Recurrence CFD simulations of time-evolving particle size segregation in gas-solid fluidized beds

We numerically assess time-evolving particle size segregation in a lab-scale gas-solid fluidized bed. Based on expensive short-
term unresolved CFD-DEM simulations we deduce characteristic flow patterns for extremely fast data-based recurrence CFD
(rCFD) simulations.
For that purpose, we further developed rCFD to account for segregating particle fractions, hence extrapolating short-term
databases not only in time but also in terms of poly-dispersity. While for the case of mild segregation rCFD predictions agree
very well with corresponding CFD-DEM simulations, the method fails for more substantial segregation.
In terms of computational performance, rCFD simulations are more than four orders of magnitude faster than corresponding
CFD-DEM simulations, eventually allowing for real-time simulations of segregation in poly-disperse fluidized beds.
Keywords: gas-solid fluidized bed, recurrence CFD, particle size segregation