Design and optimization of a safflower filament cyclone collection device based on CFD-DEM coupling method

To address the bottleneck problems of low collection efficiency and high impurity content in the mechanical harvesting of safflower filaments, this study takes Xinjiang dryland safflower as the research object and designs an efficient cleaning system for safflower filaments based on the cyclone separation principle. To optimize system performance, this study integrates Computational Fluid Dynamics and Discrete Element Method (CFD-DEM) coupled simulations with the Response Surface Methodology (RSM), systematically investigating the effects of three key parameters—volute wrap angle, separation drum diameter, and inlet air velocity—on cleaning performance. Through coupled simulations, the internal flow field characteristics (indicated by pressure drop and resultant velocity) and the movement trajectories of the filaments were thoroughly analyzed. Based on a three-factor, three-level response surface experimental design, accurate regression models for the collection rate and impurity rate of the filaments were established, and the interactive effects among the parameters were revealed. Optimization results show that under the optimal parameter combination of a 180° volute wrap angle, 140 mm separation drum diameter, and 2.5 m/s inlet air velocity, the cleaning performance is optimal, with a measured filament collection rate of 98.53% and an impurity rate of 2.30%, which are in close agreement with the model predictions (relative error <5%). This study not only verifies the effectiveness and accuracy of the CFD-DEM coupled method in the design of agricultural material cleaning equipment but also provides a reliable theoretical basis and technical solution for the efficient and low-loss mechanical harvesting of safflower filaments.