Establishment of a Discrete-Element Model and Verification of a Seeder Bench Experiment for Sesbania Seeds

In light of the absence of discrete-element simulation characteristic parameters for the precision-seeding simulation process associated with Sesbania, this study investigated the physical properties of “CAS Sesbania No. 6” Sesbania seeds and measured the characteristic parameters of the Sesbania seeds in order to build a discrete-element model. The inter-seed contact parameters were calibrated to improve the seed model’s accuracy through stacking experiments and Box–Behnken response surface methodology. The stacking angle of Sesbania seeds was determined to be 34.5°; this was used as the evaluation index when simulating the stacking experiments. The results indicated that the coefficient of collision recovery, static friction, and rolling friction (inter-seed) were 0.168, 0.339, and 0.083, respectively. Substituting the calibrated parameters into the model, the simulation yielded a simulated stacking angle of 34.9°. The simulation results and bench experiment outcomes were compared to validate the applicability of the Sesbania seed discrete-element model in the context of seeder-simulation experiments. The results indicated that the percentage errors for the replay-seeding rate and missed-seeding rate between the simulation experiment and the bench experiment were 5.90% and 5.84%, respectively. All test results satisfied the technical requirements, confirming the calibration results’ reliability and providing a theoretical basis for the optimization of the design and simulation of Sesbania precision-seeding devices.