Random generation of 2D PFC microstructures through DEM gravimetric methods

Abstract

Computational mechanics has proven to be a useful method to understand and assess the mechanical behaviour and main degradation mechanisms of Porous Friction Courses (PFC) at the microstructural level. However, a common limitation when preparing these models is the difficulty to achieve realistic microstructures. The ideal model to evaluate the mechanical behaviour and the degradation mechanisms of PFCs would use a three-dimensional (3D) geometry of the mixture with realistic aggregate shape, angularity, and gradation. However, these models are computationally expensive and typically related with numerical instabilities. This paper proposes a novel methodology to generate two-dimensional (2D) random PFC microstructures using gravimetric methods through Discrete Element Method (DEM) modelling. The proposed methodology was calibrated to generate coarse aggregates with realistic and controlled morphological properties. The produced 2D microstructures were implemented in the commercial Finite Element Method (FEM) software Abaqus® and, after coating the particles with asphalt mortar, they were used to verify that the microstructures were able to reproduce the mechanical response –measured through the dynamic axial modulus– of an actual PFC mixture tested in the laboratory. The results show that the proposed methodology is able to produce microstructures with the same dynamic modulus than an actual PFC mixture at different loading frequencies. Among other applications, this methodology could be used to estimate the mechanical properties of PFC mixtures or to assess the expected performance and durability of a specific PFC under realistic field conditions.