Analysis of Flow Pattern in a Gasoline Particulate Filter using CFD

• Author / Creator

  Vega Mesquida, Ileana Magdalena

• The rise in motor vehicle ownership has become not only an environmental issue, but also a public health threat as pollutants emitted negatively affect human health (Deng et al., 2017). One such pollutant is particulate matter, which has been regulated by a variety of countries because it aggravates respiratory diseases such as chronic obstructive disease (Ling and van Eeden, 2009). Therefore, the automotive industry has implemented particulate filters in their products in an effort to capture these particles. This thesis analyzes flow characteristics within a representative element of a gasoline particulate filter that consists of one inlet channel, surrounded by outlet channels. The analysis considers a flow without particles that enters the inlet channel, crosses the porous walls and reaches the filter exit through outlet channels. The flow pattern is analyzed in two-dimensional and three-dimensional domains. To complete such analysis, computer fluid dynamics (ANSYS-Fluent) is applied using laminar, k-ε, k-ω and SST- transition models. Further, experimental data is compared
  with numerical pressure drop to determine which model best describes the flow pattern within
  a gasoline particulate filter. Additionally, the impact of including an inlet and outlet zone to quantify the effects of contraction and expansion is evaluated.
  This work concludes the possible existence of turbulence in the channels of a gasoline particulate monolith. Although two-dimensional and three dimensional domains are solved using different parameters, both k-ε and SST-transition models prognosticate turbulence in the channels for Reynolds number above 220 in the channel. Nevertheless, the SST-transition
  model predicts turbulence at the end of the inlet channel in both domains, whereas the k-ε model only does so in 3D when the Reynolds number of the channel is equal to 2317. In contrast, the k-ω viscous model does not appear to describe flow pattern within the monolith. Moreover, the velocity profile is almost linear in 2D, whereas in 3D, the flow tends to cross the porous wall at the end of the filter. Further, unlike the turbulent models, the laminar model demonstrates high dependency on inflow conditions.
  The integration of an inlet and outlet zone increases the need for computer resources as well as increasing the difficulty in obtaining a converged solution. However, the modification of the domain does not significantly affect the numerical pressure drop. Finally, this investigation demonstrates that although the assumption of laminar flow is most accepted, it may not be the most accurate method of analyzing gasoline particulate filter. To the best of the author’s knowledge, there are no known published works that have yielded the same
  conclusion.

• Subjects / Keywords

  • GPF
  • Turbulence
  • Gasoline Particulate Filter
  • CFD
  • Flow Pattern
  • Channel

• Graduation date

  Fall 2019

• Type of Item

  Thesis

• Degree

  Master of Science

• DOI

  https://doi.org/10.7939/r3-yb55-jd04

• License

  Permission is hereby granted to the University of Alberta Libraries to reproduce single copies of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only. Where the thesis is converted to, or otherwise made available in digital form, the University of Alberta will advise potential users of the thesis of these terms. The author reserves all other publication and other rights in association with the copyright in the thesis and, except as herein before provided, neither the thesis nor any substantial portion thereof may be printed or otherwise reproduced in any material form whatsoever without the author's prior written permission.