During the long-term service of structures, the environmental temperature and humidity fields can lead to the evolution of the pore structure within concrete materials, thereby affecting changes in mechanical properties. However, the mechanisms of their influence still need further clarification and are challenging to correlate for prediction. Addressing this specific issue, the current study focuses on evaluating 78 cubic concrete specimens with C30 and C40 strength grades. These specimens were subjected to a 95 % humidity environment at temperatures of 40 °C, 60 °C, and 80 °C for durations of 30, 60, and 90 days. The research involved conducting both microstructural and macroscopic mechanical tests to quantify the evolution of internal hydrates, pore structures, and pore distribution, as well as to assess the changes in the macroscopic mechanical properties of the specimens. To accurately predict the process, a multiscale finite element model (FEM) based on the bilinear continuum cohesive damage model was established. This model quantifies the changes in mechanical properties of mesoscopic three-phase concrete due to microcrack evolution under hygrothermal coupling. To facilitate multiscale simulations between the Representative Volume Element (RVE) and macroscopic scale models, relevant subroutines were written using Umath, and parallel computing was employed to reduce the computational burden. For the macroscopic model in this study, a homogenized concrete cube model with dimensions identical to the experimental specimens was selected. The effectiveness of the parallel computational model was validated by comparing and analyzing the simulation results under uniaxial compression with the experimental data. The findings indicate that hygrothermal action leads to the initiation and expansion of microcracks within the concrete, thereby impacting the alteration of its mechanical properties. In addition, the parallel computing-based multiscale FEM can more accurately quantify the typical damage process of crack propagation while reducing the computational load and may provide relatively accurate predictions of the mechanical characteristics of concrete under the influence of hygrothermal coupling, ensuring effective modeling of macroscopic structural specimens in various temperature and humidity environments.
Read full abstract