Evaluation of the discrete element method (DEM) and of the experimental evidence on concrete behaviour under static 3D compression

Abstract

AbstractThe authors successfully employed the discrete element method (DEM) in numerical determinations of the response up to and beyond failure of reinforced concrete structures subjected to impact and impulsive loadings in which tensile fracture, which is reliably predicted by DEM models, often controls the dominant failure modes. However, in impact problems when penetration occurs, the reliability of the approach in predictions of the structural response of the 3D compression zone that develops at the tip of the projectile has not yet been explicitly confirmed. In this context, in view of its complexity, the performance of the method is herein assessed and compared with available experimental results in static tests. By means of numerical simulations, it was previously verified that DEM models do predict, but overestimate, the strength increase observed on concrete cubes subjected to static multiaxial compression in relation with the unconfined strength, for confining (lateral) pressures up to about 20% of the unconfined compressive stress. For higher confining stresses, however, the DEM formulation underestimates the compressive strength increase observed in cubic and cylindrical samples, for the reasons examined in the paper, in which limitations of both the numerical predictions and experimental observations are thoroughly discussed.