From discrete element simulation data to process insights

Industrial-scale discrete element simulations typically generate Gigabytes of data per time step, which implies that even opening a single file may require 5 - 15 minutes on conventional magnetic storage devices. Data science’s inherent multi-disciplinary nature makes the extraction useful information challenging, often leading to undiscovered details or new insights. This study explores potential statistical learning identify regions interest for large scale simulations. We demonstrate our in-house knowledge discovery and mining system (KDS) can decompose datasets into i) analyst, ii) multiple decompositions highlight different aspects data, iii) simplify interpretation DEM generated by focusing attention automatically decomposed regions, iv) streamline analysis raw letting analyst control number decomposition way are performed. Multiple be automated in parallel compressed, enabling agile engagement with analyst’s processed data. focuses spatial not temporal inferences.