Fontes sísmicas ao longo da fronteira de placas tectónicas entre os Açores e a Argélia: um modelo sismotectónico

Portugal apresenta uma actividade sísmica que resulta em grande parte da sua proximidade à fronteira entre as placas tectónicas Euro-asiática (EA) e Nubia (NU), numa faixa que se estende desde Gibraltar até ao arquipélago dos Açores. A fractura litosférica contida nessa faixa é habitualmente designada por Fractura Açores-Gibraltar. A forte interacção entre os dois blocos, manifesta-se por um aumento da sismicidade nesta faixa, fortemente influenciada pela interacção entre os dois blocos tectónicos. No prolongamento para Ocidente deste acidente atinge-se a Crista Média Atlântica (CMA) num ponto localizado a noroeste do arquipélago dos Açores, e que constitui a fronteira entre a placa Americana (AM) e as placas EA e NU. Este ponto também é conhecido por Junção Tripla dos Açores. A interacção entre os três limites de placas confere à região dos Açores a actividade sísmica que se lhe conhece, uma das mais significativas no contexto nacional. Toda esta zona, devido ao potencial e efectivo risco sísmico testemunhado pelos eventos sísmicos recentes e pelos grandes terremotos historicamente documentados, é alvo de um elevado esforço que nos últimos anos resultou, de forma integrada, em avanços como: 1) melhoria da capacidade de observação do fenómeno sísmico –a existência dos meios mínimos de monitorização sísmica é hoje uma realidade que reconhecemos, salientando contudo, o muito que há a fazer em domínios como: a instalação de estações sísmicas submarinas (OBS) capazes de suprir as lacunas verificadas; a compatibilização de dados e o livre acesso aos mesmos; a criação de uma rede de acelerómetros capaz de registar os movimentos fortes; 2) aumento da capacidade de investigação – o recrutamento de novos investigadores através de projectos multi-disciplinares em domínios como a sismicidade, fonte sísmica e mecanismos focais, geomagnetismo, gravimetria, geodesia e análise estrutural....

A finite-strain solid–shell using local Löwdin frames and least-squares strains

A finite-strain solid–shell element is proposed. It is based on least-squares in-plane assumed strains, assumed natural transverse shear and normal strains. The singular value decomposition (SVD) is used to define local (integration-point) orthogonal frames-of-reference solely from the Jacobian matrix. The complete finite-strain formulation is derived and tested. Assumed strains obtained from least-squares fitting are an alternative to the enhanced-assumed-strain (EAS) formulations and, in contrast with these, the result is an element satisfying the Patch test. There are no additional degrees-of-freedom, as it is the case with the enhanced-assumed-strain case, even by means of static condensation. Least-squares fitting produces invariant finite strain elements which are shear-locking free and amenable to be incorporated in large-scale codes. With that goal, we use automatically generated code produced by AceGen and Mathematica. All benchmarks show excellent results, similar to the best available shell and hybrid solid elements with significantly lower computational cost.

Isogeometric analysis of large-deformation thin shells using RHT-splines for multiple-patch coupling

We present an isogeometric thin shell formulation for multi-patches based on rational splines over hierarchical T-meshes (RHT-splines). Nitsche’s method is employed to efficiently couple the patches. The RHT-splines have the advantages of allowing a computationally feasible local refine- ment, are free from linear independence, possess high order continuity and satisfy the partition of unity and non-negativity, properties. In addition, C 1 continuity of the RHT-splines obviates to use of rotational degrees of freedom. The good performance of the present method is demonstrated by a number of numerical examples.

Phase-field analysis of finite-strain plates and shells including element subdivision

With the theme of fracture of finite-strain plates and shells based on a phase-field model of crack regularization, we introduce a new staggered algorithm for elastic and elasto-plastic materials. To account for correct fracture behavior in bending, two independent phase-fields are used, corresponding to the lower and upper faces of the shell. This is shown to provide a realistic behavior in bending-dominated problems, here illustrated in classical beam and plate problems. Finite strain behavior for both elastic and elasto-plastic constitutive laws is made compatible with the phase-field model by use of a consistent updated-Lagrangian algorithm. To guarantee sufficient resolution in the definition of the crack paths, a local remeshing algorithm based on the phase- field values at the lower and upper shell faces is introduced. In this local remeshing algorithm, two stages are used: edge-based element subdivision and node repositioning. Five representative numerical examples are shown, consisting of a bi-clamped beam, two versions of a square plate, the Keesecker pressurized cylinder problem, the Hexcan problem and the Muscat-Fenech and Atkins plate. All problems were successfully solved and the proposed solution was found to be robust and efficient.

A finite-strain solid–shell using local Löwdin frames and least-squares strains

A finite-strain solid–shell element is proposed. It is based on least-squares in-plane assumed strains, assumed natural transverse shear and normal strains. The singular value decomposition (SVD) is used to define local (integration-point) orthogonal frames-of- reference solely from the Jacobian matrix. The complete finite-strain formulation is derived and tested. Assumed strains obtained from least-squares fitting are an alternative to the enhanced-assumed-strain (EAS) formulations and, in contrast with these, the result is an element satisfying the Patch test. There are no additional degrees-of-freedom, as it is the case with the enhanced- assumed-strain case, even by means of static condensation. Least-squares fitting produces invariant finite strain elements which are shear-locking free and amenable to be incorporated in large-scale codes. With that goal, we use automatically generated code produced by AceGen and Mathematica. All benchmarks show excellent results, similar to the best available shell and hybrid solid elements with significantly lower computational cost.