Predicting the pressure drop of a filter is necessary, especially regarding the jump in the dynamic pressure drop of a filter without pre-filters when clogged by rain or fog droplets. However, the efficient prediction method for the pressure drop in cylinder filters loading with water mists in the pore network faces limitations due to the high requirements for microscopic images of materials and the need to consider for the increasing holding droplet capacity of filters. In this study, a systematic method that takes into account the distributions of the deposited droplet mass within the water-absorbed filter medium layers in the pore network model (PNM) was developed. The pore geometry and distributions were determined from 2D microstructure images, and the dynamic deposited droplet mass and water-absorbing capacity of fibers were combined with the changing saturation to optimize the PNM for calculating the dynamic pressure drop of the filter media. Specifically, the empirical equation of the self-defined equivalent single fiber filtration efficiency (ESFFE) describes the relationship between the changing single fiber filtration efficiency and the deposited droplet mass. The optimized PNM is suitable for predicting the behavior of bibulous wood cellulose fiber filter media. The dynamic pressure drop shows a slight increase before surging, following the typical “channel and jump” model. Although the calculations deviate slightly from the experimental results with 25 % deviations, this systematic method effectively predict the jump in pressure drop of the tested filter medium. Therefore, this systematic method for predicting pressure drop of filters holds promise in anticipating unsafe pressure drop jumps in filters installed in coastal areas.
Read full abstract