Multi-super-ellipsoid model for non-spherical particles in DEM simulation
Multi-super-ellipsoid model for non-spherical particles in DEM simulation
复制标题
离散元模拟中用于非球形颗粒的多超椭球模型
DOI：
10.1016/j.powtec.2019.09.042
复制
发表时间：
2020-02-01
影响因子：
5.2
通讯作者：
Zhao, Yongzhi
中科院分区：
文献类型：
--
作者：
Liu, Zihan;Zhao, Yongzhi
离散元法广泛应用于颗粒系统的研究，颗粒的形状对颗粒的动力学行为有很大的影响。为了更准确地描述某些非球形颗粒，本文提出了一种类似于多球模型的多超椭球模型。该模型通过连接一定数量的超椭球体来模拟非球形颗粒，并且组成超椭球体可以相互重叠。多超椭球体模型可以较准确地描述片剂、胶囊等非球形颗粒。为了验证多超椭球模型的准确性，本研究采用了四种非球形颗粒，包括三种片剂和一种胶囊。采用数值模拟和实验相结合的方法，对四种非球形颗粒在圆柱形容器中的装填和三种片剂在水平旋转滚筒中的流动进行了研究，并将模拟结果与相应的实验结果进行了比较。结果表明，模拟结果能够较准确地预测这类非球形颗粒的基本特性，包括填充高度和动态休止角，表明多超椭球模型具有较高的精度。对新模型的计算效率进行了测试，结果表明计算效率与颗粒形状和复合颗粒中的子颗粒数有关。(C)2019爱思唯尔B.V.保留所有权利。
The discrete element method (DEM) is extensively used in the investigation of granular systems and the particle shape has a great influence on the dynamics behaviors of granular material. To describe some kinds of non-spherical particles accurately, a multi-super-ellipsoid model, somewhat similar to the multi-sphere model, is proposed in this paper. With this model, non-spherical particles are modeled by connecting a certain number of super-ellipsoids, and the component super-ellipsoids can overlap with each other. Some non-spherical particles, such as tablets and capsules, can be described accurately by the multi-super-ellipsoid model. To verify the accuracy of the multi-super-ellipsoid model, four kinds of non-spherical particles including three kinds of tablets and one kind of capsule, are adopted in this research. The packing of four kinds of non-spherical particles in a cylinder container and the flow of three kinds of tablets in a horizontal rotating drum are done by both numerical simulation and experiment, and the simulation results are compared with the corresponding experiments. The results show that the simulation can predict the basic characteristics of such kinds of non-spherical particles accurately, including fill heights and the dynamic angles of repose, which show that the accuracy of the multi-super-ellipsoid model is quite high. The computational efficiency of the new model is also tested, and the results indicate that the computational efficiency depends on the particle shape and the number of sub-particles in the composite particle. (C) 2019 Elsevier B.V. All rights reserved.