DEM Modelling and Simulation of Cone Crushers and High Pressure Grinding Rolls

Doctoral thesis, 2017

The comminution of rock and ore materials consumes ~1.5-1.8 % of the total energy production in mining intensive countries (Tromans, 2008). Several research findings show that there are ways of utilizing compressive breakage modes that are more energy efficient compared to conventional grinding circuits based on large inefficient tumbling mills. Circuits using Cone Crushers and High Pressure Grinding Rolls (HPGR) have proven to be more energy efficient. These comminution devices have during the last two decades been implemented for hard rock materials. These machines are hence suitable subjects for further performance improvement and optimization. In this thesis a simulation platform, based on the Discrete Element Method (DEM), for simulation of compressive breakage machines such as Cone Crushers and HPGRs is presented. The research notion is that in order to further develop compressive breakage machines and operations, fundamental understanding is needed with regard to specific details inside the machines. The rock particles are modelled using the Bonded Particle Model (BPM) and particle shapes are based on 3D scanned rocks. The machine geometry is based on CAD modelling and 3D scanning. The interactions between rock particles and between rock particles and the machine boundaries are modelled using contact models that determine the reaction forces. A novel DEM calibration and validation framework based on design of experiments, surrogate modelling and multi-objective optimization has been developed for calibrating and validating particle flow and breakage. Seven different simulation case studies are included in the thesis work.  The simulation and modelling capabilities have successively progressed for each study conducted. The result and findings are both attributed to machine specific insights and generic modelling findings. The work shows that the most vital machine and process performance responses, such as product size distribution, pressure distributions and power draw can be predicted using the DEM simulation platform.