Les assemblages sont des éléments critiques pour les structures industrielles. De fortes non-linéarités de type contact frottant, ainsi que des précharges mal maîtrisées rendent complexe tout dimensionnement précis. Présents en très grand nombre sur les structures industrielles (quelques millions pour un A380), cela implique de rafiner les modèles localement et donc de gérer des problèmes numé-riques de très grandes tailles. Les nombreuses interfaces de contact frottant sont des sources de difficultés de convergence pour les simulations numériques. Il est donc nécessaire de faire appel à des méthodes robustes. Il s’agit d’utiliser des méthodes itératives de décomposition de domaine, permettant de gérer des modèles numériques extrêmement grands, couplées à des techniques adaptées afin de prendre en compte les non-linéarités de contact aux interfaces entre sous-domaines. Ces méthodes de décomposition de domaine restent encore très peu utilisées dans un cadre industriel. Des développements internes aux codes éléments finis sont souvent nécessaires et freinent ce transfert du monde académique au monde industriel.Nous proposons, dans ces travaux de thèse, une mise-en-oeuvre non intrusive de ces méthodes de décomposition de domaine : c’est-à-dire sans développement au sein du code source. En particulier, nous nous intéressons à la méthode Latin dont la philosophie est particulièrement adaptée aux problèmes non linéaires. La structure est décomposée en sous-domaines reliés entre eux au travers d’interfaces. Avec la méthode Latin, les non-linéarités sont résolues séparément des aspects linéaires. La résolution est basée sur un schéma itératif à deux directions de recherche qui font dialoguer les problèmes linéaires globaux etles problèmes locaux non linéaires.Au cours de ces années de thèse, nous avons développé un outil totalement non intrusif sous Code_Aster permettant de résoudre par une technique de décomposition de domaine mixte des problèmes d’assemblage. Les difficultés posées par le caractère mixte de la méthode Latin sont résolues par l’introduction d’une direction de recherche non locale. Des conditions de Robin sur les interfaces des sous-domaines sont alors prises en compte simplement sans modifier les sources de Code_Aster. Nous avons proposé une réécriture algébrique de l’approche multi-échelle assurant l’extensibilité de la méthode. Nous nous sommes aussi intéressés à coupler la méthode Latin en décomposition de domaine à un algorithme de Krylov. Appliqué uniquement à un problème sous-structuré avec interfaces parfaites, ce couplage permet d’accélérer la convergence. Des structures préchargées avec de nombreuses interfaces de contact frottant ont été traitées. Des simulations qui n’auraient pu être menées par un calcul direct sous Code_Aster ont été réalisées via cette stratégie de décomposition de domaine non intrusive. / Abstract : Assemblies are critical elements for industrial structures. Strong non-linearities such as frictional contact, as well as poorly controlled preloads make complex all accurate sizing. Present in large numbers on industrial structures (a few million for an A380), this involves managing numerical problems of very large size. The numerous interfaces of frictional contact are sources of difficulties of convergence for the numerical simulations. It is therefore necessary to use robust but also reliable methods. The use of iterative methods based on domain decomposition allows to manage extremely large numerical models. This needs to be coupled with adaptedtechniques in order to take into account the nonlinearities of contact at the interfaces between subdomains. These methods of domain decomposition are still scarcely used in industries. Internal developments in finite element codes are often necessary, and thus restrain this transfer from the academic world to the industrial world.In this thesis, we propose a non-intrusive implementation of these methods of domain decomposition : that is, without development within the source code. In particular, we are interested in the Latin method whose philosophy is particularly adapted to nonlinear problems. It consists in decomposing the structure into sub-domains that are connected through interfaces. With the Latin method the non-linearities are solved separately from the linear differential aspects. Then the resolution is based on an iterative scheme with two search directions that make the global linear problems and the nonlinear local problems dialogue.During this thesis, a totally non-intrusive tool was developed in Code_Aster to solve assembly problems by a mixed domain decomposition technique. The difficulties posed by the mixed aspect of the Latin method are solved by the introduction of a non-local search direction. Robin conditions on the subdomain interfaces are taken into account simply without modifying the sources of Code_Aster. We proposed an algebraic rewriting of the multi-scale approach ensuring the extensibility of the method. We were also interested in coupling the Latin method in domain decomposition to a Krylov algorithm. Applied only to a substructured problem with perfect interfaces, this coupling accelerates the convergence. Preloaded structures with numerous contact interfaces have been processed. Simulations that could not be carried out by a direct computationwith Code_Aster were performed via this non-intrusive domain decomposition strategy.

The experimental identification of mechanical properties is crucial in mechanics for understanding material behavior and for the development of numerical models. Classical identification procedures employ standard shaped specimens, assume that the mechanical fields in the object are homogeneous, and recover global properties. Thus, multiple tests are required for full characterization of a heterogeneous object, leading to a time consuming and costly process. The development of non-contact, full-field measurement techniques from which complex kinematic fields can be recorded has opened the door to a new way of thinking. From the identification point of view, suitable methods can be used to process these complex kinematic fields in order to recover multiple spatially varying parameters through one test or a few tests. The requirement is the development of identification techniques that can process these complex experimental data. This thesis introduces a novel identification technique called the constitutive compatibility method. The key idea is to define stresses as compatible with the observed kinematic field through the chosen class of constitutive equation, making possible the uncoupling of the identification of stress from the identification of the material parameters. This uncoupling leads to parametrized solutions in cases where 5 the solution is non-unique (due to unknown traction boundary conditions) as demonstrated on 2D numerical examples. First the theory is outlined and the method is demonstrated in 2D applications. Second, the method is implemented within a domain decomposition framework in order to reduce the cost for processing very large problems. Finally, it is extended to 3D numerical examples. Promising results are shown for 2D and 3D problems

No description available.

Unstructured simplicial mesh is an integral and critical part of many computational electromagnetics methods (CEM) such as the finite element method (FEM) and the boundary element method (BEM). Mesh quality and robustness have direct impact on the success of these CEM methods. A combined quadtree/Delunay 2D mesh generator, based on the early work of Schroeder (1991, PhD), is presented. The method produces a triangulation that approximates the original geometric model but is also topologically consistent. The key advantages of the method are: (a) its robustness, (b) ability to create a-priori graded meshes, and (c) its guaranteed mesh quality. The method starts by recursively refining the grid and using a 2:1 balanced quadtree data structure to index each cell. Once the quadtree grid is refined at a user-defined level associated with each geometrical model topological entity, each cell in the grid is successively triangulated using the Delaunay method. Finally, the method handles some modeling errors by merging vertices and allowing overlapped faces.

<div>Many Engineering structures, including electronic component assemblies, are inherently multi-scale in nature. These structures often experience complex local nonlinear behavior such as plasticity, damage or fracture. These local behaviors eventually lead to the failure at the macro length scale. Connecting the behavior across the length scales to develop an understanding of the failure mechanism is important for developing reliable products.</div><div><br></div><div>To solve multi-scale problems in which the critical region is much smaller than the entire structure, an iterative solution approach based on domain decomposition techniques is proposed. Two independent models are constructed to model the global and local substructures respectively. The unbalanced force at the interface is iteratively reduced to ensure force equilibrium of the overall structure in the final solution. The approach is non-intrusive since only nodal values on the interface are transferred between the global and local models. Solution acceleration using SR1 and BFGS updates is also demonstrated. Equally importantly, the two updates are applied in a non-intrusive manner, meaning that the technique is implemented without needing access to the codes using which the sub-domains are analyzed. Code- and mesh-agnostic solutions for problems with local nonlinear material behavior or local crack growth are demonstrated. Analysis in which the global and local models are solved using two different commercial codes is also demonstrated.</div><div><br></div><div>Engineering analysis using numerical models are helpful in providing insight into the connection between the structure, loading history, behavior and failure. Specifically, Isogeometric analysis (IGA) is advantageous for engineering problems with evolving geometry compared to the traditional finite element method (FEM). IGA carries out analysis by building behavioral approximations isoparametrically on the geometrical model (commonly NURBS) and is thus a promising approach to integrating Computer-Aided Design (CAD) with Computer-Aided Engineering (CAE).</div><div><br></div><div>In enriched isogeometric Analysis (EIGA), the solution is enriched with known behavior on lower dimensional geometrical features such as crack tips or interfaces. In the present research, enriched field approximation techniques are developed for the application of boundary conditions, coupling patches with non-matching discretizations and for modeling singular stresses in the structure.</div><div><br></div><div>The first problem solution discussed is to apply Dirichlet and Neumann boundary conditions on boundary representation (B-rep) CAD models immersed in an underlying domain of regular grid points. The boundary conditions are applied on the degrees of freedom of the lower dimensional B-rep part directly. The solution approach for the immersed analysis uses signed algebraic level sets constructed from the B-rep surfaces to blend the enriched</div><div>field with the underlying field. The algebraic level sets provide a surrogate for distance, are non-iteratively (or algebraically) computed and allow implicit Boolean compositions.</div><div><br></div><div>The methodology is also applied to couple solution approximations of decomposed patches by smoothly blending incompatible geometries to an arbitrary degree of smoothness. A parametrically described frame or interface is introduced to “stitch” the adjacent patches. A hierarchical blending procedure is then developed to stitch multiple unstructured patches including those with T-junctions or extraordinary vertices.</div><div><br></div><div>Finally, using the EIGA technique, a computational method for analyzing general multimaterial sharp corners that enables accurate estimations of the generalized stress intensity factors is proposed. Explicitly modeled geometries of material junctions, crack tips and deboned interfaces are isogeometrically and hierarchically enriched to construct approximations with the known local behavior. specifically, a vertex enrichment is used to approximate the asymptotic field near the re-entrant corner or crack tip, Heaviside function is used to approximate the discontinuous crack face and the parametric smooth stitching technique is used to approximate the behavior across material interface. The developed method allows direct extraction of generalized stress intensity factors without needing a posteriori evaluation of path independent integrals for decisions on crack propagation. The numerical implementation is validated through analysis of a bi-material corner, interface crack and growth of an inclined crack in a homogeneous solid. The developed procedure demonstrates rapid convergence to the solution stress intensity factors with relatively fewer degrees of freedom, even with uniformly coarse discretizations.</div>

The Finite Element Tearing and Interconnecting (FETI) and its variants are probably the most celebrated domain decomposition algorithms for partial differential equation (PDE) scientific computations. In electromagnetics, such methods have advanced research frontiers by enabling the full-wave analysis and design of finite phased array antennas, metamaterials, and other multiscale structures. Recently, closer scrutiny of these methods have revealed robustness and numerical scalability problems that prevent the most memory and time efficient variants of FETI from gaining widespread acceptance. This work introduces a new class of FETI methods and preconditioners that lead to exponential iterative convergence for a wide class of problems, are robust and numerically scalable. First, a two Lagrange multiplier (LM) variant of FETI with impedance transmission conditions, the FETI-2λ, is introduced to facilitate the symmetric treatment of non-conforming grids while avoiding matrix singularites that occur at the interior resonance frequencies of the domains. A thorough investigation on the approximability and stability of the Lagrange multiplier discrete space is carried over to identify the correct LM space basis. The resulting method, although accurate and flexible, exhibits unreliable iterative convergence. To accelerate the iterative convergence, the Locally Exact Algebraic Preconditioner (LEAP), which is responsible for improving the information transfer between neighboring domains is introduced. The LEAP was conceived by carefully studying the properties of the Dirichlet-to-Neumann (DtN) map that is involved in the sub-structuring process of FETI. LEAP proceeds in a hierarchical way and directly factorizes the signular and near-singular interactions of the DtN map that arise from domain-face, domain-edge and domain-vertex interactions. For problems with small number of domains LEAP results in scalable implementations with respect to the discretization. On problems with large domain numbers, the numerical scalability can only be obtained through ``global'' preconditioners that directly convey information to remotely separated domains at every DDM iteration. The proposed ``global" preconditiong stage is based on the new Multigrid FETI (MG-FETI) method. This method provides a coarse grid correction mechanism defined in the dual space. Macro-basis functions, that satisfy thecurl-curl equation on each interface are constructed to reduce the size of the coarse problem, while maintaining a good approximation of the characteristic field modes. Numerical results showcase the performance of the proposed method on one-way, 2D and 3D decomposed problems, with structured and unstructured partitioning, conforming and non-conforming interface triangulations. Finally, challenging, real life computational examples showcase the true potential of the method.

No description available.

No description available.

No description available.

No description available.