Abstract Intrinsic flaws are an important factor affecting particle strength and breakage behavior and are closely related to the macroscopic behavior of granular soils, e.g., packing characteristics, strength, deformation, localization and failure behaviors, etc. In this paper, a novel DEM modeling approach for investigating the role of flaws on the particle crushing behavior is proposed. The idea is to use the realistic fracture surfaces obtained from the experiment as a basis for generating initial flaws in DEM in a statistical way. Firstly, nano-focus X-ray computed tomography technology was used to obtain the realistic particle information including particle morphology and a set of fracture surfaces. The particle information was then stored in several arrays and imported into the DEM to construct the particle model containing a reasonable set of initial flaws. Seven simulation cases were performed to study the role of initial flaws on the particle breakage behavior. Simulation results show that the proposed modeling approach is capable of reproducing the realistic particle breakage behaviors, including the small-strain stiffness, peak crushing load, fracture pattern and evolution of fragment shape parameters.

中文翻译：

---

含固有缺陷砂粒破碎的 DEM 研究

摘要 固有缺陷是影响颗粒强度和破碎行为的重要因素，与颗粒土的宏观行为密切相关，如填充特性、强度、变形、局部化和破坏行为等。本文提出了一种新的DEM建模方法。为研究缺陷对颗粒破碎行为的作用提出了建议。这个想法是使用从实验中获得的真实断裂表面作为以统计方式在 DEM 中生成初始缺陷的基础。首先，利用纳米焦点X射线计算机断层扫描技术获取真实的颗粒信息，包括颗粒形态和一组断裂面。然后将粒子信息存储在多个阵列中并导入 DEM 以构建包含一组合理的初始缺陷的粒子模型。进行了七个模拟案例来研究初始缺陷对颗粒破碎行为的作用。仿真结果表明，所提出的建模方法能够再现真实的颗粒破碎行为，包括小应变刚度、峰值破碎载荷、断裂模式和碎片形状参数的演变。