Numerical simulation of a GDI engine flow using LES and POD

This paper presents the findings from a numerical study of a gasoline direct injection engine flow using the Large Eddy Simulation (LES) modelling technique. The study is carried out over 30 successive engine cycles. The study illustrates how the more simple but robust Smagorinsky LES sub-grid scale...

Full description

Saved in:
Bibliographic Details
Main Authors: Nicholas J. Beavis, Salah Ibrahim, Weeratunge Malalasekera
Format: Default Conference proceeding
Published: 2016
Subjects:
Other engineering not elsewhere classified
untagged
Engineering not elsewhere classified
Online Access:https://hdl.handle.net/2134/21155
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper presents the findings from a numerical study of a gasoline direct injection engine flow using the Large Eddy Simulation (LES) modelling technique. The study is carried out over 30 successive engine cycles. The study illustrates how the more simple but robust Smagorinsky LES sub-grid scale turbulence model can be applied to a complex engine geometry with realistic engineering mesh size and computational expense whilst still meeting the filter width requirements to resolve the majority of large scale turbulent structures. Detailed description is provided here for the computational setup, including the initialisation strategy. The mesh is evaluated using a turbulence resolution parameter and shows the solution to generally resolve upwards of 80% of the turbulence kinetic energy. The calculated mean and fluctuating velocity components have been validated across multiple cutting planes at key crank angles within the intake stroke with good agreement obtained against experimental data and compared with RANS model predictions. A Proper Orthogonal Decomposition (POD) technique is then used to evaluate the in-cylinder flow field with the results focusing around the eigenvalue/energy content and time coefficients associated with each mode. The findings have shown how this technique can be used to assess the amount of small scale turbulence generated at the point of spark timing, the level of flow field cyclic variability and the degree of statistical convergence to be expected from an ensemble average result based on the number of cycles and the level of cyclic variability present in the flow field.

Similar Items