Análise do efeito de escala em materiais quase-frágeis utilizando a teoria peridinâmica

Abstract

Laboratory experiments, in general, represent the reality about the behavior of a given component, however in the case of very large structures, the physical size delimits the space available for carrying out the tests and, in this way, small scale models are used. The results of these experiments can give indications of the expected behavior of the component, as long as the results obtained in the laboratory are not influenced by the size difference. However, this condition is not met for a class of materials, called quasi-brittle. These materials, such as polymers reinforced with fibers, rocks and mainly concrete, have specific characteristics and are used in a large number of engineering applications. However, depending on the scale of analysis of the structures formed by these materials, their behavior is influenced by their size. This effect is called the structural scale effect. Using expanded polystyrene as study material, beams with three different dimensions, proportional to each other, subjected to a three-point bending test are analyzed numerically using peridynamics (PD). The peridynamic theory is a particular formulation of the discrete element method. It is a non-local theory that takes into account the forces between material points in a given neighborhood, defined by a region called the horizon. The results are generally compared with experimental studies available in the literature. The scale effect will be analyzed on the mechanical quantities of the flexural stress, critical rotation and fracture energy of the material, using the fractal approach described by Carpinteri. In addition, the simulated samples were monitored using the acoustic emission technique and the scale effect was also analyzed from this point of view. The results show that the PD is an adequate tool for the simulation of quasi-brittle materials and that the scale effect captured is consistent with what was expected by the theory and predicted by the experimental data.