PyBONDEM-GPU: A discrete element bonded particle Python research framework – Development and examples

Discrete element modelling (DEM) is widely used to simulate granular systems, nowadays routinely on graphical processing units. Graphics units (GPUs) are inherently designed for parallel computation, and recent advances in the architecture, compiler design language development allowing general-purpose computation be computed multiple GPUs. Application of DEM bonded particle systems much less common, with a number open research questions remaining. This study outlines Bonded-Particle Research Framework, PyBONDEM-GPU, written Python. framework leverages nature GPUs computational speed-up rapid prototype flexibility Python faster easier learn than classical compiled languages, making simulation accessible undergraduate graduate engineers. PyBONDEMGPU Numba-CUDA module compile syntax execution The enables fibre pull-out from fibre-matrix embeddings. Bonds simulated between all interacting particles. performance PyBONDEM-GPU compared against CPU implementations PyBONDEM using Numpy Numba-CPU modules. was found 1000 times implementation 4 resolve forces integrate equations motion.