skip to main content
DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Application of deep artificial neural networks to multi-dimensional flamelet libraries and spray flames

Abstract

The “curse of dimensionality” has limited the applicability and expansion of tabulated combustion models. While the tabulated flamelet model and other multi-dimensional manifold approaches have shown predictive capability, the associated tabulation involves the storage of large lookup tables, requiring large memory as well as multi-dimensional interpolation subroutines, all implemented during runtime. As such, this work investigates the use of deep artificial neural networks to replace lookup tables in order to reduce the memory footprint and increase the computational speed of tabulated flamelets and related approaches. Specifically, different strategic approaches to training the artificial neural network models are explored and a grouped multi-target artificial neural network is introduced, which takes advantage of the ability of artificial neural networks to map an input space to multiple targets by classifying the species based on their correlation to one another. The grouped multi-target artificial neural network approach is validated by applying it to an n-dodecane spray flame using conditions of the Spray A flame from the Engine Combustion Network and comparing global flame characteristics for different ambient conditions using a well-established large-eddy simulation framework. The same framework is then extended to the simulations of methyl decanoate combustion in a compression ignition engine. The validationmore » studies show that the grouped multi-target artificial neural networks are able to accurately capture flame liftoff, autoignition, two-stage heat release and other quantitative trends over a range of conditions. The use of neural networks in conjunction with the grouping mechanism as performed in the grouped multi-target artificial neural network produces a significant reduction in the memory footprint and computational costs for the code and, thus, widens the operating envelope for higher fidelity engine simulations with detailed mechanisms.« less

Authors:
ORCiD logo [1];  [2];  [2];  [3];  [2]
  1. North Carolina State Univ., Raleigh, NC (United States); Argonne National Lab. (ANL), Lemont, IL (United States)
  2. Argonne National Lab. (ANL), Lemont, IL (United States)
  3. North Carolina State Univ., Raleigh, NC (United States)
Publication Date:
Research Org.:
Argonne National Lab. (ANL), Argonne, IL (United States)
Sponsoring Org.:
USDOE Office of Science (SC); USDOE Office of Energy Efficiency and Renewable Energy (EERE), Vehicle Technologies Office (EE-3V)
OSTI Identifier:
1605138
Grant/Contract Number:  
AC02-06CH11357
Resource Type:
Accepted Manuscript
Journal Name:
International Journal of Engine Research
Additional Journal Information:
Journal Volume: 21; Journal Issue: 1; Journal ID: ISSN 1468-0874
Publisher:
SAGE
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING; artificial neural networks; chemistry tabulation; flamelets

Citation Formats

Owoyele, Opeoluwa, Kundu, Prithwish, Ameen, Muhsin M., Echekki, Tarek, and Som, Sibendu. Application of deep artificial neural networks to multi-dimensional flamelet libraries and spray flames. United States: N. p., 2019. Web. https://doi.org/10.1177/1468087419837770.
Owoyele, Opeoluwa, Kundu, Prithwish, Ameen, Muhsin M., Echekki, Tarek, & Som, Sibendu. Application of deep artificial neural networks to multi-dimensional flamelet libraries and spray flames. United States. https://doi.org/10.1177/1468087419837770
Owoyele, Opeoluwa, Kundu, Prithwish, Ameen, Muhsin M., Echekki, Tarek, and Som, Sibendu. Fri . "Application of deep artificial neural networks to multi-dimensional flamelet libraries and spray flames". United States. https://doi.org/10.1177/1468087419837770. https://www.osti.gov/servlets/purl/1605138.
@article{osti_1605138,
title = {Application of deep artificial neural networks to multi-dimensional flamelet libraries and spray flames},
author = {Owoyele, Opeoluwa and Kundu, Prithwish and Ameen, Muhsin M. and Echekki, Tarek and Som, Sibendu},
abstractNote = {The “curse of dimensionality” has limited the applicability and expansion of tabulated combustion models. While the tabulated flamelet model and other multi-dimensional manifold approaches have shown predictive capability, the associated tabulation involves the storage of large lookup tables, requiring large memory as well as multi-dimensional interpolation subroutines, all implemented during runtime. As such, this work investigates the use of deep artificial neural networks to replace lookup tables in order to reduce the memory footprint and increase the computational speed of tabulated flamelets and related approaches. Specifically, different strategic approaches to training the artificial neural network models are explored and a grouped multi-target artificial neural network is introduced, which takes advantage of the ability of artificial neural networks to map an input space to multiple targets by classifying the species based on their correlation to one another. The grouped multi-target artificial neural network approach is validated by applying it to an n-dodecane spray flame using conditions of the Spray A flame from the Engine Combustion Network and comparing global flame characteristics for different ambient conditions using a well-established large-eddy simulation framework. The same framework is then extended to the simulations of methyl decanoate combustion in a compression ignition engine. The validation studies show that the grouped multi-target artificial neural networks are able to accurately capture flame liftoff, autoignition, two-stage heat release and other quantitative trends over a range of conditions. The use of neural networks in conjunction with the grouping mechanism as performed in the grouped multi-target artificial neural network produces a significant reduction in the memory footprint and computational costs for the code and, thus, widens the operating envelope for higher fidelity engine simulations with detailed mechanisms.},
doi = {10.1177/1468087419837770},
journal = {International Journal of Engine Research},
number = 1,
volume = 21,
place = {United States},
year = {2019},
month = {3}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record

Citation Metrics:
Cited by: 4 works
Citation information provided by
Web of Science

Save / Share:

Works referenced in this record:

An Algorithm for Least-Squares Estimation of Nonlinear Parameters
journal, June 1963

  • Marquardt, Donald W.
  • Journal of the Society for Industrial and Applied Mathematics, Vol. 11, Issue 2
  • DOI: 10.1137/0111030

Small scales, many species and the manifold challenges of turbulent combustion
journal, January 2013


Tabulation of combustion chemistry via Artificial Neural Networks (ANNs): Methodology and application to LES-PDF simulation of Sydney flame L
journal, November 2017


Simulation of combustion in direct injection diesel engines using a eulerian particle flamelet model
journal, January 2000


Reduced-Order Modeling of Turbulent Reacting Flows with Application to Ramjets and Scramjets
journal, March 2011

  • Torrez, Sean M.; Driscoll, James F.; Ihme, Matthias
  • Journal of Propulsion and Power, Vol. 27, Issue 2
  • DOI: 10.2514/1.50272

Importance of turbulence-chemistry interactions at low temperature engine conditions
journal, September 2017


Progress-variable approach for large-eddy simulation of non-premixed turbulent combustion
journal, January 1999


A method for the solution of certain non-linear problems in least squares
journal, January 1944

  • Levenberg, Kenneth
  • Quarterly of Applied Mathematics, Vol. 2, Issue 2
  • DOI: 10.1090/qam/10666

Flamelet LES of a semi-industrial pulverized coal furnace
journal, November 2016


Assessment of flamelet versus multi-zone combustion modeling approaches for stratified-charge compression ignition engines
journal, February 2015

  • Pal, Pinaki; Keum, SeungHwan; Im, Hong G.
  • International Journal of Engine Research, Vol. 17, Issue 3
  • DOI: 10.1177/1468087415571006

Modeling diesel engine combustion using pressure dependent Flamelet Generated Manifolds
journal, January 2011

  • Bekdemir, C.; Somers, L. M. T.; de Goey, L. P. H.
  • Proceedings of the Combustion Institute, Vol. 33, Issue 2
  • DOI: 10.1016/j.proci.2010.07.091

Unsteady flamelet modeling of turbulent hydrogen-air diffusion flames
journal, January 1998


DNS of a turbulent lifted DME jet flame
journal, July 2016


Development and validation of an n-dodecane skeletal mechanism for spray combustion applications
journal, March 2014


A direct numerical simulation of cool-flame affected autoignition in diesel engine-relevant conditions
journal, January 2017

  • Krisman, Alex; Hawkes, Evatt R.; Talei, Mohsen
  • Proceedings of the Combustion Institute, Vol. 36, Issue 3
  • DOI: 10.1016/j.proci.2016.08.043

Comparison of well-mixed and multiple representative interactive flamelet approaches for diesel spray combustion modelling
journal, January 2014


Understanding the ignition mechanism of high-pressure spray flames
journal, January 2017

  • Dahms, Rainer N.; Paczko, Günter A.; Skeen, Scott A.
  • Proceedings of the Combustion Institute, Vol. 36, Issue 2
  • DOI: 10.1016/j.proci.2016.08.023

Modeling Diesel Spray Ignition Using Detailed Chemistry with a Progress Variable Approach
journal, September 2006

  • Lehtiniemi, Harry; Mauss, Fabian; Balthasar, Michael
  • Combustion Science and Technology, Vol. 178, Issue 10-11
  • DOI: 10.1080/00102200600793148

An integrated PDF/neural network approach for simulating turbulent reacting systems
journal, January 1996


Investigation of Methyl Decanoate Combustion in an Optical Direct-Injection Diesel Engine
journal, November 2014

  • Cheng, A. S. (Ed); Dumitrescu, Cosmin E.; Mueller, Charles J.
  • Energy & Fuels, Vol. 28, Issue 12
  • DOI: 10.1021/ef501934n

Modelling of Premixed Laminar Flames using Flamelet-Generated Manifolds
journal, December 2000


Manifold resolution study of the FGM method for an igniting diesel spray
journal, November 2013


Large eddy simulation of extinction and reignition with artificial neural networks based chemical kinetics
journal, March 2010


Detailed chemical kinetic oxidation mechanism for a biodiesel surrogate
journal, August 2008


Toward computationally efficient combustion DNS with complex fuels via principal component transport
journal, January 2017


Principal component transport in turbulent combustion: A posteriori analysis
journal, May 2015


Modeling heat loss effects in the large eddy simulation of a model gas turbine combustor with premixed flamelet generated manifolds
journal, January 2015


Prediction of autoignition in a lifted methane/air flame using an unsteady flamelet/progress variable model
journal, October 2010


Novel Tabulated Combustion Model Approach for Lifted Spray Flames with Large Eddy Simulations
journal, April 2016

  • Ameen, Muhsin M.; Kundu, Prithwish; Som, Sibendu
  • SAE International Journal of Engines, Vol. 9, Issue 4
  • DOI: 10.4271/2016-01-2194

Evaluation of an Unsteady Flamelet Progress Variable Model for Autoignition and Flame Lift-Off in Diesel Jets
journal, February 2013


An equivalent dissipation rate model for capturing history effects in non-premixed flames
journal, February 2017


Optimal artificial neural networks and tabulation methods for chemistry representation in LES of a bluff-body swirl-stabilized flame
journal, January 2009

  • Ihme, Matthias; Schmitt, Christoph; Pitsch, Heinz
  • Proceedings of the Combustion Institute, Vol. 32, Issue 1
  • DOI: 10.1016/j.proci.2008.06.100

Simultaneous formaldehyde PLIF and high-speed schlieren imaging for ignition visualization in high-pressure spray flames
journal, January 2015

  • Skeen, Scott A.; Manin, Julien; Pickett, Lyle M.
  • Proceedings of the Combustion Institute, Vol. 35, Issue 3
  • DOI: 10.1016/j.proci.2014.06.040

Large eddy simulation of pressure and dilution-jet effects on soot formation in a model aircraft swirl combustor
journal, June 2018


Laminar diffusion flamelet models in non-premixed turbulent combustion
journal, January 1984


Datasets of seed mucilage traits for Arabidopsis thaliana natural accessions with atypical outer mucilage
journal, March 2021


    Works referencing / citing this record:

    An Efficient Machine-Learning Approach for PDF Tabulation in Turbulent Combustion Closure
    journal, November 2019