Evaluating concrete material models for blast analysis using 3D finite element analysis

Abstract

Blast loads are extremely detrimental to structural elements of a building and can result in minor damage to complete failure of a building. Finite element analysis (FEA) is an effective tool to study the behavior of structural elements exposed to blast loading. The focus of this study was to evaluate the performance of three concrete material models available in LS-DYNA and compare the results to experimental results from blast loaded reinforced concrete (RC) panel supported on four sides (two-way bending). Concrete material models chosen for this study were Winfrith (MAT_84-85), Karagozian & Case (MAT_072R3), and Continuous Surface Cap Model (MAT_159). These are widely used in finite element modeling related to civil structures and can account for strain rate dependent properties. Each concrete material model was implemented with and without strain rate effects. The finite element model had concrete material as solid elements and steel reinforcement material as beam elements with simply supported boundary conditions. Results from two different blast loading methods are presented: (1) blast pulse load curve from experimental data (2) blast input by defining type, location, and amount of explosive in terms of TNT. The finite element models were evaluated by comparing fundamental period, peak displacement, and residual displacement to experimental results. In this study, the model using MAT_072R3 most closely matched experimental behavior in comparison to MAT_84-85 and MAT_159. Additional discussion is presented about: (1) single degree of freedom (SDOF) analysis, (2) comparison of numerical and experimental crack patterns, and (3) behavior of concrete material models accounting for compressive and tensile dynamic increase factors (DIFs).