Authors

Abstract

In view of the lack of accurate contact parameters and the difficulty of measuring contact parameters in the discrete element simulation of mechanized harvesting of highland barley, this study took the wax-ripening highland barley as the object, carried out the discrete element simulation of highland barley stem based on EDEM, and calibrated the discrete element simulation parameters of highland barley stem by response surface optimization. In this paper, Plackett-Burman test was used to screen 8 initial parameters. It was found that the static friction coefficient between highland barley stems, the rolling friction coefficient between highland barley stems, and the rolling friction coefficient between highland barley stems and steel plates have significant effects on the particle angle of repose. Based on the optimal value range of significant factors determined by the steepest ascent test, a second-order regression model of the angle of repose and significant parameters was established and optimized based on the results of Box-Behnken test. The optimal parameter combination of significant parameters was obtained as follows: static friction coefficient between highland barley stems is 0.27, rolling friction coefficient between highland barley stems is 0.07, and rolling friction coefficient between highland barley stems and steel is 0.26. Finally, the simulation results under the optimal parameter combination are compared with the actual test angle of repose. The relative error is 0.52 %. That indicates that the parameters of the simulation calibration are credible, which can provide a reference for the future research on the cleaning device in the mechanized harvesting of highland barley.

Abstract in Chinese

针对目前青稞机械化收获离散元仿真缺乏准确的接触参数、接触参数测量难度大的问题，本研究以蜡熟期的青稞为对象，基于EDEM开展青稞茎秆离散元仿真，通过响应面优化标定了青稞茎秆离散元仿真参数。研究应用Plackett-Burman试验对8个初始参数进行筛选，发现青稞茎秆间静摩擦系数、青稞茎秆间滚动摩擦系数与青稞茎秆-钢板滚动摩擦系数对颗粒堆积角有显著影响。以最陡爬坡试验确定的显著性因素最优取值区间为基础，基于Box-Behnken试验结果建立堆积角与显著性参数的二阶回归模型并对其进行优化，得到显著性参数的最佳参数组合为青稞茎秆间静摩擦系数0.27、青稞茎秆间滚动摩擦系数0.07、青稞茎秆-钢板滚动摩擦系数0.26。最后将最佳参数组合下的仿真结果与真实试验堆积角对比，二者相对误差为0.52%，误差很小表明仿真标定的参数是可信的，可以为以后的青稞机械化收获中清选装置的研究提供参考。