Modeling the Behavior of an Aggregate Skeleton during Static Creep of an Asphalt Mixture Based on a Three-Dimensional Mesoscale Random Model

The aggregate skeleton and its evolution under external loads are crucial to evaluate the mechanical properties of the asphalt mixture. This study developed a method to identify the three-dimensionally (3D) aggregate skeleton based on geometry information. The proposed method can identify the evolution process of the aggregate skeleton during loading in the finite element method (FEM). Random models based on the Gilbert-Johnson-Keerthi (GJK) algorithm are used to study the influence of the concave-convex degree of the gradation curve and the magnitude of the creep load on the aggregate skeleton during the creep process. The results indicate that the aggregate skeleton undergoes three stages during creep: aggregate redistribution, skeleton formation, and skeleton failure after reaching the ultimate load, while the increase of load causes the aggregate skeleton to lose its bearing capacity faster. More coarse aggregate raises the strength of the aggregate skeleton, but it is more likely to break down quickly once it reaches the strength. More fine aggregate enhances the stability of the aggregate skeleton. The proposed identification method can serve as a general tool to investigate the evolution process of the internal aggregate structure of the asphalt mixture.