基于颗粒缩放理论的沙棘茶毫离散元参数标定

为解决沙棘叶茶茶毫离散元仿真中缺乏准确接触参数的问题，利用物理和仿真休止角堆积试验标定茶毫接触参数；将茶毫颗粒简化为软质球形颗粒，通过量纲分析和颗粒缩放理论将茶毫空气动力学当量粒径从231.37μm放大至1.8 mm。利用EDEM软件，选定“Hertz–Mindlin with JKR”接触模型，以休止角为响应值，通过Plackett-Burman试验筛选出对休止角影响最显著的3个参数：茶毫-茶毫恢复系数、茶毫-茶毫滚动摩擦系数、茶毫-不锈钢滚动摩擦系数；利用最陡爬坡试验，确定各参数最佳取值范围；根据Box-Behnken试验建立并优化3个显著性参数与休止角的二阶回归方程，对回归方程进行寻优求解。得到3个显著性参数的最优组合：茶毫-茶毫恢复系数为0.159,茶毫-茶毫滚动摩擦系数为0.290,茶毫-不锈钢滚动摩擦系数为0.239。通过对比休止角仿真试验值与物理试验值，二者相对误差为1.97%,表明仿真试验预测效果良好。

Discrete element parameters calibration of Hippophae rhamnoides based on particle scaling theory

To address the lack of accurate contact parameters in the discrete element simulation of sea buckthorn leaf tea hair, a physical and simulation resting angle heap experiment was conducted to calibrate the contact parameters of tea hair. The tea hair particles were simplified as soft spherical particles. By dimensional analysis and particle scaling theory, the aerodynamic equivalent diameter of tea hair was enlarged from 231.37 μm to 1.8 mm. The EDEM software was used with the “Hertz|Mindlin with JKR” contact model. The repose angle was chosen as the response variable, and a Plackett|Burman experiment was performed to select the three most significant parameters affecting the resting angle: the tea hair|tea hair restitution coefficient, the tea hair|tea hair rolling friction coefficient, and the tea hair|stainless steel rolling friction coefficient. The optimal parameter ranges were determined using a steepest ascent test. Based on Box|Behnken experimental design, a second|order regression equation was established and optimized for the three significant parameters and repose angle. The three significant parameters for the optimal combination were as follow: tea hair|tea hair restitution coefficient was 0.159, tea hair|tea hair rolling friction coefficient was 0.290, and tea hair|stainless steel rolling friction coefficient was 0.239. A comparative analysis between the simulated repose angle values and the real values from physical experiments yielded a relative error was 1.97%, indicating good predictive performance in simulation experiments.