A Hybrid PBM-DEM Model of High-Pressure Grinding Rolls Applied to Iron Ore Pellet Feed Pressing

Abstract

Mathematical models of high-pressure grinding rolls (HPGR) have attracted great attention, owing to their role in optimization of operating machines as well as in the design and selection of new ones. Although population balance models (PBM) and the discrete element method (DEM) have been used in this task, both suffer from important limitations. Whereas PBMs have challenges associated to the prediction of operating gap and to the validity of several of its assumptions in different formulations in the literature, application of DEM has its own challenges, in particular when fed with distributions containing large amounts of fines. This work proposes a hybrid approach in which the coupling of DEM to particle replacement models and multibody dynamics is used to predict operating gap, throughput and power, as well as providing information along the rolls length that is used in PBM to predict the product fineness. The hybrid approach is then compared to both DEM and a PBM (Modified Torres and Casali), demonstrating similar results to the later when applied to simulating a pilot-scale machine operating under different conditions, but improved prediction when applied in scale-up to an industrial-scale HPGR.