Development of Effective Voidage Correlations in Pilot-Scale Liquid-Solid Fluidized Bed Based on Data-Driven Modeling

Liquid-solid fluidized beds can meet many differentdemandsfrom chemical, environmental, and pharmaceutical processes, and itis very crucial to accurately predict their voidages, which, however,are usually underestimated by the classical Richardson-Zakiequation. In this study, a data-driven modeling method was appliedto develop the voidage correlation based on the experimental datafrom pilot-scale liquid-solid fluidized bed where the fluidizationexperiments of spherical glass beads were performed at different particlesizes, particle densities, liquid superficial velocities, and operatingtemperatures. By selecting different parametric spaces, various high-accuracyvoidage correlations based on different dimensionless numbers (i.e.,Re, Re-t, Ar, Fr, St) were discovered and all of the valuesof R (2) were approximately 0.99. In comparison,the combination of Re and Ar numbers could better characterize thevoidage and the mean square error was as low as 2.29 x 10(-4). The proposed correlation also performed very wellin predicting the voidage of the fluidization of irregular calcitepellets in a pilot-scale fluidized bed and the voidage of the fluidizationof coarse coal particles in a laboratory-scale fluidized bed, thusindicating its universality.

Automated summary

A data-driven modeling method was used to develop a voidage correlation for liquid-solid fluidized beds, which accurately predicts the voidages. Different parametric spaces were selected, and various high-accuracy voidage correlations based on different dimensionless numbers were discovered. The combination of Re and Ar numbers performed the best, with a mean square error as low as 2.29 x 10(-4). The proposed correlation also showed good performance in predicting the voidage of different particles in fluidized beds, indicating its universality.