Abstract – The flooding velocity is an important but difficult
to accurately predict parameter for the packed column design.
With the appearance of new packing shapes, traditional
empirical models are insufficient to satisfy the requirement of
engineering applications. In this paper, a novel approach using
least squares-support vector machine (LS-SVM) is proposed to
predict the flooding velocity in the randomly dumped packed
towers. To evaluate the performance of the LS-SVM model
applied to predict the flooding velocity, it is compared with the
traditional empirical models and the neural network models. It is
found that the LS-SVM model can provide the best performance
of all, with an average absolute relative error less than 8 %. The
results demonstrate that LS-SVM offers an alternative approach
to model and predict the flooding velocity in the randomly
dumped packed towers.