Flow Simulation and Identification of Important Model Parameters in Industrial Packed Beds for High-Performance Random Packings

The goals of this work were: first, to simulate the liquid flow distribution in a large diameter (1.2 m) packed column with an RMSR 70-5 high performance random packing (layer height up to 3 m), by a dispersion model. Second, to find and estimate the important model parameters and flow maldistribution factor, using experimental data and two different optimization approaches. A three-parameter dispersion model for prediction of radial liquid distribution and two different approaches to determine some of the model parameters from experimental data were used. In parallel, a two-parameter optimization procedure for model parameters identification was performed based on the minimum of residual variance between model and experimental liquid velocities over a column crosssection. The simulated and experimental flow maldistribution, were estimated by means of an integral estimation – a maldistribution factor. The comparison between the model and experimental liquid distribution and respective maldistribution factors at packing heights H = 1 m and H = 2.5 m for liquid load 16.6∙10-3 m3/m2∙s showed very good agreement, even for a high packing layer. In conclusion, the presented model predictions and estimations about RSRM 70-packing 5 characteristics and behaviour will complement the information about its efficiency and operation in industrial processes.