Book Section records
https://feeds.library.caltech.edu/people/Harstad-K-G/book_section.rss
A Caltech Library Repository Feedhttp://www.rssboard.org/rss-specificationpython-feedgenenTue, 16 Apr 2024 13:41:04 +0000Novel Subgrid Modeling of the LES Equations Under Supercritical Pressure
https://resolver.caltech.edu/CaltechAUTHORS:20200506-071823327
Authors: {'items': [{'id': 'Selle-L-C', 'name': {'family': 'Selle', 'given': 'Laurent C.'}}, {'id': 'Bellan-J', 'name': {'family': 'Bellan', 'given': 'Josette'}, 'orcid': '0000-0001-9218-7017'}, {'id': 'Harstad-K-G', 'name': {'family': 'Harstad', 'given': 'Kenneth G.'}}]}
Year: 2007
DOI: 10.2514/6.2007-568
Transitional states obtained from Direct Numerical Simulation (DNS) of a supercritical mixing layer are analyzed for studying small-scale behavior and assessing the ability of Subgrid Scale (SGS) models to duplicate that behavior. Initially, the mixing layer contains a single chemical species in each of the two streams, and a perturbation promotes rollup and a double pairing of the four spanwise vortices initially present. The database encompasses three combinations of chemical species, several perturbation wavelengths and amplitudes, and several initial Reynolds numbers specifically chosen for the sole purpose of achieving transition. The Large Eddy Simulation (LES) equations are derived from the DNS ones through filtering. This filtering leads to two types of additional terms in the LES compared to the DNS equations : SGS fluxes and other terms for which either assumptions or models are necessary. The magnitude of all terms in the LES conservation equations is analyzed on the DNS database, with special attention to terms that could possibly be neglected. It is shown that in contrast to atmospheric-pressure gaseous flows, there are two new terms that must be modeled: one in each of the momentum and the energy equations. Discussed is a model for the momentum-equation additional term. This model performs well at small filter size but deteriorates as the filter size increases, highlighting the necessity of ensuring appropriate grid resolution in LES.https://authors.library.caltech.edu/records/dc26g-0ct41A New Method in Modeling and Simulations of Complex Oxidation Chemistry
https://resolver.caltech.edu/CaltechAUTHORS:20200505-073936057
Authors: {'items': [{'id': 'Harstad-K-G', 'name': {'family': 'Harstad', 'given': 'Kenneth G.'}}, {'id': 'Bellan-J', 'name': {'family': 'Bellan', 'given': 'Josette'}, 'orcid': '0000-0001-9218-7017'}]}
Year: 2007
DOI: 10.2514/6.2007-1433
A simplified model is proposed for the kinetics of alkane oxidation in air, based on a decomposition of heavy (carbon number ≥3) hydrocarbons into a 13 constituent radical base. The behavior of this base is examined in test computations for heptane utilizing Chemkin II with LLNL data inputs. Emphasis is placed on prediction of the heat release and temperature evolution. At stoichiometric conditions, the total constituent molar density was found to follow a quasi-steady rate which is a simplification in the modeling of its reaction rate.https://authors.library.caltech.edu/records/sf8zg-8g646Modeling of the Energy Equation for LES of Flows at Supercritical Pressure
https://resolver.caltech.edu/CaltechAUTHORS:20200330-131934118
Authors: {'items': [{'id': 'Selle-L-C', 'name': {'family': 'Selle', 'given': 'L. C.'}}, {'id': 'Bellan-J', 'name': {'family': 'Bellan', 'given': 'J.'}, 'orcid': '0000-0001-9218-7017'}, {'id': 'Harstad-K-G', 'name': {'family': 'Harstad', 'given': 'K. G.'}}]}
Year: 2008
DOI: 10.2514/6.2008-948
A database of transitional Direct Numerical Simulation (DNS) realizations of a supercritical mixing layer is analyzed for understanding small-scale behavior and examining Subgrid Scale (SGS) models duplicating that behavior. Initially, the mixing layer contains a single chemical species in each of the two streams, and a perturbation promotes roll-up and a double pairing of the four spanwise vortices initially present. The database encompasses three combinations of chemical species, several perturbation wavelengths and amplitudes, and several initial Reynolds numbers specifically chosen for the sole purpose of achieving transition. The DNS equations are the Navier Stokes, total energy and species equations coupled to a real gas equation of state; the fluxes of species and heat include the Soret and Dufour effects. The Large Eddy Simulation (LES) equations are derived from the DNS ones through Altering. Compared to the DNS equations, two types of additional terms are identified in the LES equations: SGS fluxes and other terms for which either assumptions or models are necessary. The focus is here on the energy equation. The magnitude of all terms in this filtered DNS equation is analyzed on the DNS database, with special attention to terms that could possibly be neglected. It is shown that in contrast to atmospheric-pressure gaseous flows, there is a new term that must be modeled in this equation. This new term can be thought to result from the filtering of the strongly nonlinear equation of state, and is associated with high density-gradient magnitude regions both found in DNS and observed experimentally in fully-turbulent high-pressure flows. A priori modeling approaches for the energy-equation additional term are proposed, all of which must ultimately be tested in LES to show viability.https://authors.library.caltech.edu/records/kt3q6-frf90A Simplified Model of Alkane Oxidation
https://resolver.caltech.edu/CaltechAUTHORS:20200421-083907781
Authors: {'items': [{'id': 'Harstad-K-G', 'name': {'family': 'Harstad', 'given': 'Kenneth G.'}}, {'id': 'Bellan-J', 'name': {'family': 'Bellan', 'given': 'Josette'}, 'orcid': '0000-0001-9218-7017'}]}
Year: 2008
DOI: 10.2514/6.2008-975
A simplified model is proposed for the kinetics of alkane oxidation in air, based on a decomposition of heavy (carbon number greater or equal to 3) hydrocarbons into a 13 constituent radical base. The behavior of this base is examined in test computations for n-heptane utilizing Chemkin II with LLNL data inputs, placing emphasis on modeling to predict the heat release and temperature evolution. A normalized temperature was constructed which when used to plot the total constructed molar density divided by the product of the equivalence ratio and a nondimensional pressure, reveals a self-similar behavior of the plotted variable over a wide range of initial pressures and equivalence ratios. Examination of the LLNL kinetics shows that the total constituent molar density rate follows a quasi-steady behavior. This reaction rate was curve fitted along with the corresponding enthalpy production. The fits are shown against the normalized temperature for various equivalence ratios and initial nondimensional pressures and comparisons with the LLNL kinetics are very favorable. The model reduces the LLNL n-heptane mechanism from 160 species (progress variables) and 1540 reactions to 12 progress variables, 16 quasi-steady rates (associated with heavy species), 162 conventional reaction rates (light species) and 11 other functional forms. (i.e. fits for the mean heavy-species heat capacity at constant pressure, the enthalpy release rate of the heavy species, and the molar fraction of quasi-steady light species). The proposed kinetic mechanism is valid over a pressure range from atmospheric to 60 bar, temperatures from 600 K to 2500 K and equivalence ratios from 0.125 to 8. This range encompasses diesel, HCCI and gas turbine engines, including cold ignition; and NO_x, CO and soot pollutant formation in the lean and rich regimes, respectively.https://authors.library.caltech.edu/records/g9qg0-nf078Modeling of Alkane Oxidation using Constituents and Species
https://resolver.caltech.edu/CaltechAUTHORS:20200117-081103466
Authors: {'items': [{'id': 'Harstad-K-G', 'name': {'family': 'Harstad', 'given': 'Kenneth G.'}}, {'id': 'Bellan-J', 'name': {'family': 'Bellan', 'given': 'Josette'}, 'orcid': '0000-0001-9218-7017'}]}
Year: 2009
DOI: 10.2514/6.2009-1368
A chemical kinetics reduction model is proposed for alkane oxidation in air that is baaed on a parallel methodology to that used in turbulence modeling in the context of Large Eddy Simulation. The objective of kinetic modeling is to predict the heat release and temperature evolution. In an a priori step, a categorization of time scales is first conducted to identify scales that must be modeled and scales that must be computed using progress variables based on the model for the other scales. First, a decomposition of heavy (carbon number greater or equal to 3) hydrocarbons into constituents is proposed. Examination of results obtained using the LLNL heptane-oxidation database in conjunction with Chemkin II shows that (i) with appropriate scaling, the total constituent mole fraction behaves in a self-similar manner and the total constituent molar density rate follows a quasi-steady behavior, and (ii) the light species can be partitioned into two subsets according to whether
they are quasi-steady (nine species) or unsteady (11 species). The twelve progress variables represented by the total constituent molar density and the molar densities of the unsteady light species are defined to be a base from which the system's behavior can be reproduced. This is a dramatic reduction from the 160 species (progress variables) and 1540 reactions in the LLNL set to 12 progress variables, 16 quasi-steady rates (associated with heavy species), 162 conventional reaction rates (light species) and 11 other functional forms (i.e. fits for the mean heavy-species heat capacity at constant pressure, the enthalpy release rate of the heavy species, and the molar fraction of quasi-steady light species). A summary of the model is presented explaining the curve fits that constitute the model, namely (1) for the constituent molar density rate a long with the corresponding enthalpy production rate, (2) for the quasi-steady species mole fraction, and (3) for the contribution from the heavy species to the unsteady light species reaction rates. The proposed kinetic mechanism is valid over a pressure range from atmospheric to 60 bar, temperatures from 600 K to 2500 K and equivalence ratio1 from 0.125 to 8. This range encompasses diesel, HCCI and gas turbine engines, including cold ignition; and NO_x, CO and soot pollutant formation in the lean and rich regimes, respectively. Highlights of the a priori model results are illustrated for a variety of initial conditions. Results from a posteriori tests are shown in which the model predictions for the unsteady light species and the temperature are compared to the equivalent quantities baaed on the LLNL dataset.https://authors.library.caltech.edu/records/4jve6-r0p63Alkane Kinetics Reduction Consistent with Turbulence Modeling using Large Eddy Simulation
https://resolver.caltech.edu/CaltechAUTHORS:20191016-111202608
Authors: {'items': [{'id': 'Harstad-K-G', 'name': {'family': 'Harstad', 'given': 'Kenneth G.'}}, {'id': 'Bellan-J', 'name': {'family': 'Bellan', 'given': 'Josette'}, 'orcid': '0000-0001-9218-7017'}]}
Year: 2010
DOI: 10.2514/6.2010-1514
A methodology for deriving a reduced kinetic mechanism for alkane oxidation is described, inspired by n-heptane oxidation. The model is based on partitioning the species of the skeletal kinetic mechanism into lights, defined as those having a carbon number smaller than 3, and heavies, which are the complement of the species ensemble. For modeling purposes, the heavy species are mathematically decomposed into constituents, which are similar but not identical to groups in the group additivity theory. From analysis of the n-heptane LLNL skeletal mechanism in conjunction with CHEMKIN II, it is shown that a similarity variable can be formed such that the appropriately non-dimensionalized global constituent molar density exhibits a self-similar behavior over a very wide range of equivalence ratios, initial pressures and initial temperatures that is of interest for predicting n-heptane oxidation. Furthermore, the oxygen and water molar densities are shown to display a quasi-linear behavior with respect to the similarity variable. The light species ensemble is partitioned into quasi-steady and unsteady species. The reduced model is based on concepts consistent with those of Large Eddy Simulation in which functional forms are used to replace the small scales eliminated through Altering of the governing equations; these small scales are unimportant as far as dynamic energy is concerned. Here, we remove the scales deemed unimportant for recovering the thermodynamic energy. The concept is tested by using tabular information from the n-heptane LLNL skeletal mechanism in conjunction with CHEMKIN II utilized as surrogate ideal functions replacing the necessary functional forms. The test reveals that the similarity concept is indeed justified and that the combustion temperature is well predicted, but that the ignition time is overpredicted, which is traced to neglecting a detailed description of the processes lending to the heavies chemical decomposition. To palliate this deficiency, functional modeling is incorporated into our conceptual reduction. This functional modeling includes the global constituent molar density, the enthalpy evolution of the heavies, the contribution to the reaction rate of the unsteady lights from other light species and from the heavies, the molar density evolution of oxygen and water, and the mole &actions of the quasi-steady light species. The model is compact in that there are only nine species-related progress variables. Results are presented showing the performance of the model for predicting the temperature and species evolution for n-heptane. The model reproduces the ignition time over a wide range of equivalence ratios, initial pressure and initial temperature. Preliminary results for iso-octane using the full mechanism are also presented, showing encouragingly that the concept may be generalized to other alkanes. The utility of the model and possible improvements are discussed.https://authors.library.caltech.edu/records/31y6z-srt68Computation of Laminar Premixed Flames Using Reduced Kinetics Based on Constituents and Species
https://resolver.caltech.edu/CaltechAUTHORS:20190930-110459291
Authors: {'items': [{'id': 'Harstad-K-G', 'name': {'family': 'Harstad', 'given': 'Kenneth G.'}}, {'id': 'Bellan-J', 'name': {'family': 'Bellan', 'given': 'Josette'}, 'orcid': '0000-0001-9218-7017'}]}
Year: 2011
DOI: 10.2514/6.2011-415
A model is proposed for quasi-one-dimensional steady flame development in the configuration of an inviscid, premixed fuel jet injected into air. The governing equations are written within the framework of a reduced kinetic model based on constituents and species. The reduced kinetic model, previously exercised in a constant-volume perfectly-stirred reactor mode, has been successful at predicting ignition and combustion product and temperature evolution for n-heptane, iso-octane, PRF fuel combinations, and mixtures of iso-octane with either n-pentane or iso-hexane. The differential governing equations have the option of an axially variable area and they are coupled with a real gas equation of state. The flame development model accounts for a full diffusion matrix, and thermal conductivity computed for the species mixture. Preliminary results are presented.https://authors.library.caltech.edu/records/k5cjx-eej12Modeling of Steady Laminar Flames for One-dimensional Premixed Jets of Heptane/Air and Octane/Air Mixtures
https://resolver.caltech.edu/CaltechAUTHORS:20190828-102318692
Authors: {'items': [{'id': 'Harstad-K-G', 'name': {'family': 'Harstad', 'given': 'Kenneth G.'}}, {'id': 'Bellan-J', 'name': {'family': 'Bellan', 'given': 'Josette'}, 'orcid': '0000-0001-9218-7017'}]}
Year: 2012
DOI: 10.2514/6.2012-340
A model is proposed for quasi-one-dimensional steady flame development in the configuration of an inviscid, premixed fuel jet injected into air. The governing equations are written within the framework of a reduced kinetic model based on constituents and species. The reduced kinetic model, previously exercised in a constant-volume perfectlystirred reactor mode, has been successful at predicting ignition and combustion product and temperature evolution for n-heptane, iso-octane, PRF fuel combinations, and mixtures of iso-octane with either n-pentane or iso-hexane. The differential governing equations have the option of an axially variable area and they are coupled with a real gas equation of state. The flame development model accounts for a full diffusion matrix, and thermal conductivity computed for the species mixture. Results from four simulations at various conditions are presented.https://authors.library.caltech.edu/records/7b2pw-t5720Direct Numerical Simulation of High-Pressure Multispecies Turbulent Mixing in the Cold Ignition Regime
https://resolver.caltech.edu/CaltechAUTHORS:20190828-102318779
Authors: {'items': [{'id': 'Masi-E', 'name': {'family': 'Masi', 'given': 'Enrica'}}, {'id': 'Bellan-J', 'name': {'family': 'Bellan', 'given': 'Josette'}, 'orcid': '0000-0001-9218-7017'}, {'id': 'Harstad-K-G', 'name': {'family': 'Harstad', 'given': 'Kenneth'}}]}
Year: 2012
DOI: 10.2514/6.2012-351
A model is proposed for describing mixing of several species under high pressure conditions that relies on a previously proposed model based on governing equations for multispecies mixing that has so far only been exercised for two-species mixing. For the two-species mixing simulations, transport properties were computed from correlated Schmidt (Sc) and Prandtl (Pr) numbers, accurately calculated as functions of the thermodynamic variables, and from a specified Reynolds number value from which an adjusted viscosity value was calculated so as to enable Direct Numerical Simulation (DNS). One of the novelties of the present study is the modeling and computation of multispecies mixing based on a full mass-diffusion matrix, a full thermal-diffusion-factor matrix necessary to include Soret and Dufour effects, and thermal conductivity computed for the species mixture. The scaling of the viscosity necessary for conducting DNS induces a scaling of the other transport properties that respects the accurate values of the Sc numbers and of the Pr number. Computations are performed with five species in the configuration of a temporal mixing layer and the effect of transport properties on species mixing and layer development are analyzed and discussed.https://authors.library.caltech.edu/records/y01wa-vt468Pressure Effects from Direct Numerical Simulation of High-Pressure Multispecies Mixing
https://resolver.caltech.edu/CaltechAUTHORS:20190826-092411849
Authors: {'items': [{'id': 'Masi-E', 'name': {'family': 'Masi', 'given': 'Enrica'}}, {'id': 'Bellan-J', 'name': {'family': 'Bellan', 'given': 'Josette'}, 'orcid': '0000-0001-9218-7017'}, {'id': 'Harstad-K-G', 'name': {'family': 'Harstad', 'given': 'Kenneth'}}]}
Year: 2013
DOI: 10.2514/6.2013-711
The focus of this study is the understanding of effects of pressure increase or Reynolds number increase in supercritical-pressure flows. To this effect, Direct Numerical Simulations are conducted for supercritical-pressure flows in which five species undergo mixing. The computation of multispecies mixing is based on a full mass-diffusion matrix, a full thermal-diffusion-factor matrix necessary to include Soret and Dufour effects, and both viscosity and thermal conductivity computed for the species mixture. The scaling of the physical viscosity, necessary for conducting DNS, induces a scaling of the other transport properties that respects the accurate values of the Schmidt (Sc) numbers and of the Prandtl (Pr) number. Computations are performed in the configuration of a temporal mixing layer and the results are analyzed to reveal the separate effect of pressure or Reynolds number increase on the flow. The analysis consists of examining vortical aspects of the flow, the fluxes and relevant thermodynamic properties. It is found that a larger pressure has an opposite effect to a larger Reynolds number, mainly by increasing the fluid density and making it more difficult to entrain and mix.https://authors.library.caltech.edu/records/esqwj-88k62Modeling of Steady High-Pressure Laminar Premixed Flames of n-Heptane and Iso-Octane
https://resolver.caltech.edu/CaltechAUTHORS:20190826-092411436
Authors: {'items': [{'id': 'Harstad-K-G', 'name': {'family': 'Harstad', 'given': 'Kenneth G.'}}, {'id': 'Bellan-J', 'name': {'family': 'Bellan', 'given': 'Josette'}, 'orcid': '0000-0001-9218-7017'}]}
Year: 2013
DOI: 10.2514/6.2013-1168
A model is proposed for quasi-one-dimensional steady flame development in the configuration of an inviscid, premixed fuel jet injected into air. The governing equations are written within the framework of a reduced kinetic model based on constituents and species. The reduced kinetic model, previously exercised in a constant-pressure perfectlystirred reactor mode, has been successful at predicting ignition and combustion product and temperature evolution for n-heptane, iso-octane, PRF fuel combinations, and mixtures of iso-octane with either n-pentane or iso-hexane. The differential governing equations are coupled with a real gas equation of state. The flame development model accounts for a full diffusion matrix, and thermal conductivity computed for the species mixture. Results are presented for both n-heptane and iso-octane at stoichiometric conditions over the pressure range of 1 - 40 bar. Additionally, an equivalence ratio study is conducted for iso-octane at a pressure of 40 bar.https://authors.library.caltech.edu/records/ze2ws-2gd52Dimensionality Reduction Using a Dominant Dynamic Variable, Self Similarity and Data Tabulation: Application to Hydrocarbon Oxidation
https://resolver.caltech.edu/CaltechAUTHORS:20190819-150530040
Authors: {'items': [{'id': 'Kourdis-P-D', 'name': {'family': 'Kourdis', 'given': 'Panagiotis D.'}}, {'id': 'Bellan-J', 'name': {'family': 'Bellan', 'given': 'Josette'}, 'orcid': '0000-0001-9218-7017'}, {'id': 'Harstad-K-G', 'name': {'family': 'Harstad', 'given': 'Kenneth'}}]}
Year: 2014
DOI: 10.2514/6.2014-0819
A dimensionality reduction method is developed for autonomous dynamical systems exploiting the local (near) self similarity due to the presence of a dominant dynamic variable. The method is coupled with a simple tabulation scheme to take advantage of the computationally more efficient local partial self similarity. The proposed methodology is used to construct reduced kinetics models of hydrocarbon oxidation and is tested for a n-dodecane/air mixture.https://authors.library.caltech.edu/records/0gwhe-4d922Towards direct simulations of counterflow flames with consistent differential-algebraic boundary conditions
https://resolver.caltech.edu/CaltechAUTHORS:20190816-144341538
Authors: {'items': [{'id': 'Kourdis-P-D', 'name': {'family': 'Kourdis', 'given': 'Panagiotis D.'}}, {'id': 'Bellan-J', 'name': {'family': 'Bellan', 'given': 'Josette'}, 'orcid': '0000-0001-9218-7017'}, {'id': 'Harstad-K-G', 'name': {'family': 'Harstad', 'given': 'Kenneth'}}]}
Year: 2015
DOI: 10.2514/6.2015-1383
A new approach for the formulation of boundary conditions for the counterflow configuration is presented. Upon discretization of the steady-state Navier-Stokes equations at the inflow boundaries, numerically algebraic equations are imposed as boundary conditions, while upon discretization of the unsteady Navier-Stokes equations at the outflow, differential boundaries result. It is demonstrated that the resulting numerical differential-algebraic boundary conditions are suitable to account for the multi-directional character of the flow at the boundaries of the counterflow configuration.https://authors.library.caltech.edu/records/dag14-qec60