Discrete element method for modeling the long-term aging viscoelastic behavior of concrete considering its mesostructure

Detailed models of concrete mesostructure can be used to understand the interactions between its components and predict complex deterioration scenarios. The discrete or distinct element method (DEM) is currently being used for modeling the fracture process of quasibrittle materials, such as rock and concrete. An explicit formulation of a DEM contact model that includes aging viscoelastic behavior based on the solidification theory is proposed, allowing the DEM particle model to be applied to delayed concrete analysis. Because of the timestep constraints of the DEM, a fast numerical procedure for the analysis of long-term aging viscoelastic behavior of concrete is also proposed. A calibration procedure for the aging viscoelastic contact model parameters is presented, including new expressions for the delayed deformability macro properties. The presented validation tests using a one-contact particle assembly show good agreement between the fast numerical procedure, the fully explicit DEM procedure with small timestep, and the creep compliance analytical solution. The contact aging model validation tests using larger regular and random particle assemblies show that the fast numerical procedure significantly reduces the computational costs by introducing large timesteps in which the solution is computed while giving the same accuracy as the fully explicit procedure. The DEM aging concrete model is validated using a B3 model fit to Ward et al.'s experimental results for different loading ages. The obtained numerical results show that the DEM aging viscoelastic particle model, considering the concrete mesostructure can predict the long-term behavior of concrete once the contact properties of each component are properly calibrated (mortar, aggregate, and interfacial transition zone).