Concrete is a material whose behavior is complex, especially in cases of extreme loads. The objective of this thesis is to carry out an experimental characterization of the behavior of concrete under impact-generated stresses (confined compression and dynamic traction) and to develop a robust numerical tool to reliably model this behavior.
In the experimental part, we have studied concrete samples from the VTT center (Technical Research Center of Finland). At first, quasi-static triaxial compressions with the confinement varies from 0 MPa (unconfined compression test) to 600 MPa were realized. The stiffness of the concrete increases with confinement pressure because of the reduction of porosity. Therefore, the maximum shear strength of the concrete is increased. The presence of water plays an important role when the degree of saturation is high and the concrete is subjected to high confinement pressure. Beyond a certain level of confinement pressure, the maximum shear strength of concrete decreases with increasing water content. The effect of water also influences the volumic behavior of concrete. When all free pores are closed as a result of compaction, the low compressibility of the water prevents the deformation of the concrete, whereby the wet concrete is less deformed than the dry concrete for the same mean stress.
The second part of the experimental program concerns dynamic tensile tests at different loading velocities, and different moisture conditions of concrete. The results show that the tensile strength of concrete C50 may increase up to 5 times compared to its static strength for a strain rate of about 100 s-1.
In the numerical part, we are interested in improving an existing constitutive coupled model of concrete behavior called PRM (Pontiroli-Rouquand-Mazars) to predict the concrete behavior under impact. This model is based on a coupling between a damage model which is able to describe the degradation mechanisms and cracking of the concrete at weak confinement pressure and a plasticity model which allows to reproduce the concrete behavior under strong confinement pressure. The identification of the model was done using the results of experimental tests. The improvement of this model, especially the plasticity part, focuses on three main points : taking into account the effect of the deviatoric stress in the calculation of the mean stress; better accounting for the effect of water using poromechanical law instead of mixing law, improvement of the coupling variable between the damage model and the elastoplastic model with consideration of the Lode angle. These improvements were then validated by comparing numerical results and impact tests. The improved model is capable of reproducing the behavior of concrete under different loading paths and at different levels of confinement pressure while taking into account the degree of saturation of concrete.