Motivated by the virtual testing of asphalt concrete, the North Carolina State University research team has developed an algorithm to computationally generate air voids. After examining the X-ray tomographic images of real asphalt concrete microstructure, we concluded that the air void's shape and size are affected primarily by the surrounding local aggregate structure. Building on this observation, we developed an algorithm to generate random but representative air void configurations inside a given microstructure. By applying the algorithm to scanned aggregate structures, we show that the generated air voids not only look visually similar to actual air voids, but also are effective in capturing modulus reduction. The algorithm is included in a virtual aggregate structure generation framework, resulting in a streamlined virtual fabrication procedure for asphalt concrete that can qualitatively capture the effects of accelerated degradation due to the presence of air voids.
Read full abstract