Virtual fabrication and computational simulation of asphalt concrete microstructure

Abstract

This study presents the development and verification of an in-house microstructure generator that can be used to fabricate virtual, two-dimensional microstructures of asphalt mixtures. The proposed methodology takes advantage of a modern image analysis system, i.e. Aggregate Image Measurement System 2, to obtain geometric characteristics of real aggregate particles. During the generation process, images of real aggregates are stored in a database and further positioned in the virtual microstructures in a way that the gradation, area fraction and orientation of the virtual aggregate skeleton closely reproduce properties of real asphalt concrete samples fabricated in the laboratory. To demonstrate the capabilities of the generator, a microstructural analysis procedure used to quantify key volumetric characteristics of virtual samples is first proposed and verified. Then, using the verified analysis methodology, microstructural characteristics of virtual samples are determined and compared to those of real specimens. To further demonstrate the applicability of the proposed methodology, dynamic modulus simulations of virtually generated samples of asphalt concrete are performed using the finite element software ABAQUS and considering different aggregate characteristics. The results obtained indicate that the proposed methodology can be used as an efficient tool to fabricate and analyse virtual microstructures of asphalt concrete mixtures. This, in turn, becomes an essential step for the improvement of the computational microstructural model under development by the authors, which aims to accurately predict the mechanical behaviour of asphalt mixtures from known constituent properties and geometric characteristics, and to provide significant savings in the time and cost required for laboratory testing of actual specimens.

Keywords:

• Asphalt mixtures
• virtual microstructure generation
• representative volume element
• finite element method
• computational microstructure modelling