A hybrid LB-DPM method to predict collection efficiency of woven screens for microparticles and nanoparticles

碩士 === 國立中山大學 === 機械與機電工程學系研究所 === 106 === The present study uses a 2D hybrid lattice Boltzmann-discrete phase model method to predict screen collection efficiency in woven wire screen. The computational results are used to compare with published experimental data that have not been computationally...

Full description

Bibliographic Details
Main Authors: Tsung-hsien Yu, 余宗憲
Other Authors: Kuang C. Lin
Format: Others
Language:en_US
Published: 2018
Online Access:http://ndltd.ncl.edu.tw/handle/dn3cuy
Description
Summary:碩士 === 國立中山大學 === 機械與機電工程學系研究所 === 106 === The present study uses a 2D hybrid lattice Boltzmann-discrete phase model method to predict screen collection efficiency in woven wire screen. The computational results are used to compare with published experimental data that have not been computationally verified in the literature. The filtration processes are investigated for particles that are micron (0.3-100 μm) and nanometer (3-20 nm) in size. This thesis is composed of two subjects. The first study analyzes the filtration of microparticles at gas flow velocities of 0.48 cm/s via mesh screens with five different pore sizes in the range of 11-160 μm. The second study investigates the filtration of nanoparticles via stainless steel mesh screen with a pore size of 126 μm, where the filtration process is conducted with gas flow velocities of 4.17, 5.63 and 7.04 cm/s. Besides, the thermal effect on the filtration performance is revealed in the second study by varying the gas temperature (296K, 400K and 500K). For these two subjects, the computational domain contains a circular fiber that represents the cross-section of a semi-infinite fiber array. In order to simulate the 3-D woven screen structures, several 2-D models are proposed using different physical specifications of mesh screens including open area, pore size and packing density. The results show that the data predicted with a modified pore-size-based model are able to show excellent agreement with experiments. In addition, the particle distributions along the fiber surface are used to estimate the particle capture mechanisms together with the deposition patterns in the corresponding operational conditions.