Abstract Particle morphology is an inherent characteristic that has a significant influence on breakage behaviors of natural sands. Based on X-ray micro-computed tomographic scanning and image-processing, the three-dimensional (3D) particle surfaces of natural sand particles were first reconstructed by the spherical harmonic (SH) analysis. The traditional morphological parameters of the particles, including sphericity, roundness, and aspect ratio were then calculated based on the SH-reconstructed particle surfaces. Furthermore, a 3D local roundness descriptor was used to characterize the local angularity of the contact area between a particle and its loading platens. To model particle breakage, this study developed a combined finite-discrete element method (FDEM) in which the cohesive interface elements were inserted between the finite elements to simulate solid fracture by defining a traction-separation damage law. By using the developed FDEM, a series of single particle crushing simulations were conducted for ellipsoid particles and two types of sand particles. The results show that the developed FDEM model is capable of simulating breakage behaviors of sand particles. Statistical analysis showed that local roundness is the most significant factor in determining the particle fracture pattern, with the strongest positive correlation with the particle crushing strength of sand particles subjected to uniaxial loading.

中文翻译：

---

结合有限离散元法研究颗粒形态对单颗粒破碎的影响

摘要 颗粒形态是一种内在特征，对天然砂的破碎行为具有重要影响。基于 X 射线显微计算机断层扫描和图像处理，首先通过球谐 (SH) 分析重建天然砂颗粒的三维 (3D) 颗粒表面。然后基于SH重建的颗粒表面计算颗粒的传统形态参数，包括球形度、圆度和纵横比。此外，使用 3D 局部圆度描述符来表征粒子与其加载压板之间接触区域的局部角度。为了模拟颗粒破碎，本研究开发了一种组合有限离散元方法 (FDEM)，其中将粘性界面元素插入有限元之间，通过定义牵引-分离损伤规律来模拟固体断裂。通过使用开发的 FDEM，对椭圆体颗粒和两种类型的砂颗粒进行了一系列单颗粒破碎模拟。结果表明，开发的 FDEM 模型能够模拟砂粒的破碎行为。统计分析表明，局部圆度是决定颗粒破裂模式的最重要因素，与单轴载荷作用下砂颗粒的颗粒抗压强度正相关最强。对椭球体颗粒和两种类型的砂颗粒进行了一系列单颗粒破碎模拟。结果表明，开发的 FDEM 模型能够模拟砂粒的破碎行为。统计分析表明，局部圆度是决定颗粒破裂模式的最重要因素，与单轴载荷作用下砂颗粒的颗粒抗压强度正相关最强。对椭球体颗粒和两种类型的砂颗粒进行了一系列单颗粒破碎模拟。结果表明，开发的 FDEM 模型能够模拟砂粒的破碎行为。统计分析表明，局部圆度是决定颗粒破裂模式的最重要因素，与单轴载荷作用下砂颗粒的颗粒抗压强度正相关最强。