T-Space at The University of Toronto Libraries >
School of Graduate Studies - Theses >
Please use this identifier to cite or link to this item:
|Title: ||Subfilter Scale Modelling for Large Eddy Simulation of Lean Hydrogen-enriched Turbulent Premixed Combustion|
|Authors: ||Hernandez Perez, Francisco Emanuel|
|Advisor: ||Groth, Clinton P. T.|
Gulder, Omer L.
|Department: ||Aerospace Science and Engineering|
Large Eddy Simulation
Computational Fluid Dynamics
Subfilter Scale Modelling
|Issue Date: ||30-Aug-2011|
|Abstract: ||Hydrogen (H2) enrichment of hydrocarbon fuels in lean premixed systems is desirable since it can lead to a progressive reduction in greenhouse-gas emissions, while paving the way towards pure hydrogen combustion. In recent decades, large-eddy simulation (LES) has emerged as a promising tool to computationally describe and represent turbulent combustion processes.
However, a considerable complication of LES for turbulent premixed combustion is that chemical reactions occur in a thin reacting layer at small scales which cannot be entirely resolved on computational grids and need to be modelled.
In this thesis, subfilter-scale (SFS) modelling for LES of lean H2-enriched methane-air turbulent premixed combustion was investigated. Two- and three-dimensional fully-compressible LES solvers for a thermally perfect reactive mixture of gases were developed and systematically
validated. Two modelling strategies for the chemistry-turbulence interaction were pursued: the artificially thickened flame model with a power-law SFS wrinkling approach and the presumed conditional moment (PCM) coupled with the flame prolongation of intrinsic low-dimensional manifold (FPI) chemistry tabulation technique. Freely propagating and Bunsen-type flames
corresponding to stoichiometric and lean premixed mixtures were considered. Validation of the LES solvers was carried out by comparing predicted solutions with experimental data and other published numerical results.
Head-to-head comparisons of different SFS approaches, including a transported flame surface density (FSD) model, allowed to identify weaknesses and strengths of the various models. Based on the predictive capabilities of the models examined, the PCM-FPI model was selected for the study of hydrogen-enrichment of methane. A new progress of reaction variable was proposed
to account for NO. The importance of transporting species with different diffusion coefficients was demonstrated, in particular for H2. The proposed approach was applied to a Bunsen-type configuration, reproducing key features observed in the experiments: the enriched flame was shorter, which is attributed to a faster consumption of the blended fuel; and the enriched flame displayed a broader two-dimensional curvature probability density function. Furthermore, reduced levels of carbon dioxide (CO2), increased levels of nitrogen monoxide (NO), and a slight increase in the carbon monoxide (CO) levels in areas of fully burned gas were predicted for the
|Appears in Collections:||Doctoral|
This item is licensed under a Creative Commons License
Items in T-Space are protected by copyright, with all rights reserved, unless otherwise indicated.