Artificial neural network prediction of minimum fluidization velocity for mixtures of biomass and inert solid particles

ANDRÉS REYES URRUTIA; JUAN PABLO CAPOSSIO; CÉSAR M. VENIER; ERICK DAVID TORRES; ROSA ANA RODRIGUEZ; GERMÁN MAZZA

FLUIDS

Multidisciplinary Digital Publishing Institute ? MDPI

Lugar: Basilea; Año: 2023 vol. 8 p. 1 - 18

The fluidization of certain biomasses used in thermal processes, like sawdust, is particularly difficult dueto their irregular shapes, varied sizes, and low densities, causing highminimum fluidization velocities (U_mf). The addition of an inertmaterial causes its U_mf to drop significantly. The determination ofthe U_mf of the binary mixture is however hard to obtain. Generally,predictive correlations are based on a small number of specific experiments, andsphericity is seldom included. In the present work, three models, i.e., anempirical correlation and two artificial neural networks (ANN) models were usedto predict the U_mf of biomass-inert mixtures. An extensivebibliographical survey of more than 200 datasets was conducted, with completedata about densities, particle diameters, sphericities, biomass fraction, and U_mf.With the combined application of the Partial Dependence Plot (PDP) and the ANNmodels, the average effect of sphericity on U_mf was quantitatively determined(inverse relationship) together with the average impact of the biomass fractionon U_mf (direct relationship). In comparison with the empiricalcorrelations, results showed that both ANN models can accurately predict the U_mfof the presented binary mixtures with errors lower than 25%.