Show simple item record

dc.contributor.authorBacker, Lara
dc.date.accessioned2019-10-15T16:51:20Z
dc.date.available2020-02-29T07:01:17Z
dc.date.issued2019-08-30
dc.identifier.otherBacker_cornellgrad_0058F_11597
dc.identifier.otherhttp://dissertations.umi.com/cornellgrad:11597
dc.identifier.otherbibid: 11050743
dc.identifier.urihttps://hdl.handle.net/1813/67757
dc.description.abstractThe race to reduce pollutant emissions from hydrocarbon combustion while simultaneously increasing fuel efficiency and optimizing engine performance calls for the use of numerical simulations in parallel with, or in lieu of, expensive and time-consuming experiments. To explore the efficacy of emerging alternative fuels and additives in numerical simulations and to predict the effects of the fuel description on emissions, the fuel should be treated as one of the optimization parameters. This necessitates an accurate and detailed description of the fuel and its breakdown, as combustion kinetics are exceedingly dependent on fuel constituents. However, the combustion of even a single fuel component can involve hundreds of species and thousands of reactions, requiring prohibitively high CPU times for realistic simulations of complex fuels with detailed chemistry. An advantageous strategy to combat this difficulty is to employ reduced-order modeling by replacing the realistic fuel blend with a simplified description called a surrogate, in tandem with reducing the chemical kinetic mechanism. In recent years, a component library framework has been proposed to facilitate the creation of reduced-order models for practical applications. The idea is that chemical models for single-component fuels can be reduced separately and combined at-will to represent any surrogate blend of interest. However, this approach fails when individual fuel molecules have significant non-linear interactions with one another during combustion, or when the prediction of pollutant formation is of interest, since the kinetics involved strongly depend on the details of the multi-component fuel mixture. In this work, two new strategies are presented to automatically facilitate the generation of compact, reduced-order models for multi-component fuels. The first addresses the drawbacks of the component library framework by efficiently allowing for the automatic creation of reduced fuel component oxidation mechanisms and the addition of secondary pathways of interest onto existing component library modules, directly at the reduced level. The second generates a compact description of multi-component fuel decomposition chemistry, significantly reducing the computational cost of simulating fuels with numerous constituents. Reduced-order models created with these techniques are shown to reproduce the behavior of detailed kinetic models reasonably well. Subsequent studies leverage the strategies presented here to produce reduced kinetic mechanisms for multi-component fuel chemistry. A preliminary analysis highlights relevant combustion regimes and useful canonical problems to consider when reducing models for turbulent combustion applications. Results from this analysis are used to guide the creation of a compact reduced-order model for jet fuel.
dc.language.isoen_US
dc.rightsAttribution 4.0 International
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectFluid Mechanics
dc.subjectKinetics
dc.subjectComputer science
dc.subjectCombustion
dc.subjectFuels
dc.subjectMechanism Reduction
dc.subjectThermodynamics
dc.titleGenerating Reduced Mechanisms for Realistic Multi-Component Fuel Combustion
dc.typedissertation or thesis
thesis.degree.disciplineMechanical Engineering
thesis.degree.grantorCornell University
thesis.degree.levelDoctor of Philosophy
thesis.degree.namePh.D., Mechanical Engineering
dc.contributor.chairPepiot, Perrine
dc.contributor.committeeMemberPope, Stephen Bailey
dc.contributor.committeeMemberDesjardins, Olivier
dcterms.licensehttps://hdl.handle.net/1813/59810
dc.identifier.doihttps://doi.org/10.7298/nw6z-d370


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record

Except where otherwise noted, this item's license is described as Attribution 4.0 International

Statistics