Computational simulation of soybean particles flow in a hopper using computational fluid dynamics (CFD) and discrete elements method (DEM)

Abstract

The hoppers are the most common structures used in storage units for agricultural products such as grains and cereals. The soybean, which is one of the most common products in Brazil spend most of their time in a hopper between the stages of picking and shipment. Problems such as damage to the hopper structures during the outflow are factors that have been the subject of studies using computational models. Computational Fluid Dynamics (CFD) has played a big role in gas-solid systems study, together with the Discrete Element Method (DEM). This method manages both fluid phase as the solid phase, which in this case is granular, through the Eulerian and Lagrangian approach. The DEM is based on the interaction between the particles and each one is separately monitored. This work aims to calibrate the parameters of the spring-dashpot model, in the granular dynamics of fluids study, which influences the contact between the soy particles in the silo. For this purpose, a comparison was made of the experimental discharge time of soybeans into a hopper, with the time resulting from 27 simulations generated by a central composite design (CCD). Through the analysis of the simulations and statistics, it was possible to identify the factors that influence whether or not the time of discharge and establish a calibration of these parameters that best describe the experimental results.