Multi-layered mesh generation and user subroutine development based on an Eulerian–Eulerian approach for soot filtration visualization in a three-dimensional particulate filter model

Abstract

A three-dimensional model is developed for computational fluid dynamics analysis of soot filtration processes in a wall flow–type particulate filter based on an Eulerian–Eulerian numerical approach. The primary objective of this study is to accurately capture the local values of filtration parameters, such as volume porosity, collection efficiency, and soot mass deposited, through isotropically discretized computational grids within the multi-layered porous wall regions. Most commercial computational fluid dynamics codes do not have the ability to generate structured mesh with ordered cell index or allow expressing mathematical recursive operation or handling time array through defined user field functions. For these reasons, it is difficult to utilize the code in situations where complex algorithms and mathematical treatments are required. Therefore, custom-build user subroutines, written with C++ programming language, are developed and integrated with the model to calculate localized soot mass distributions in each layer of porous wall. New recursive and computationally efficient algorithms are developed utilizing the functional attributes of the computational fluid dynamics code and coupled with the modified unit collector mechanism in order to obtain temporal and local filtration efficiency. Simulations enabled quantitative visualization of the detailed filtration processes along the filter wall and channel, including the time evolution of filtration parameters, which is difficult to detect experimentally. The model revealed correlations between wall flow pattern and rearrangement behaviors of filtration parameters and provided insights into the soot cake layer profiles with respect to both the length and the width of the channel.