Please login to view abstract download link
Vertically stirred regrind mills are used in the mining process industry to improve the process yield. The sizing and analysis have been done mostly by experience from field data and by scaling. This study presents the first steps to investigate the grinding process with the discrete element method (DEM) Yet, the DEM is used to compare different rotor designs of high intensity grinding mills in mining [1] and vertical stirred mills in the pharmaceutical industry [2]. Mostly the simulations are used to make relative comparisons with best guess material parameters. The aim of this project is to simulate the real mill behaviour with a coupled DEM and Computational Fluid Dynamics (CFD) simulation to gain knowledge of the dynamic behaviour of the grinding process and predict power draw and efficiency of various mill scales. Figure 1 shows a two-way coupled CFD-DEM simulation with a fluid inlet below the rotor tip. The particles are dragged upwards centrally along the rotor (red particles). An important part of the project is to calibrate the material parameter used for the DEM contact models. Therefore, certain values are measured and the rolling resistance and adhesion between the particles are calibrated using a surface response approach as shown on Figure 2. The green area shows the target value of the dynamic angle of repose determined from experiments. This study gives insight of the calibration of the bulk mass behaviour with simple methods that can be transferred to industrial applications. The calibrated model leads to better results of the real process, which is validated via experiments on a test mill. The results of the study are used to predict the behaviour of large-scale mills.