Experimental results on the flooding capacity of randomly dumped packed bed
s were collected from the literature to generate a working database. The re
ported measurements were first used to review the accuracy of existing pred
ictive tools in that field. A total of 14 correlations were extracted from
the literature and cross-examined with the database. Many limitations regar
ding the level of accuracy and generalization came to light with this inves
tigation. Artificial neural network modeling was then proposed to improve t
he broadness and accuracy in predicting the flooding capacity, which is an
important design parameter for packed towers. A combination of six dimensio
nless groups, namely, the Lockhart-Martinelli parameter (chi); the liquid R
eynolds (Re-L), Galileo (Ga-L) and Stokes (St(L)) numbers; the packing sphe
ricity (phi); and one bed number (SB) outlining the tower dimensions were u
sed as the basis of the neural network correlation. With an initial databas
e containing 1019 measurements, the correlation yielded an absolute average
relative error (AARE) of 16.1% and a standard deviation of 20.4%. Another
database containing over 100 measurements on the flooding capacity was used
to validate the correlation. The prediction based on these results yielded
an AARE of 11.6% and a standard deviation of 13.7%. Through a sensitivity
analysis, the Stokes number in the liquid phase was found to exhibit the st
rongest influence on the prediction, while the liquid velocity, gas density
, and packing shape factor were determined to be the leading physical prope
rties defining the flooding level. As a matter of fact, the neural correlat
ion remains in accordance with the design recommendations and trends report
ed in the literature.