Search

A pullout analytical model is introduced to simulate the Double Edge Wedge Splitting (DEWS) test to obtain the tensile post-peak response of steel fiber reinforced concretes. One of the main barriers to the widespread utilization of FRC as a structural system is the anisotropy of the material. This augments the uncertainty of the material behavior (fibers random distribution and orientation in the bulk concrete), increases the dispersion of the test results, and requires the performance of characterization tests to check the residual tensile stresses at specific crack opening intervals. The characterization tests proposed in the literature (e.g., Barcelona, Montevideo, Wedge Splitting Test, EN 14651) are time-consuming and relatively complex, demanding careful preparation and hydraulic presses with data acquisition systems. This paper aims to contribute to the discussion of alternatives for FRC characterization procedures by the introduction of a pullout analytical model to predict FRC constitutive law based on DEWS tests. The pullout analytical model is calibrated with pullout tests considering straight and hooked-end steel fibers with different strengths, embedment lengths, inclinations, and diameters. A 3D digital model reproducing the experimental DEWS samples is developed, considering the fibers' randomness distribution, orientation, and wall effect. The analyses are compared with four different experimental tests. The results define a linear post-peak tensile constitutive law, considering the mean residual stress values at different COD values. The model was able to reproduce the specimens' production procedures and the fibers' distribution and orientation correctly. The results demonstrated a good agreement with the experimental data. [Display omitted] • Pullout analytical model for straight and hooked-end steel fibers. • 3D parametric digital model considering fibers random distribution and wall effect. • Prediction of DEWS tests considering the manufacturing process. • Determination of FRC linear tensile post-peak constitutive model. [ABSTRACT FROM AUTHOR]

A new methodology to investigate transient creep at the material level in terms of Load Induced Thermal Strain (LITS) by means of numerical mesoscopic analysis is proposed. 2D three-phase axisymmetric analyses for plain concrete cylindrical specimens are performed. Concrete phases are represented by the ITZ, the inclusions (coarse aggregates) and the matrix (cement paste including fine aggregates). The aggregates are placed using the take and place method and normal random distribution. An appropriate finite element mesh with coupled temperature-displacement is adopted using concrete damaged plasticity model in Abaqus. The numerical values are compared with a LITS semi-empirical model and experimental results available in the literature. For the analyses, LITS is uncoupled into matrix strain (dehydration of Calcium-Silicate-Hydrates), aggregate strain (geomechanical properties degradation) and mechanical strain (thermal expansion restraint). A parametric analysis is performed to investigate the sensitivity of the model due to different aggregate distributions and mechanical properties variations. A comparison of concretes with circular and polygonal aggregate shapes is performed, highlighting the main differences between the materials behavior and some challenges imposed to the model calibration. In general, a good correlation between the numerical and the experimental results was observed. The numerical model was able to represent LITS nonlinear behavior due to variations of the applied load. In addition, the numerical results, in terms of load path analysis with preloaded and preheated specimens, are in agreement with experimental observations. LITS acceleration is clearly demonstrated and the results showed that the boundary conditions play an important role in the total deformation. [ABSTRACT FROM AUTHOR]