No abstract

Mist elimination is a widely used separation technique in many industrial processes. Mist elimination is commonly carried out using knitted wires, and the droplet impact and capture on these wires are governed by the impact hydrodynamics. Predicting the maximum impact velocity that results in the droplet being captured on a horizontal, stationary wire (V t,max ) has previously been based on the solution of the force balance equation over an impacting droplet. However, these correlations were often based on varying simplifications, particularly in using a constant value of drag coefficient (C d ) in calculating the drag force in the force balance equations. The result of such simplifications has contributed to the discrepancy between predictions and experimental data ranging from 15 to 40%. The current work proposes a new analytical model with a correlation that was developed to predict the C d as a function of fundamental physical properties of the fluid (the ratio of viscosities) and operating parameters (the ratio of diameters). Droplet capture experiments with different wire materials and different wire and droplet sizes were also carried out. High-speed imaging was used to measure the V t,max . Results from the experiment were used to estimate the parameters in the newly developed correlation for the variable C d using a hybrid genetic algorithm (GA) and Levenberg−Marquadt algorithm (LMA) method. The new model was tested for a wide range of experimental data. It was observed that the new model was able to predict the experimental data more accurately than previously reported predictions. The model was also used to investigate the effect of wire size and fluid properties on the V t,max .