Authors

Abstract

This study measured the intrinsic and contact parameters through physical experiments to improve the accuracy of discrete element simulation analysis of peanut seeds. Discrete element models for five different peanut seed filling ball numbers were established. The simulation parameters were calibrated through a combination of physical and simulation experiments. Firstly, the Plackett-Burman test was used to screen the significance of simulation parameters. Then, the steepest climbing test was conducted to determine the optimal range of significance parameters using the relative error be-tween the simulated and the physical experimental as the evaluation index. Finally, a response surface experiment with three factors and three levels was conducted using the repose angle as the response value. The static and rolling friction coefficients among peanut seeds were set as 0.43 and 0.50 separately, and the rolling coefficient between peanut seeds and steel plate was set as 0.12. During verification experiments, the simulated repose angle was 25.18°, with a relative error of 2.42% compared to the physical repose angle, further verifying the reliability of the simulation model. The re-search group used different numbers of filling balls with optimal parameters in the repose angle experiment. Then they evaluate the simulation time and the error value of repose angle between the simulated and physical experiment. The optimal number of filled balls is the Sphere 1178. The research results indicate that discrete element model of peanut seeds and calibration parameters are reliable. Based on the results of this research, an intelligent peanut precision sowing machine can be developed.

Abstract in Chinese

为了提高花生籽粒离散元仿真分析过程中的仿真精度，该研究通过物理试验测定了花生籽粒的本征参数和接触参数，建立了五种不同填充球数的花生籽粒的离散元模型，通过物理试验与仿真试验相结合的方式对仿真参数进行了标定。首先，采用Plackett–Burman试验对仿真参数进行显著性筛选。然后，以仿真堆积角与物理试验堆积角之间的相对误差为评价指标，进行最陡爬坡试验，确定了显著性参数的最优取值范围。最后，以堆积角为响应值，进行了三因素三水平的响应面试验，得到最优参数组合为：花生籽粒间的静摩擦系数0.43、花生籽粒间滚动摩擦系数0.50、花生籽粒-钢板滚动摩擦系数0.12。并进行验证试验，得出仿真堆积角为25.18，与物理堆积角之间的相对误差为2.42%，进一步验证了仿真模型的可靠性。以最优参数进行不同填充球数的堆积试验，以仿真堆积角与物理实验堆积角的误差值和仿真时间为评价指标，得到最佳填充球数为1178。研究结果表明该花生籽粒离散元模型及标定参数可靠，可为智能花生精量播种机具的研制提供依据。