Artificial Neural Network Prediction of Minimum Fluidization Velocity for Mixtures of Biomass and Inert Solid Particles
Andres Reyes-Urrutia,
Juan Pablo Capossio,
Cesar Venier,
Erick Torres,
Rosa Rodriguez,
Germán Mazza
Affiliations
Andres Reyes-Urrutia
Institute for Research and Development in Process Engineering, Biotechnology and Alternative Energies, PROBIEN, CONICET-National University of Comahue, Buenos Aires 1400 St., Neuquén 8300, Argentina
Juan Pablo Capossio
Institute for Research and Development in Process Engineering, Biotechnology and Alternative Energies, PROBIEN, CONICET-National University of Comahue, Buenos Aires 1400 St., Neuquén 8300, Argentina
Cesar Venier
Research Center for Computational Methods, CIMEC, CONICET-National University of the Litoral, Paraje “El Pozo”, Santa Fe 3000, Argentina
Erick Torres
Chemical Engineering Institute, Faculty of Engineering, National University of San Juan, Research Group Associated with PROBIEN Institute, CONICET-National University of Comahue, San Juan 5400, Argentina
Rosa Rodriguez
Chemical Engineering Institute, Faculty of Engineering, National University of San Juan, Research Group Associated with PROBIEN Institute, CONICET-National University of Comahue, San Juan 5400, Argentina
Germán Mazza
Institute for Research and Development in Process Engineering, Biotechnology and Alternative Energies, PROBIEN, CONICET-National University of Comahue, Buenos Aires 1400 St., Neuquén 8300, Argentina
The fluidization of certain biomasses used in thermal processes, such as sawdust, is particularly difficult due to their irregular shapes, varied sizes, and low densities, causing high minimum fluidization velocities (Umf). The addition of an inert material causes its Umf to drop significantly. The determination of the Umf of the binary mixture is however hard to obtain. Generally, predictive correlations are based on a small number of specific experiments, and sphericity is seldom included. In the present work, three models, i.e., an empirical correlation and two artificial neural networks (ANN) models were used to predict the Umf of biomass-inert mixtures. An extensive bibliographical survey of more than 200 datasets was conducted with complete data about densities, particle diameters, sphericities, biomass fraction, and Umf. With the combined application of the partial dependence plot (PDP) and the ANN models, the average effect of sphericity on Umf was quantitatively determined (inverse relationship) together with the average impact of the biomass fraction on Umf (direct relationship). In comparison with the empirical correlations, the results showed that both ANN models can accurately predict the Umf of the presented binary mixtures with errors lower than 25%.