Abstract: In order to ensure the accuracy of the parameters used in the simulation test of peanut pods, the study uses the method of combining actual physical test and simulation test to calibrate the discrete element simulation parameters. First of all, the basic physical parameters of peanut pods (including dimensions, density, moisture content, bulk density, elastic modulus, Poisson wave and shear modulus) are measured through actual physical experiments, and the simulation is determined based on the results of various physical quantities measured by actual physical experiments. The value range of the test parameters is carried out, and the PlackettBurman test is carried out. The factors that have a significant influence on the stacking angle are selected: peanut podpeanut pod static friction coefficient, rolling friction coefficient, peanut podsteel plate static friction coefficient. The value range of the significant factor is further determined by the steepest climb test. BoxBehnken test is carried out to establish the quadratic regression equation between the accumulation angle and the significant factor, and solve the equation with the actual physical test accumulation angle (31.63°) as the target value, and obtain the best simulation parameters:the static friction coefficient and rolling friction coefficient of peanut pods, and the static friction coefficient between peanut pods and steel plates are 0.74, 0.24 and 0.58, respectively. Finally, a simulation test on the best simulation parameters determined after the experimental analysis is performed, and an independent sample T test on the obtained simulation value and the actual test value is performed, with P>0.05, which shows that there is no significant difference between the actual test accumulation angle and the simulation test accumulation angle, and the relative error is 2.877%,  the accuracy of the simulation test is also verified. By comparing the methods and test results used in the calibration of other material parameters, it further shows that the calibrated parameters can provide references for related research.

Key words: peanut pods, stacking angle, DEM simulation, parameter optimization, parameter calibration

摘要： 为保证花生荚果在仿真模拟试验时所用参数的准确性，通过实际物理试验与仿真模拟试验相结合的方法对离散元仿真参数进行标定。首先，通过实际物理试验测得花生荚果基本物理参数(外形尺寸、密度、含水率、容重、泊松比、弹性模量和剪切模量)，依据实际物理试验测得的各物理量结果确定仿真模拟试验参数取值范围，开展PlackettBurman试验，筛选出对堆积角存在显著性影响的因素为：花生荚果—花生荚果静摩擦系数、滚动摩擦系数，花生荚果—钢板静摩擦系数。进一步通过最陡爬坡试验确定显著性因素的取值范围。开展BoxBehnken试验，建立堆积角与显著性因素之间的二次回归方程，并以实际物理试验堆积角(31.63°)为目标值对方程进行求解，得到最佳仿真模拟参数：花生荚果间静摩擦系数、滚动摩擦系数，花生荚果—钢板静摩擦系数分别为0.74、0.24和0.58。最后，对试验分析后确定的最佳仿真参数进行仿真模拟试验，对取得的仿真模拟值与实际试验值进行独立样本T检验得出P>0.05，表明实际试验堆积角与仿真模拟试验堆积角无显著性差异，且相对误差为2.877%，验证了仿真模拟试验的准确性。通过对比其他物料参数标定时所用方法及试验结果，进一步表明标定后的参数可为相关研究提供参考。

关键词: 花生荚果, 堆积角, 离散元仿真, 参数优化, 参数标定