Application of neural networks to mass-transfer predictions in a fast fluidized bed of fine solids

✍ Scribed by Piroz Zamankhan; Pekka Malinen; Hannu Lepomäki

• Publisher: American Institute of Chemical Engineers
• Year: 1997
• Tongue: English
• Weight: 626 KB
• Volume: 43
• Category: Article
• ISSN: 0001-1541
• DOI: 10.1002/aic.690430705

✦ Synopsis

In this study back-propagation, feed-forward neural networks are applied to estimate mass-transfer parameters in fast fluidized beds of fine solids. These networks are trained to predict mass-transfer rates using measurements of the sublimation rate of coarse naphthalene balls in fast fluidized beds of fine glass beads at several solid-to-gas mass flow rates within the relevant superficial gas-velocity range. When tested to predict the effective difisivities from a coarse particle to the bulk of the fast bed of fine solids, trained neural networks calculated the Sherwood number with high accuracy. It is demonstrated that back-propagation, feed-forward neural networks provide a more accurate correlation for the mass-transfer coeficient compared to those obtained by the currently used heuristic models.