Establishment of a three-dimensional particle library for graded crushed stone based on a new aggregate morphology characterization method

The morphology of graded crushed stone (GCS) particles has an essential influence on the performance of aggregate mixtures. The impact of particle shape is a comprehensive effect that cannot be considered separately, leading to difficulties in establishing the relationship between the mixture properties and the aggregate morphology by using laboratory methods. The discrete element method (DEM) is an effective way widely adopted to reconstruct the morphology of particles and simulate performance tests of granular materials. However, selecting limited particles characterizing a real particle-assembly for simulation is still a challenge in current research due to the inherent rich variability of particle shapes. In this study, based on the acquisition of three-dimensional (3D) aggregate shapes by using laser scanning, ellipsoid index (EDI) translating the particle shape as a function of surface area, volume, and contour length is proposed to comprehensively evaluate aggregate morphology. Further, a particle library capable of characterizing aggregate morphology distribution is established based on the statistics of the corresponding morphological characteristics of particle samples. The model reliability is validated by carrying out a series of experimental and numerical penetration tests with nine different gradations. The established particle library can be used to model aggregate mixtures and the proposed simulation framework is promising for optimizing the mixture gradation design numerically.