Modelling segregation phenomena in large industrial silos: A cellular automaton approach

Segregation presents a significant challenge in the handling of bulk materials across various industries, including the handling of wood pellets. Wood pellets, which degrade over time, develop a wide particle size distribution, leading to increased segregation during handling. This can result in fines and dust spikes in silo discharge streams, negatively affecting operational efficiency and safety. Accurate prediction of segregation during silo filling and discharging is critical for ensuring safe handling and efficient operations. To address these challenges, this study develops and validates a Cellular Automata (CA) model to simulate segregation in wood pellet silos. The CA approach provides computational efficiency while capturing the essential physics of particle behaviour. The model was initially calibrated through laboratory experiments and subsequently validated against data from a 2D glass-walled silo. Following successful validation, a 3D CA model was developed and tested against industrial-scale wood pellet silos. The model demonstrated an accurate prediction of segregation patterns and fines content in discharge streams, offering a valuable tool for optimising silo operations and mitigating associated risks. • Segregation is a prevalent issue in wood pellet handling. • A Cellular Automaton model was developed to simulate segregation within silos. • Model predictions were validated against data from laboratory and industrial silos. • The model demonstrated strong correlation with experimental observations.