Eficácia de modelos micromecânicos na modelagem de estruturas de concreto de ultra alto desempenho reforçado com fibras de aço

Abstract

Ultra-High-Performance Fiber Reinforced Concrete (UHPFRC) is a material that has exceptional compressive strength, durability and ductility. Therefore, it presents to be an alternative in the improvement of structural designs that have specific conditions of aggressiveness due to deleterious agents or high structural loads. UHPFRC is currently used predominantly in the rehabilitation and repair of structures. The large-scale application of UHPFRC faces barriers mainly due to the high cost and the lack of normative parameters to design requirements. The characterization of the mechanical behavior of UHPFRC depends primarily on the interaction between the phases that compose the mixture, that induce a requirement of a multiscale analysis to define the overall performance of the material. The present work aims to evaluate the accuracy of the principal micromechanical homogenization models to obtain the mechanical parameters of UHPFRC. This is done by comparing the results of the micromechanical approach with experimental results obtained from the literature. These models are implemented as UMAT user subroutines in ABAQUS® software, which is then used to performs evaluate of the mechanical behavior of a beam by using the finite element method. The homogenization models adopted in the present work are: Mori-Tanaka, Self Consistent, Voigt (upper limit) and Reuss (lower limit). The results obtained by Mori-Tanaka and Auto Consistent models indicate a difference for the experimental values of the Young modulus between 2% and 3%, while for the Poisson's ratio this difference is around 2%. On the other hand, the Voigt and Reuss models presented to be inaccurate to determine the Young modulus and should not be used to determine the Poisson's ratio due to the assumptions adopted in its formulation. The mechanical behavior of the beam denotes high stiffness, producing small deformations for high loads. furthermore, the inner stresses have been shown to vary linearly at a steel fiber volumetric fraction variation of 2% for the Mori-Tanaka model. Finally, the micromechanical approach can be applied to UHPFRC with a satisfactory degree of accuracy and can be used as a complement to lower costs in an extensive experimental dosing process.