Study on collaborative optimization of vibration parameters and efficient screening of pumpkin seed based on discrete element method

Pumpkin seeds are widely valued for their nutritional and economic benefits; however, conventional grain separators used in centralized processing often suffer from low efficiency and high loss ratio due to the similar physical properties of seeds and pulp fragments. Furthermore, while discrete element method (DEM) simulations are widely used for generic vibrating screens, targeted theoretical and structural design studies addressing the unique material characteristics of pumpkin seeds remain scarce. To address this gap, this study systematically analyzes the relative motion constraints of particles on a vibrating screen to identify key factors governing screening behavior. Subsequently, a DEM model was established to simulate particle motion on a specialized double-layer rectangular perforated screen. A three-factor, three-level Box–Behnken response surface design was employed to construct regression models evaluating the interactive effects of vibration amplitude, deck inclination, and vibration frequency. Numerical optimization identified the optimum operating parameters as a 26 mm vibration amplitude, 2.5° deck inclination, and 4.0 Hz vibration frequency, yielding a predicted dynamic screening efficiency of 88.7% and a loss ratio of approximately 3.7%. Subsequent bench tests conducted under these optimized conditions demonstrated high consistency with DEM predictions, exhibiting small and stable deviations. Ultimately, unlike existing generic screening studies, this research provides a tailored equipment design paradigm and a collaborative parameter optimization framework specifically for pumpkin seeds, offering a reliable approach to resolve systemic bottlenecks in centralized processing.