Search

Purpose |!|#8211; The purpose of this paper is to present a novel computational framework capable of simulating the block cave phenomenon of fines migration in two dimensions. Fines migration is characterised by the faster movement of fine and often low-grade material towards the draw point in comparison to larger, blocky material. A greater understanding of the kinematic behaviour of fines and ore within the cave during draw is integral to the solution of this problem. Design/methodology/approach |!|#8211; The lattice Boltzmann method (LBM) is employed in a nonlinear form to represent the fines as a continuum, and it is coupled to the discrete element method (DEM) which is used to represent large blocks. The issues relevant to this approach, such as fluid-solid interaction, the synchronisation of explicit schemes, and the characterisation of a bulk material as a non-Newtonian fluid are discussed. Findings |!|#8211; Results of the 2D simulations reveal migration trends for the geometries, material properties and operational sequences analysed. By executing an extensive programme of numerical experiments the influence of these and other relevant block cave factors on the migration of fines could be isolated. Originality/value |!|#8211; To the authors|!|#39; knowledge, this is the first time the LBM has been used to simulate the flow of bulk materials. The non-Newtonian LBM-DEM framework is also a novel approach to the investigation of fines migration, which until now has been limited to scale models, cellular automata or pure DEM simulations. The results of the 2D migration analyses highlight the potential for this novel approach to be applied in an industrial context and also encourage the extension of the framework to 3D. [ABSTRACT FROM AUTHOR]

Purpose |!|#8211; The purpose of this paper is to present a novel computational framework based on the lattice Boltzmann method (LBM) and discrete element method (DEM) capable of simulating fines migration in three dimensions. Fines migration occurs in a block cave mine, and is characterised by the faster movement of fine and often low-grade material towards the draw point in comparison to larger, blocky material. Design/methodology/approach |!|#8211; This study builds on the foundations and applications outlined in a companion paper, in which the non-Newtonian LBM-DEM framework is defined and applied in 2D simulations. Issues relevant to the extension to 3D, such as spatial discretisation, fluid boundary conditions and the definition of synthetic bulk material parameters using a power law model, are discussed. Findings |!|#8211; The results of the 3D DEM percolation replication showed that migration is predominantly limited to within the draw zone, and that the use of a low-cohesion material model resulted in a greater amount of fines migration. The draw sensitivity investigation undertaken with the two bell partial block cave analysis did not show a significant difference in the amount of migration, despite the two draw strategies being deliberately chosen to result in isolated and interactive draw of material. Originality/value |!|#8211; Along with the companion paper, this paper presents a novel application of the developed non-Newtonian LBM-DEM framework in the investigation of fines migration, which until now has been limited to scale models, cellular automata or pure DEM simulations. The results highlight the potential for this approach to be applied in an industrial context, and indicate a number of potential avenues for further research. [ABSTRACT FROM AUTHOR]

This paper outlines a dynamic domain decomposition-based parallel strategy for combined finite/discrete element analysis of multi-fracturing solids and discrete systems. Attention is focused on the parallelised interaction detection between discrete objects. Two graph representation models for discrete objects in contact are proposed which lay the foundation of the current development. In addition, a load imbalance detection and re-balancing scheme is also suggested to enhance the parallel performance. Finally, numerical examples are provided to illustrate the parallel performance achieved with the current implementation. [ABSTRACT FROM AUTHOR]