Analisi teorica, numerica e sperimentale dei processi di grandi deformazioni nei materiali duttili

A broad area of the troublesome bound to the mechanical design of structures is represented by the ability of including in the engineering analysis all the non linearity effects expressed by the ductile materials, often disregarded to obtain a clearer and simpler description of the phenomena. The frequently used “isotropic approach” and the choice of focusing only on the linear-elastic phase of the materials are the most important reasons behind the success of these design choices. Years after years, following the constant improvements in structures, in the material’s geometries and innovative shapes the designers are compelled to take into account all the situations in which the materials behave in a non-linear fashion, especially when they’re loaded with critical stress states. Baseline or induced anisotropic phenomena, shear bandind, cinematic and isotropic hardening.,strain rate dependency, thermal effects, the loosing of the original mechanical properties, are just a part of a bigger list of factors that affect the deformation properties of a body, leading to many complications in the theoretical analysis of solids and structures. To solve this problematic, the mathematical models used to include into the calculation the above cited effects (obtaining important results in terms of accuracy and reality of the description) lead to a much more complicate analytical formulation and a large increasing of all the experimental procedures connected with the parameters identification. Even if there’s an undeniable positive effect in the theoretical description of the material behavior and the phenomena investigated, on the other hand an important increase of all the numerical and experimental costs and time consumption required is evident. Between all the possible non linearity effects in this thesis we focused only on the damage process and the anisotropic elasto-plasticity that, with no doubts, have a fundamental role to correctly modelling the ductile material behavior utilized in the gas/oil onshore and offshore pipelines. After a short historical review of the most important solutions that were presented in the specialized literature, two different macro-mechanical damage model are analyzed in details: the the Lemaitre and Wierzbicki-Xue models. This choice was made to evaluate from different point of view the differences in the determination procedures of the material features along the deformation pattern. Both the models were implemented and tested through two customized damage user-subroutines for the FE software Msc.Marc. The results obtained from the numerical simulations were checked using an optical method to evaluate the real displacement maps. The strain/ displacement maps were recorded in situ using a Digital Image Correlation technique on some particular aluminum specimens designed to highlight the inconsistencies of the two selected solutions. While the first part of the work only pertain to the ductile damage procedures, the second part of the thesis, is focused on the large strain formulation for elasto-plastic anisotropic solids: the theoretical basis and the numerical algorithms are presented here and used to describe the constitutive relations for both the isotropic and anisotropic cases. This background will be an essential part to essential part to move forward to "small deformation" case and introduce the Bathe-Montans plasticity model for the anisotropic solids. The Bathe-Montans model was implemented writing an UMAT subroutine for the FE code Abaqus Standard and numerically tested using different material and geometry/meshes. To validate the results, some numerical comparisons will be presented and moreover some numerical-experimental results obtained simulating particular classes of steel used in the onshore and offshore oil/gas pipeline will be shown to demonstrate the description ability and the accuracy of the model proposed.