Affiliation:
1. Colegio de Ciencias e Ingenierías, Universidad San Francisco de Quito, Casilla Postal: 17-12-841, Quito, Ecuador
Abstract
Concrete is a compound material where aggregates are randomly placed within the cement paste. To describe the behavior of concrete structures at the ultimate, it is necessary to use nonlinear finite element models, which for shear and torsion problems do not always give satisfactory results. The current study aims at improving the modeling of concrete at the meso-level, which eventually can result in an improved assessment of existing structures. Concrete as a heterogeneous material is modeled consisting of hydrated cement paste and aggregates. The stress–strain curves of the hydrated cement paste and aggregates are described with results from the literature. A three-dimensional (3D) finite element model was developed to determine the influence of individual phases on the inelastic stress–strain distribution of concrete structures. A random distribution and morphology of the cement and aggregate fractions are achieved by using DREAM.3D. Two affordable computational dual-phase representative volume elements (RVEs) are imported to ABAQUS to be studied in compression and tension. The virtual specimens (concrete mesh) subjected to continuous monotonic strain loading conditions were constrained with 3D boundary conditions. Results demonstrate differences in stress–strain mechanical behavior in both compression and tension test simulations. A strong dependency of flow stress and plastic strain on phase type, aggregate (andesite) size, shape and distribution upon the composite local response are clearly observed. It is noted that the resistance to flow is higher in concrete meshes composed of finer and homogeneous aggregate particles because the Misses stresses and effective plastic strains are better distributed. This study shows that at the meso-level, concrete can be modeled consisting of aggregates and hydrated cement paste.
Publisher
World Scientific Pub Co Pte Lt
Subject
Computer Science Applications,Modeling and Simulation
Cited by
10 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献