Artificial Neural Networks for Modeling of Chemical Source Terms in CFD Simulations of Turbulent Reactive Flows

Abstract:

The main goal of the work presented here was to develop, implement and test a highly efficient numerical algorithm for the evaluation of the chemical reaction source terms that appear in the Navier - Stokes equations when a turbulent, premixed, reactive flow is simulated using a finite rate chemistry combustion model. The approach was based on employing Artificial Neural Networks (ANN) that were designed, trained and incorporated into an existing LEM – LES numerical algorithm. Two numerical simulations of reacting flows have been carried out using several techniques for the estimation of the LES filtered reaction rate for the chemical species in laminar and turbulent, premixed, reactive flows, and the results were compared in terms of numerical accuracy and computational speed. It was concluded that the ANN approach provides a significant speedup of the numerical simulation while preserving acceptable accuracy.

• Periodical: Applied Mechanics and Materials (Volume 555)
• Pages: 395-400
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.555.395
• Online since: June 2014
• Authors: Ionuţ Porumbel*, Andreea Cristina Petcu, Florin Gabriel Florean, Constantin Eusebiu Hritcu
• Keywords: Artificial Neural Network (ANN), Combustion Modeling, Computational Fluid Dynamics (CFD), Turbulent Combustion

References

[1] N. Peters: Turbulent Combustion (Cambridge Univ. Press, Cambridge, U.K. 2000).

[2] B. Gherman, F.G. Florean, I. Porumbel and C. Cârlănescu: On the influence of the combustion model on the results of turbulent flames numerical simulations, GT2012-69255: Proc. ASME Turbo Expo Conf. (2012).

DOI: 10.1115/gt2012-69255

[3] F.C. Christo, A.R. Masri and E.M. Nebot: Artificial neural networks implementation of chemistry with PDF simulation of H2/CO2 flames: Comb. Flame, Vol. 106, p.406 (1996).

DOI: 10.1016/0010-2180(95)00250-2

[4] J.A. Blasco, N. Fueyo, C. Dopazo and J. Ballester: Modelling the temporal evolution of a reduced combustion chemical system with an artificial neural network: Comb. Flame, Vol. 113, p.38 (1998).

DOI: 10.1016/s0010-2180(97)00211-3

[5] J.A. Blasco, N. Fueyo, J. C. Larroya, C. Dopazo and J.Y. Chen: A single-step time-integrator of a methane air chemical system using artificial neural networks: Comp. Chem. Engr., Vol. 23, p.1127 (1999).

DOI: 10.1016/s0098-1354(99)00278-1

[6] F.C. Christo, A.R. Masri, E.M. Nebot and S. Pope: An integrated PDF / neural network approach for simulating turbulent reactive systems: Proc. Comb. Inst., Vol. 26, p.43, (1995).

DOI: 10.1016/s0082-0784(96)80198-6

[7] J.Y. Chen, J.A. Blasco, N. Fueyo and C. Dopazo: A single-step time-integrator of a methane air chemical system using artificial neural networks: Proc. Comb. Inst., Vol. 28, p.115 (2000).

DOI: 10.1016/s0098-1354(99)00278-1

[8] R. Kapoor, A. Lentati and S. Menon: Simulations of methane - air flames using ISAT and ANN. AIAA-2001-3847(2001).

DOI: 10.2514/6.2001-3847

[9] R. Kapoor and S. Menon: Computational issues for simulating finite - rate kinetics in LES: Proc. Turbo Expo Conf. (2002).

DOI: 10.1115/gt2002-30608

[10] F. Flemming, A. Sadiki and J. Janicka: LES using artificial neural networks for chemistry representation: Proc. 21st German Flameday Conf. (2003).

[11] F. Sarghini, G. de Felice and S. Santini: Neural networks based subgrid scale modeling in large eddy simulations: Comp. Fluids, 32, p.97 (2003).

DOI: 10.1016/s0045-7930(01)00098-6

[12] Y. Choi and J.Y. Chen: Fast prediction of start-of-combustion in HCCI with combined artificial neural networks and ignition delay model: Proc. Comb. Inst., Vol. 30, p.2711 (2005).

DOI: 10.1016/j.proci.2004.08.143

[13] B.A. Sen and S. Menon: Representation of Chemical Kinetics by Artificial Neural Networks for Large Eddy Simulations: AIAA – 2007 – 5635 (2007).

DOI: 10.2514/6.2007-5635

[14] B.A. Sen and S. Menon: Artificial Neural Networks Based Chemistry-Mixing Subgrid Model for LES: AIAA – 2009 – 241 (2009).

DOI: 10.2514/6.2009-241

[15] B.A. Sen and S. Menon: Linear eddy mixing based tabulation and artificial neural networks for large eddy simulations of turbulent flames: Proc. Comb. Inst., Vol. 32, p.1605 (2009).

DOI: 10.1016/j.combustflame.2009.06.005

[16] B.A. Sen, E.R. Hawkes and S. Menon: Large eddy simulation of extinction and reignition with artificial neural networks based chemical kinetics: Comb. Flame, Vol. 157, p.566 (2010).

DOI: 10.1016/j.combustflame.2009.11.006

[17] F. Rosenblatt: Principles of Neurodynamics (Spartan Books 1962).

[18] D.B. Parker: Learning-logic: TR-47, MIT Center Comp. Res. Ec. Mgmt. Sc. (1985).

[19] NeuralWare: Neural computing. A technology handbook for NeuralWorks Prof. II Plus (2001).

[20] R.J. Kee, F.M. Rupley, and J.A. Miller: CHEMKIN-III a FORTRAN chemical kinetics package for the analysis of gas phase chemical kinetics: SAND96-8216, Sandia Nat. Lab. (1996).

DOI: 10.2172/481621

[21] I. Porumbel, C. Cârlănescu, F.G. Florean and C.E. Hriţcu: LES Algorithm for Turbulent Reactive Flows Simulation: Proc. Appl. Comp. Conf. (2010).

[22] P. N. Brown, G. D. Bryne, and A. C. Hindmarsh: VODE, a variable-coefficient ode solver: SIAM J. Sc. Stat. Comp., Vol. 10, no. 5, p.1038 (1989).

DOI: 10.1137/0910062

[23] C.J. Sung, C.K. Law, and J.Y. Chen: An augmented reduced mechanism for methane oxidation with comprehensive global parametric validation: 27thSymp. (Intl. ) Comb., p.295 (1998).

DOI: 10.1016/s0082-0784(98)80416-5

[24] G. Eggenspieler. Numerical Simulation of Pollutant Emission and Flame Extinction in Lean Premixed Systems: PhD thesis, Georgia Institute of Technology, Atlanta, Georgia, USA (2005).

[25] S.B. Pope: Computationally efficient implementation of combustion chemistry using in situ adaptive tabulation: Comb. Th. Mod., Vol. 1, p.41 (1997).

DOI: 10.1080/713665229

[26] D. Isvoranu, V. Badescu, C. Berbente, P. Parvu and D. Crunteanu: Numerical simulation of NOx emissions in the exhaust gas re-burn process: 4th Intl. Coll. Math. Engr. Num. Phys. (2006).