DETERMINATION OF PHYSICAL PROPERTIES AND CALIBRATION OF DISCRETE ELEMENT SIMULATION PARAMETERS FOR YIMENG MOUNTAIN JUJUBES

Accurate discrete element method (DEM) parameters are essential for simulating key operations in jujube postharvest processing, such as pneumatic cleaning during mechanical harvesting. The fundamental physical properties of Yimeng Mountain jujubes were systematically measured. A jujube particle template was constructed using the multi-sphere method, and the Hertz–Mindlin (no slip) contact model was adopted. The jujube simulation model was generated by filling the particle template with multiple spheres of different diameters. The static friction coefficient between jujubes ranged from 0.41 to 0.69, while that between jujubes and a steel plate ranged from 0.40 to 0.54. The rolling friction coefficient between jujubes ranged from 0.16 to 0.30, whereas that between jujubes and a steel plate ranged from 0.11 to 0.24. The coefficient of restitution between Yimeng Mountain jujubes, determined using a high-speed camera, ranged from 0.41 to 0.54, while that between jujubes and a steel plate ranged from 0.46 to 0.61. A Plackett–Burman screening design was conducted, and the results showed that the Poisson’s ratio of jujube, the static friction coefficient between jujubes, and the static friction coefficient between jujubes and the steel plate were the three dominant factors. The steepest ascent method was then used to narrow the ranges of these significant parameters. Subsequently, a Box–Behnken design was employed to establish a quadratic regression model relating the angle of repose to the three significant factors. Using the experimentally measured angle of repose of 28.08° as the optimization target, the optimal simulation parameters were determined as follows: a Poisson’s ratio of 0.32, a jujube–jujube static friction coefficient of 0.52, and a jujube–steel static friction coefficient of 0.47. The relative error between the simulated and experimentally measured angles of repose was 2.17%, further confirming the accuracy and reliability of the calibrated simulation parameters.