Calibration and experimental verification of discrete element parameters of oil sunflower seeds

Abstract: Discrete element method (DEM) is widely used to simulate the key work process of oil sunflower seeding and harvesting. However, feature parameters of oil sunflower seeds are still lacking, particularly on the contact parameters between seeds and mechanized planting/harvesting equipment. Taking the oil sunflower seeds as the research object, this research aims to establish a discrete element model using a 3D scanning reverse simulation in the EDEM software, and then to calibrate the specific parameters of oil sunflower seeds. The average length, width, and thickness of 567DW oil sunflower seeds were 13.36, 6.94, and 4.63 mm, respectively, in an absolute digital caliper. The weight of 100 567DW oil sunflower seeds was measured at 9.547 g using an electronic analytical balance. The density of 567DW oil sunflower seeds was 849.4 kg/m3, and the water content was 10.13% in a B0D-75-Ⅱ electric thermostatic drier and pycnometer. Poisson's ratio, elastic modulus, and shear modulus of oil sunflower seeds were 0.413, 304.69, and 107.82 MPa, respectively. A universal tensile testing machine was also used. The static friction coefficients between oil sunflower seeds and steel plate/plexiglass were 0.423 and 0.525 on the inclined plane, respectively. A physical stacking test was performed on a cylinder lifting for the oil sunflower seeds. MATLAB platform was selected to carry out a gray-scale and binarization processing on the image of stacking angle. Bwperim and imfill functions were selected to extract the outline of binary image, and then perform the holes filling and expansion on the inner edge, finally to obtain the edge contour. The edge contour was imported into the Origin software, where the converted coordinate data was obtained for the linear fitting slope, thereby obtaining a stacking angle of 48.858°. Taking the stacking angle as the evaluation index, the Plackett-Burman test was used to screen out the factors with a significant impact on the stacking angle and the static and dynamic friction coefficients between the oil sunflower seeds. The steepest ascending path was used to determine the range of factors with a significant impact on the stacking angle. The variance analysis was also made to optimize the response surface. Optimal static/dynamic friction coefficients between oil sunflower seeds were determined to be 0.41 and 0.05, respectively. A stacking test was simulated with a better parameter combination, where the measured stacking angle was 48.976°, only 0.24% in error with the stacking angle, compared with the physical stacking test. The electromagnetic hopper vibration and simulation were carried out to obtain the mass flow rate curves of oil sunflower seeds under different working voltages. The correlation coefficient of the two curves was 0.998, indicating the changing trend was all the same. Consequently, the reliable model of oil sunflower seeds and the calibrated discrete element parameters can provide significant theoretical support to the discrete element simulation.