Generation of virtual asphalt mixture porosity for computational modelling

Abstract

A note on versions:The version presented here may differ from the published version or from the version of record. If you wish to cite this item you are advised to consult the publisher's version. Please see the repository url above for details on accessing the published version and note that access may require a subscription. These elements are then grown by mimicking the biological mechanism called contact inhibition until a target air voids content (AVC), chosen by the user, is met. In addition, multiple 2D domains can be converted to 3D and combined to generate a multi-layered realistic representation of the porosity present in asphalt mixtures. In this paper, the working mechanism of the algorithm is described and its efficiency is assessed. Moreover, the validity of the results is discussed, and virtual domains, in both 2 and 3 dimensions, are compared with real CT scans in order to show the efficacy of this approach. It was found that the virtual representations of the asphalt mixture porosity show realistic characteristics in terms of air voids content and that the air voids size distribution is consistent with that of real specimens.

Stochastic generation of virtual air pores in granular materials

“…In [6] a method to generate virtual asphalt samples is described. The method starts with the generation of 2D planes of virtual particles based on the biological mechanism of contact inhibition and can be addressed as a packing algorithm.…”

“…A set of conditions similar to relation 1 can be developed in the case of ellipses, too. For further details on the packing method for ellipses, see [6]. The randomization of the coordinates of the centres of the particles in the selected portion of the xy plane (set A in relation 1) is performed with a random number generator using the standard uniform distribution [6].…”

“…The model of growth introduced in [6] and used in this paper belongs to the field of stochastic geometry, therefore, it is relevant to compare it to similar mathematical methods mentioned in the scientific literature. The conditions for growth defined in relation 1 and used in [6] follow the so-called touch-and-stop model of growth [2], which can be defined as a pattern formation process intermediate between the random sequential absorption (RSA) model [2] [7] and the JohnsonMehl-Avrami-Kolmogorov (JMAK) model [2].…”

“…The current models describing porous media are all based on advanced theoretical principles, therefore, their implementation is complex, unless commercial software is used [21] or the researchers are expert in mathematical modelling. In [6] a new model based on geometry only is described and analysed by the present authors. The method was developed in order to allow a theoretically-based and simple way to generate granular material virtual representations.…”

“…The method was developed in order to allow a theoretically-based and simple way to generate granular material virtual representations. In [6], the authors proved that the method provides realistic results, however, a further investigation on the features of the void space is necessary, mainly because the 3D model generated previously only provided a proof of concept and was obtained through a manual trial and error method. In this paper, we describe the automation of the mathematical method presented in [6] and analyse the void patterns that it generates, providing topological information on the virtual samples that are created in terms of isotropy and coordination number.…”