Minerals (Nov 2023)

Discrete Element Modelling of a Bulk Cohesive Material Discharging from a Conveyor Belt onto an Impact Plate

  • Otto C. Scheffler,
  • Corné J. Coetzee

DOI
https://doi.org/10.3390/min13121501
Journal volume & issue
Vol. 13, no. 12
p. 1501

Abstract

Read online

The discrete element method (DEM) has become the numerical method of choice for analysing and predicting the behaviour of granular materials in bulk handling systems. Wet-and-sticky materials (WSM) are especially problematic, resulting in build-up and blockages. Furthermore, due to the large number of particles in industrial-scale applications, it is essential to decrease the number of particles in the model by increasing their size (upscaling or coarse graining). In this study, the accuracy with which upscaled DEM particles can model the discharge of a cohesive material from a belt conveyor onto an inclined impact plate was investigated. Experimentally, three sand grades (particle size distributions, PSDs) were used, each in a dry (non-cohesive) state and with three levels of moisture-induced cohesion. The effects of the modelled PSDs on the material flow, build-up on the plate, the peak impact force and the residual weight were investigated. Although a linear cohesion contact model was mostly used, the results were also compared to that of the Johnson–Kendall–Roberts (JKR) and simplified JKR (SJKR) models. It was found that the general profile of the pile (build-up) could be accurately modelled, but using a more accurate (but still upscaled) PSD improved the results. The impact force and the residual weight on the plate could be accurately modelled (error 15%) if the particle size was not excessively scaled. The maximum acceptable scaling factor was found to be a geometric factor of the bulk measure of interest, and not a factor of the physical particle size. Furthermore, with an increase in cohesion, the bulk measures such as the thickness of the discharge stream and the height of the material build-up increased, which meant that the maximum acceptable scale factor also increased. The results are valuable for future accurate and efficient modelling of large industrial scale applications of WSMs.

Keywords