Development and evaluation of a pneumatic grain separation system using CFD and DEM

Abstract

As the global demand for food continues to increase, grain and seed cleaning equipment has to evolve along with it. The demand for efficient grain and seed cleaning equipment is rising, underscoring its crucial role in maintaining the quality and efficiency of agricultural processes. This equipment is essential for enhancing grain and seed quality by removing impurities like chaff, toxic seeds, and dust, thereby reducing spoilage and increasing market value. Pneumatic grain separators are particularly effective in removing these lightweight impurities, operating on principles of airflow and density to separate contaminants from grain. Despite the industry's reliance on trial-and-error methods, this study aims to examine the hydrodynamics of grain particles in pneumatic separators to contribute and assist in the development of more efficient cleaning equipment, ultimately benefiting the agricultural and food processing sectors. This study seeks to leverage the precision of numerical modelling techniques like Computational Fluid Dynamics (CFD) and Discrete Element Method (DEM), as well as coupled CFD-DEM simulations, despite the inherent complexity and challenges in their implementation to aid in both the development and increased performance of pneumatic separators. This study is stratified into two approaches: (a) concept development using CFD simulations to simulate the airflow, pressure drop, and recirculating phenomena within the aspirator and (b) aspirator analysis and optimisation using coupled CFD-DEM simulations. This research employs numerical modelling techniques to design and optimise pneumatic grain separators for agricultural and industrial applications. Utilising ANSYS Fluent® for fluid flow and Rocky® DEM for particle dynamics, the study offers a multiphase two-way coupling approach to investigate particle hydrodynamics. This study addresses a gap in the existing literature and presents an integrated methodology with the potential to transform the agricultural and food processing industry. This study investigated the development and testing of a grain aspirator and the validation of CFD, DEM and coupled CFD-DEM simulations. In the first approach, a horizontal pneumatic grain aspirator was simulated, designed, manufactured, and evaluated for cleaning maize and soya. Utilising ANSYS Fluent® simulations and real-world prototype testing, the aspirator's design was optimised for efficient impurity separation. The final model achieved a 95.9% efficiency rate, offering a reliable method to enhance grain cleaning in agriculture and the food processing industry. The second approach employed a two-way coupled CFD-DEM simulation in the ANSYS-Rocky® framework to optimise a grain aspirator system. A two-tiered validation approach was used, comparing the simulation's outputs to empirical data and experimental results. The simulation results are in good comparison to the experimental results, with a cleaning efficiency error of only 4.69% and a cleaning loss error of 6.18%. This validated coupled CFD-DEM methodology optimised the grain aspirator and provided key insights for a more time and cost-efficient design process. Therefore, this approach can form the basis for designing and optimising complex hydrodynamic systems.