Formatted abstract

The Discrete Element Method (DEM) is a tool used to model the interactions between large numbers of small particles, and compute their resulting motion. It is well suited for use in the mining industry, to simulate mineral particles that are being processed and refined. There are many pieces of equipment in a mine that can be modelled using DEM, one of which is a rotary breakage mill. A rotary breakage mill is a machine that is used to reduce the size of particles, and to separate required minerals from discharge material. These mills are used in coal mines to remove coal from waste rock found in a coal seam.
A review of literature revealed that in Australia, there is not a recognised method for designing the geometry of a rotary breakage mill. In the past, the size and shape of these mills has been determined by using empirical methods which are known to work, or by consulting with the manufacturer. If the mill geometry is designed so that minimal coal exits the mill through the discharge chute, the efficiency of the mill increases. To do this, the coal particles must all break to a size small enough to be screened out of the mill before being discharged.
This thesis aims to investigate the use of DEM in modelling the breakage mechanisms inside a rotary breakage mill. To do this, dropshatter test data is analysed to find the energies required for breakage, over 15 drops. Using these energies, the required specific breakage energy for the sample of coal is found, which can be used to determine how much energy any particular particle requires to break.
DEM software Rocky was then used to find the time that any coal particle takes to travel through the mill, called a residence time. A Computer Aided Design (CAD) model of a mill is drawn and imported into Rocky, and a simulation of a full load of coal is run. From this, it was found that the particles took an average of 77 seconds to travel from the inlet chute to the outlet chute. A longitudinal slice is then taken from the model of the mill, and simulated using just one particle. This provides a simpler method of determining the energy supplied by the mill to the particles. The specific energies supplied to the particles for three different mill speeds are found, and the periods between each of the drops of the particle are calculated from the rotational speed of each mill. An approximation of breakage is taken based on the fact that when a particle breaks, resultant particles may still be too large to be screened. The required breakage energy is then used with the supplied energy to determine the number of drops required for breakage, and this is used with the period between drops to find the time required for breakage. Since some resultant particles are still too large to be screened, this time may increase if resultant particles need to break further.
It was found that as the size of the particles and the speed of the mill varies, the time required for breakage of the particles ranges between 26s and 138s. Since a residence time of 77s is known, these values indicate that depending on mill parameters, some coal particles would break and be successfully screened by the mill, while others would exit through the discharge chute. From these results, the parameters of the mill can be chosen so that the mill breaks the desired amount of coal in a given time.
This thesis shows that DEM can be used to investigate breakage in a mill, since the calculated times required for breakage are comparable to the residence time. This demonstrates the usefulness of DEM, and suggests that it can be developed in future as a design tool to help analyse material handling processes. For any continuation of this work, it is recommended that a physical model be included in the testing, such as a scale model of a slice of a mill. The inclusion of such a physical model would provide verification of the results found in this thesis, and could improve the DEM methodology used in mill geometry selection.
