Generation of 3D geotechnical particles using random angular bend algorithm

• DOI: 10.1002/nag.3515
• Publication Date: 2023-03-20T11:11:01Z
• Authors: Wengang Zhang; Liang Han; Zhenyu Wang; Hanlong Liu; Luqi Wang; Xuecheng Gao

AbstractGeo‐materials are generally regarded as multi‐phase structural systems, which are composed of particles, pores, and fissures. Generally, the shape and size of particles have considerable influence on the physical and mechanical properties of geo‐materials. Based on the theory of random angular bend (RAB) algorithm, this paper presented a method to finely generate the three‐dimensional (3D) random particle model via the superposition of two‐dimensional (2D) particle profiles, and this proposed method was coded in Python. According to the generation procedure, the relevant point cloud data should be generated, and with the aid of the point cloud normal vector calculation and Poisson reconstruction technique, the required 3D particle profiles can be obtained. Subsequently, three shape descriptors are selected to describe the shape characteristics of generated particles, and accordingly, the quantitative relationships between the shape descriptors and the input controlling parameters are established to accurately control the shape of generated particles. Finally, the generated geo‐material particles are implemented into the particle flow code (PFC3D) to construct the clump library, and an example about a 3D soil–rock mixture model was taken to validate the performance of the proposed method for numerical simulation. The results show that this study provides a useful strategy for the simulation of the geo‐materials based on the RAB theory.