Improved Liquid Hold-up Correlation for Randomly Packed Towers

摘要: The state-of-the-art tools for the evaluation of total liquid holdup in gas-liquid counter-current randomly dumped packed beds are critically evaluated by thoroughly interrogating a wide hydrodynamic database. This database, consisting ca. 1,500 experiments on hold-up below flooding point, represents an important portion non-proprietary information released literature since 1930s. Providing access to diversified information, it is dedicated embracing wide-ranging temperature and gas density levels, packing shapes extending from classical ones modern third generation packings. Furthermore, eleven correlations extracted cross-examined using Many limitations regarding level accuracy generalization come light with this investigation. Artificial neural network modelling dimensional analysis then proposed improve predicting pre-loading loading regions beds. A combination five dimensionless groups, comprising Reynolds (ReL), Froude (FrL), Ohnesorge (OhL) numbers as well (FrG) Stokes (StG) used basis correlation. correlation yields absolute average relative error 14% whole database remains accordance trends reported literature.