÃtodo proposto obteve desempenho superior ao Algoritmo GenÃtico na otimizaÃÃo de laminados dispersos.

Neste trabalho buscou-se consolidar aspectos referentes à otimização de problemas envolvidos na mecânica dos meios contínuos, envolvendo diferentes áreas do conhecimento, tais como: otimização matemática, diferenciação automática, análise estrutural, análise aerodinâmica, parametrização de curvas, superfícies e sólidos do tipo B-spline racionais não-uniformes (NURBS, acrônimo do inglês), análise IsoGeométrica (IGA, acrônimo do inglês) e análise por Elementos Finitos (FEA, acrônimo do inglês). Como objetivo final busca-se otimizar formas de cascas estruturais e formas de corpos aerodinâmicos imersos em escoamentos compressíveis. No que concerne à análise estrutural, esta é realizada via análise IsoGeométrica utilizando elementos sólidos para modelar cascas. Uma cinemática co-rotacional abrangente e precisa baseada na exata decomposição polar é desenvolvida, para lidar com problemas estáticos e dinâmicos altamente não lineares. Na análise estática foram implementados o método de Newton-Raphson e controle de deslocamentos generalizado, para problemas dinâmicos foram implementados o método -generalizado (G) e o método energia momento generalizado (GEMM+). A análise aerodinâmica é realizada via análise por Elementos Finitos para modelar escoamentos compressíveis viscosos e não viscosos em regimes transônicos e supersônicos. Um esquema característico baseado na separação da equação de momento (CBS, acrônimo do inglês) é utilizado para obter uma adequada integração temporal. No que concerne à otimização matemática, é utilizado um método baseado em gradientes, conhecido por programação quadrática sequencial (SQP, acrônimo do inglês), onde a avaliação as derivadas de Fréchet são levadas a cabo via diferenciação automática (AD, acrônimo do inglês). No que concerne aos resultados finais é realizada a otimização estrutural de forma de cascas modeladas como sólidos são apresentados, evidenciando um desempenho ótimo com respeito à energia de deformação interna. Os resultados de otimização aerodinâmica bidimensionais apresentam perfis aerodinâmicos ótimos com respeito à relação arrasto/sustentação para uma ampla gama de número de Mach, enquanto um resultado tridimensional é apresentado evidenciando a robustez e eficiência da implementação proposta. Pretendese estabelecer com este trabalho as bases para pesquisas em problemas de otimização aeroelástica. / Consolidation of the link among optimization problems in continuum mechanics, involving different fields, such as mathematical optimization, automatic differentiation, structural analysis, aerodynamic analysis, curves, surfaces and solids parameterization using Non Uniform Rational B-spline (NURBS), IsoGeometric Analysis (IGA), Finite Element Analysis (FEA) is looked for. Structural shape optimization of shell structures and aerodynamic shape optimization of immersed bodies in compressible flows are the main goals of this work. Concerning structural analysis, the so-called IsoGeometric analysis is employed. An accurate and comprehensive corotational kinematic based on the exact polar decomposition is developed in order to study highly nonlinear static and dynamic problems. Static analysis is carried out with Newton-Raphson and Generalized Displacement Control Method, while dynamic analysis is carried out with Generalized- (G) and Generalized Energy-Momentum Method (GEMM+). Aerodynamic analysis is carried out via Finite Element Analysis (FEA) in order to solve compressible flows in transonic and supersonic regimes. A Characteristic Based Split (CBS) method is employed to obtain an accurate time integration, which is based on the splitting of the momentum equation. Concerning mathematical optimization, the so-called Sequential Quadratic Programming (SQP) is employed, which is a gradient-based method, where the Fréchet derivatives are evaluated using Automatic Differentiation (AD). Final results consisting in structural optimization shown an optimal behaviour with respect to internal strain energy. While, results concerning aerodynamic bi-dimensional shape optimization exhibit a optimal behaviour with respect drag/lift ratio, for a large range of Mach number, and a simple result for tri-dimensional case is presented in order to show the efficiency and robustness of the implementation. Bases for future research in aeroelastic optimization problems are established in this work.

Avec l’arrivée de l’Analyse IsoGéométrique (IGA), le calcul de coque est devenu possible en utilisant la géométrie exacte pour des maillages grossiers. Pour cela, les polynômes de Lagrange sont remplacés pour l’interpolation par des fonctions NURBS (technologie la plus courante en conception assistée par ordinateur). De plus, ces fonctions possèdent une continuité supérieure ce qui offre une meilleure précision qu’un calcul éléments finis à nombre de degrés de liberté égal. L’IGA a déjà été développée pour les formulations coques. Elle n’a été cependant que très peu étudiée pour les modèles massifs coque. Pourtant, cette deuxième approche est très utilisée par l’ingénieur car elle permet de calculer des structures minces à l’aide d’éléments continus 3D, c’est-à-dire en faisant intervenir uniquement des inconnues en déplacements. La difficulté en calcul de coque est de faire face au verrouillage qui conduit à une forte dégradation de la convergence de la solution. Le cadre NURBS ne permet pas lui-même de résoudre ce problème. La meilleure efficacité de l’approximation NURBS ne peut donc être atteinte sans le développement de techniques particulières pour supprimer le verrouillage. C’est le but de cette thèse dans le cadre des éléments massifs coque. Le premier travail a consisté, sur un problème de poutre courbe, à étendre les méthodes sans verrouillage habituelles au contexte NURBS. Deux nouvelles stratégies ont alors été développées pour les NURBS : la première est basée sur une technique d’intégration réduite tandis que la seconde fait appel à une projection B-bar. Le formalisme général des méthodes B-bar semblant plus adapté, c’est celui-ci que nous avons développé ensuite pour les éléments massifs coque. Plus précisément, nous avons mis en place une formulation mixte de laquelle nous avons pu dériver la projection B-bar équivalente. Cette démarche constitue d’un point de vue théorique le résultat principal du travail : une méthode systématique pour construire une projection B-bar consistante est de passer par une formulation mixte. D’un point de vue mise en œuvre, l’idée principale pour traiter le verrouillage des éléments massifs coque a été de modifier l’interpolation de la moyenne dans l’épaisseur de la coque des composantes du tenseur des contraintes. Un contrôle de hourglass a aussi été ajouté pour stabiliser l’élément dans certaines situations. L’élément obtenu est de bonne qualité pour une interpolation de bas degrés et des maillages grossiers : la version quadratique semble plus précise que des éléments standards NURBS de degré 4. La méthode proposée conduit à une matrice de rigidité globale de petite taille mais pleine. Ce problème est inhérent aux NURBS. Il a pu être limité ici en utilisant une procédure de type moindres carrés locaux pour approcher la projection B-bar. Finalement, l’élément mixte a été étendu avec succès en non linéaire géométrique ce qui témoigne du potentiel de la méthode pour mener des simulations complexes. / With the introduction of IsoGeometric Analysis (IGA), the calculation of shell has become possible using the exact geometry for coarse meshes. In order to that, Lagrange polynomials are replaced by NURBS functions, the most commonly used technology in Computer-Aided Design, to perform the analysis. In addition, NURBS functions have a higher order of continuity, which leads to higher per-degree-of-freedom accuracy of the shell solution than with classical Finite Elements Analysis (FEA). IGA has now been widely applied in shell formulations. Nevertheless, it has still rarely been studied in the context of solid-shell models. This second shell approach is, however, very useful for engineers, since it enables to calculate thin structures using 3D solid elements, i.e. involving only displacements as degrees of freedom. The difficulty in shell analysis is to deal with locking which highly deteriorates the convergence of the solution. The NURBS framework does not enable to solve the problem directly. Then, to really benefit from NURBS in shells, specific strategies need to be implemented to answer the locking issue. This is the goal of the thesis in the context of solid-shell elements. The first work has consisted, on a curved beam problem, in extending the locking-free methods usually encountered in FEA to the NURBS context. The study resulted in the development of two new strategies for NURBS: the first one is based on a selective reduced integration technique and the second one makes use of a B-bar projection. The global formalism offered by the B-bar method appearing more suitable for NURBS, it has then been investigated for solid-shell elements. More precisely, a mixed formulation has first been elaborated from which, it has been possible to derive the equivalent B-bar projection. From a theoretical point of view, this strategy constitutes the most important result of this work: a systematic method to construct a consistent B-bar projection is to write a mixed formulation. With regards to the implementation, the main idea to treat locking of the solid-shell elements has been to modify the average of the strain and stress components across the thickness of the shell. Hourglass control has also been added to stabilize the element in particular situations. The resulting element is of good quality for low order approximations and coarse meshes: the quadratic version seems to be more accurate than basic NURBS elements of order 4. The proposed method leads to a global stiffness matrix of small size but full. This problem is inherent to NURBS functions. It has been limited here by using a local least squares procedure to approach the B-bar projection. Finally, the mixed element has been successfully extended to geometric non-linearity which reflects the ability of the methodology to be used in complex simulations.

Cette thèse présente un algorithme automatique pour la construction d’un modèle NURBS volumique à partir d’un modèle représenté par ses bords (maillages ou splines). Ce type de modèle est indispensable dans le cadre de l’analyse isogéométrique utilisant les NURBS comme fonctions de forme. Le point d’entrée de l’algorithme est une triangulation du bord du modèle. Après deux étapes de décomposition, le modèle est approché par un polycube. Ensuite un paramétrage surfacique entre le bord du modèle et celui du polycube est établi en calculant un paramétrage global aligné à un champ de direction interpolant les directions de courbure principales du modèle. Finalement, le paramétrage volumique est obtenu en se basant sur ce paramétrage surfacique. Dans le contexte des études paramétriques basées sur des paramètres de formes géométriques, cette méthode peut être appliquée aux techniques de réduction de modèles pour obtenir la même représentation pour des objets ayant différentes géométries mais la même topologie. / This thesis presents an effective method to automatically construct trivariate tensor-product spline models of complicated geometry and arbitrary topology. Our method takes as input a solid model defined by its triangulated boundary. Using cuboid decomposition, an initial polycube approximating the input boundary mesh is built. This polycube serves as the parametric domain of the tensor-product spline representation required for isogeometric analysis. The polycube's nodes and arcs decompose the input model locally into quadrangular patches, and globally into hexahedral domains. Using aligned global parameterization, the nodes are re-positioned and the arcs are re-routed across the surface in a way to achieve low overall patch distortion, and alignment to principal curvature directions and sharp features. The optimization process is based on one of the main contributions of this thesis: a novel way to design cross fields with topological (i.e., imposed singularities) and geometrical (i.e., imposed directions) constraints by solving only sparse linear systems. Based on the optimized polycube and parameterization, compatible B-spline boundary surfaces are reconstructed. Finally, the interior volumetric parameterization is computed using Coon's interpolation and the B-spline surfaces as boundary conditions. This method can be applied to reduced order modeling for parametric studies based on geometrical parameters. For models with the same topology but different geometries, this method allows to have the same representation: i.e., meshes (or parameterizations) with the same topology.

Optimierungsaufgaben unter Partiellen Differentialgleichungen (PDGLs) tauchen in verschiedensten Anwendungen der Wissenschaft und Technik auf. Wenn wir ein PDGL Problem formulieren, kann es aufgrund seiner Größe unmöglich werden, das Problem mit konventionellen Methoden zu lösen. Zusätzlich noch eine Optimierung auszuführen birgt zusätzliche Schwierigkeiten. In vielen Fällen können wir das PDGL Problem in einem kompakteren Format formulieren indem wir der zugrundeliegenden Kronecker-Produkt Struktur zwischen Raum- und Zeitdimension Aufmerksamkeit schenken. Wenn die PDGL zusätzlich mit Isogeometrischer Analysis diskretisiert wurde, können wir zusätlich eine Niedrig-Rang Approximation zwischen den einzelnen Raumdimensionen erzeugen. Diese Niedrig-Rang Approximation lässt uns die Systemmatrizen schnell und speicherschonend aufstellen. Das folgende PDGL-Problem lässt sich als Summe aus Kronecker-Produkten beschreiben, welche als eine Niedrig-Rang Tensortrain Formulierung interpretiert werden kann. Diese kann effizient im Niedrig-Rang Format gelöst werden. Wir illustrieren dies mit unterschiedlichen, anspruchsvollen Beispielproblemen.:Introduction Tensor Train Format Isogeometric Analysis PDE-constrained Optimization Bayesian Inverse Problems A low-rank tensor method for PDE-constrained optimization with Isogeometric Analysis A low-rank matrix equation method for solving PDE-constrained optimization problems A low-rank tensor method to reconstruct sparse initial states for PDEs with Isogeometric Analysis Theses and Summary Bibilography / Optimization problems governed by Partial Differential Equations (PDEs) arise in various applications of science and engineering. If we formulate a discretization of a PDE problem, it may become infeasible to treat the problem with conventional methods due to its size. Solving an optimization problem on top of the forward problem poses additional difficulties. Often, we can formulate the PDE problem in a more compact format by paying attention to the underlying Kronecker product structure between the space and time dimension of the discretization. When the PDE is discretized with Isogeometric Analysis we can additionally formulate a low-rank representation with Kronecker products between its individual spatial dimensions. This low-rank formulation gives rise to a fast and memory efficient assembly for the system matrices. The PDE problem represented as a sum of Kronecker products can then be interpreted as a low-rank tensor train formulation, which can be efficiently solved in a low-rank format. We illustrate this for several challenging PDE-constrained problems.:Introduction Tensor Train Format Isogeometric Analysis PDE-constrained Optimization Bayesian Inverse Problems A low-rank tensor method for PDE-constrained optimization with Isogeometric Analysis A low-rank matrix equation method for solving PDE-constrained optimization problems A low-rank tensor method to reconstruct sparse initial states for PDEs with Isogeometric Analysis Theses and Summary Bibilography

Isogeometric analysis is aimed to mitigate the gap between Computer-Aided Design (CAD) and analysis by using a unified geometric representation. Thanks to the exact geometry representation and high smoothness of adopted basis functions, isogeometric analysis demonstrated excellent mathematical properties and successfully addressed a variety of problems. In particular, it allows to solve higher order Partial Differential Equations (PDEs) directly omitting the usage of mixed approaches. Unfortunately, complex CAD geometries are often constituted by multiple Non-Uniform Rational B-Splines (NURBS) patches and cannot be directly applied for finite element analysis.parIn this work, we presents a dual mortaring framework to couple adjacent patches for higher order PDEs. The development of this formulation is initiated over the simplest 4th order problem-biharmonic problem. In order to speed up the construction and preserve the sparsity of the coupled problem, we derive a dual mortar compatible C1 constraint and utilize the Bezier dual basis to discretize the Lagrange multipler spaces. We prove that this approach leads to a well-posed discrete problem and specify requirements to achieve optimal convergence. After identifying the cause of sub-optimality of Bezier dual basis, we develop an enrichment procedure to endow Bezier dual basis with adequate polynomial reproduction ability. The enrichment process is quadrature-free and independent of the mesh size. Hence, there is no need to take care of the conditioning. In addition, the built-in vertex modification yields compatible basis functions for multi-patch coupling.To extend the dual mortar approach to couple Kirchhoff-Love shell, we develop a dual mortar compatible constraint for Kirchhoff-Love shell based on the Rodrigues' rotation formula. This constraint provides a unified formulation for both smooth couplings and kinks. The enriched Bezier dual basis preserves the sparsity of the coupled Kirchhoff-Love shell formulation and yields accurate results for several benchmark problems.Like the dual mortaring formulation, locking problem can also be derived from the mixed formulation. Hence, we explore the potential of Bezier dual basis in alleviating transverse shear locking in Timoshenko beams and volumetric locking in nearly compressible linear elasticity. Interpreting the well-known B projection in two different ways we develop two formulations for locking problems in beams and nearly incompressible elastic solids. One formulation leads to a sparse symmetric symmetric system and the other leads to a sparse non-symmetric system.

In this thesis, a robust, reliable and efficient isogeometric analysis framework is presented that allows for an adaptive spatial discretization of non-linear and time-dependent multi-field problems. In detail, B\'ezier extraction of truncated hierarchical B-splines is proposed that allows for a strict element viewpoint, and in this way, for the application of standard finite element procedures. Furthermore, local mesh refinement and coarsening strategies are introduced to generate graded meshes that meet given minimum quality requirements. The different strategies are classified in two groups and compared in the adaptive isogeometric analysis of two- and three-dimensional, singular and non-singular problems of elasticity and the Poisson equation. Since a large class of boundary value problems is non-linear or time-dependent in nature and requires incremental solution schemes, projection and transfer operators are needed to transfer all state variables to the new locally refined or coarsened mesh. For field variables, two novel projection methods are proposed and compared to existing global and semi-local versions. For internal variables, two different transfer operators are discussed and compared in numerical examples. The developed analysis framework is than combined with the phase-field method. Numerous phase-field models are discussed including the simulation of structural evolution processes to verify the stability and efficiency of the whole adaptive framework and to compare the projection and transfer operators for the state variables. Furthermore, the phase-field method is used to develop an unified modelling approach for weak and strong discontinuities in solid mechanics as they arise in the numerical analysis of heterogeneous materials due to rapidly changing mechanical properties at material interfaces or due to propagation of cracks if a specific failure load is exceeded. To avoid the time consuming mesh generation, a diffuse representation of the material interface is proposed by introducing a static phase-field. The material in the resulting transition region is recomputed by a homogenization of the adjacent material parameters. The extension of this approach by a phase-field model for crack propagation that also accounts for interface failure allows for the computation of brittle fracture in heterogeneous materials using non-conforming meshes.

Au cours des processus de conception,l’optimisation de formes apporte aux industriels dessolutions pour l’amélioration des performances desproduits. En particulier, les structures minces quiconstituent environ 70% d’un véhicule, sont une préoccupationdans l’industrie automobile. La plupartdes méthodes d’optimisation pour ces structures surfaciquesprésentent certaines limites et nécessitent desexpertises à chaque niveau de la procédure d’optimisation.L’objectif de cette thèse est de proposer une nouvellestratégie d’optimisation de formes pour les coquesminces. L’approche présentée consiste à exploiter leséquations de coques du modèle de Koiter en se basantsur une analyse isogéométrique. Cette méthode permetde réaliser des simulations sur la géométrie exacteen définissant la forme à l’aide de patchs CAO. Lesvariables d’optimisation choisies sont alors les pointsde contrôle permettant de piloter leur forme. La définitiondes patchs permet également de dégager ungradient de forme pour l’optimisation à l’aide d’uneméthode adjointe.Cette méthode a été appliquée pour des critères mécaniquesissus des bureaux d’études Renault. Des résultatsd’optimisation pour un critère de compliance sontprésentés. La définition et l’implémentation de critèresvibro-acoustiques sont discutés à la fin de cette thèse.Les résultats obtenus témoignent de l’intérêt de la méthode.Toutefois, de nombreux développements serontnécessaires avant d’être en mesure de l’appliquer dansl’industrie. / During the design process, optimizationof shapes offers manufacturers solutions for improvingproducts performances. In particular, thin shellstructures that represent about 70 % of a vehicle, area concern in the automotive industry. Most optimizationmethods for surface structures have limitationsand require expertise at every level of the optimizationprocedure.The aim of this thesis is to propose a new strategyfor the shape optimization of thin shell structures.The approach presented rely on using the Koiter’sshell model based on an isogeometric analysis. Thismethod allows for simulations on the exact geometryby defining the shape using CAD patches. Selectedoptimization variables are the control points used tocontrol the shape of the CAD patches. Variations ofthese points allows to scan a wide design space withfew parameters. The definition of patchs also enablesto find a gradient with respect to the shape for theoptimization by using the adjoint state method.This method was applied to mechanical criteria fromthe Renault design offices. Optimization results for acompliance criterion are presented. The definition andimplementation of vibro-acoustic criteria are discussedat the end of this thesis. The results demonstratethe interest of the method. However, many developmentswill be needed before being able to apply it inthe industry.