To improve the accuracy of the parameters used in discrete element simulation tests in the process of clearing fresh sunflowers, this study took the extracts of fresh sunflower as the research object and used a combination of physical experiments and simulation experiments to calibrate the discrete element simulation parameters. First, the composition of the edible sunflower extract is determined, and then, the physical test method is used to determine the characteristics of the edible sunflower extract material. Based on the results of the physical test as a simulation basis, the Plackett–Burman test is used to screen the significance of the test parameters, and the screening results show that the shear modulus, static friction coefficient, and collision recovery coefficient of sunflower kernels have significant effects on the simulated angle of repose. On the basis of the steepest climbing test, the Box–Behnken test is carried out to obtain the simulated angle of repose and significance. The second-order regression equation of texture parameters is optimized with the actual physical angle of repose (24.68°) as the target value to obtain the optimal parameter combination. Finally, through the two-sample T test, p > 0.05 is obtained, indicating that the simulated angle of repose and the actual physical test angle of repose are not significantly different, and the relative error is 0.923%, which verifies the reliability of the optimal simulation parameter combination. The research results show that the calibrated simulation parameter combination can be used in discrete element simulation tests of the extraction process of sunflower extracts.

Existing cleaning devices for edible sunflower have a low cleaning efficiency, high cleaning loss rate, and high impurity rate; therefore, a wind-sieve-type cleaning device for edible sunflower harvesting was designed. According to the characteristics of dislodged objects, a vibrating screen for the device was designed, and the dislodged edible sunflower objects in the device were used for a mechanical analysis of the force conditions to determine the displacement of the different edible sunflower objects dislodged by the action of airflow. Using FLUENT-DEM gas–solid coupling simulation technology, the velocity of the flow field, the velocity vector, and the trajectory of the dislodged objects inside the cleaning device were analyzed, and the law of motion applied to the airflow and the dislodged objects inside the device was clarified. According to the results of the coupled simulation analysis, the key factors affecting the operation of the cleaning device were wind speed, vibration frequency, and amplitude. Based on the key factors of wind speed, vibration frequency, and amplitude, an orthogonal rotary combination test was carried out with the loss rate and impurity rate of cleaned grains as the evaluation indexes, and the test parameters were optimized to obtain the optimal combination of operating parameters of the device, which were as follows: wind speed: 30 m·s⁻¹; vibration frequency: 8.44 Hz; and amplitude: 41.35 mm. With this combination of parameters, the seed loss rate and impurity rate reached 3.47% and 6.17%, respectively. Based on the optimal combination of operating parameters, a validation test was performed, and the results of this test were compared with the results of the test bench using this combination of parameters. The results show that the relative errors of the loss rate and impurity rate between the bench test and the simulation test were 3.45% and 3.07%, respectively, which are less than 5%, proving the reliability of the simulation analysis and the reasonableness of the design of the test bench.