Show simple item record

dc.contributor.authorBisetti, Fabrizio
dc.date.accessioned2017-06-05T08:35:48Z
dc.date.available2017-06-05T08:35:48Z
dc.date.issued2015-01-07
dc.identifier.urihttp://hdl.handle.net/10754/624103
dc.description.abstractKinetic models for reactive flow applications comprise hundreds of reactions describing the complex interaction among many chemical species. The detailed knowledge of the reaction parameters is a key component of the design cycle of next-generation combustion devices, which aim at improving conversion efficiency and reducing pollutant emissions. Shock tubes are a laboratory scale experimental configuration, which is widely used for the study of reaction rate parameters. Important uncertainties exist in the values of the thousands of parameters included in the most advanced kinetic models. This talk discusses the application of uncertainty quantification (UQ) methods to the analysis of shock tube data as well as the design of shock tube experiments. Attention is focused on a spectral framework in which uncertain inputs are parameterized in terms of canonical random variables, and quantities of interest (QoIs) are expressed in terms of a mean-square convergent series of orthogonal polynomials acting on these variables. We outline the implementation of a recent spectral collocation approach for determining the unknown coefficients of the expansion, namely using a sparse, adaptive pseudo-spectral construction that enables us to obtain surrogates for the QoIs accurately and efficiently. We first discuss the utility of the resulting expressions in quantifying the sensitivity of QoIs to uncertain inputs, and in the Bayesian inference key physical parameters from experimental measurements. We then discuss the application of these techniques to the analysis of shock-tube data and the optimal design of shock-tube experiments for two key reactions in combustion kinetics: the chain-brancing reaction H + O2 ←→ OH + O and the reaction of Furans with the hydroxyl radical OH.
dc.relation.urlhttp://mediasite.kaust.edu.sa/Mediasite/Play/a08d38a04a87419e83f99876e2ad79061d?catalog=ca65101c-a4eb-4057-9444-45f799bd9c52
dc.titleSurrogate models and optimal design of experiments for chemical kinetics applications
dc.typePresentation
dc.contributor.departmentClean Combustion Research Center
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
dc.contributor.departmentMechanical Engineering Program
dc.contributor.departmentPhysical Science and Engineering (PSE) Division
dc.contributor.departmentReactive Flow Modeling Laboratory (RFML)
dc.conference.dateJanuary 6-9, 2015
dc.conference.nameAdvances in Uncertainty Quantification Methods, Algorithms and Applications (UQAW 2015)
dc.conference.locationKAUST
kaust.personBisetti, Fabrizio
refterms.dateFOA2018-06-14T05:59:03Z


Files in this item

This item appears in the following Collection(s)

Show simple item record