Hostname: page-component-78c5997874-v9fdk Total loading time: 0 Render date: 2024-11-12T00:52:34.344Z Has data issue: false hasContentIssue false
  • English
  • Français

Online in situ prediction of 3-D flame evolution from its history 2-D projections via deep learning

Published online by Cambridge University Press:  18 July 2019

Jianqing Huang Open the ORCID record for Jianqing Huang [Opens in a new window] ,
Hecong Liu  and
Weiwei Cai Open the ORCID record for Weiwei Cai [Opens in a new window]
Jianqing Huang
Affiliation:
Key Lab of Education Ministry for Power Machinery and Engineering, School of Mechanical Engineering, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, China
Hecong Liu
Affiliation:
Key Lab of Education Ministry for Power Machinery and Engineering, School of Mechanical Engineering, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, China
Weiwei Cai*
Affiliation:
Key Lab of Education Ministry for Power Machinery and Engineering, School of Mechanical Engineering, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, China
*
†Email address for correspondence: cweiwei@sjtu.edu.cn
Get access Rights & Permissions [Opens in a new window]

Abstract

Online in situ prediction of 3-D flame evolution has been long desired and is considered to be the Holy Grail for the combustion community. Recent advances in computational power have facilitated the development of computational fluid dynamics (CFD), which can be used to predict flame behaviours. However, the most advanced CFD techniques are still incapable of realizing online in situ prediction of practical flames due to the enormous computational costs involved. In this work, we aim to combine the state-of-the-art experimental technique (that is, time-resolved volumetric tomography) with deep learning algorithms for rapid prediction of 3-D flame evolution. Proof-of-concept experiments conducted suggest that the evolution of both a laminar diffusion flame and a typical non-premixed turbulent swirl-stabilized flame can be predicted faithfully in a time scale on the order of milliseconds, which can be further reduced by simply using a few more GPUs. We believe this is the first time that online in situ prediction of 3-D flame evolution has become feasible, and we expect this method to be extremely useful, as for most application scenarios the online in situ prediction of even the large-scale flame features are already useful for an effective flame control.

JFM classification

Mathematical Foundations: Computational methods Reacting Flows: Combustion Reacting Flows: Flames
Type
JFM Rapids
Information
Journal of Fluid Mechanics , Volume 875 , 25 September 2019 , R2
Copyright
© 2019 Cambridge University Press 

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

Alathur Srinivasan, P. A.2018 Deep learning models for turbulent shear flow. Thesis, KTH Royal Institute of Technology.Google Scholar
Augenstein, I., Rocktäschel, T., Vlachos, A. & Bontcheva, K.2016 Stance detection with bidirectional conditional encoding. arXiv:1606.05464.Google Scholar
Bakkouri, I. & Afdel, K. 2017 Breast tumor classification based on deep convolutional neural networks. In International Conference on Advanced Technologies for Signal & Image Processing. IEEE.Google Scholar
Baum, E., Peterson, B., Surmann, C., Michaelis, D., Böhm, B. & Dreizler, A. 2013 Investigation of the 3D flow field in an IC engine using tomographic PIV. Proc. Combust. Inst. 34 (2), 2903–2910.Google Scholar
Brunton, S. L. & Noack, B. R. 2015 Closed-loop turbulence control: progress and challenges. Appl. Mech. Rev. 67 (5), 050801.Google Scholar
Curran, H. J. 2019 Developing detailed chemical kinetic mechanisms for fuel combustion. Proc. Combust. Inst. 37 (1), 57–81.Google Scholar
Duraisamy, K., Iaccarino, G. & Xiao, H. 2019 Turbulence modeling in the age of data. Annu. Rev. Fluid Mech. 51, 357–377.Google Scholar
Fang, B., Zhang, Z., Li, G., Tao, B., Wang, S., Hu, Z. & Song, M. 2019 Simple calibrated nonlinear excitation regime two-line atomic fluorescence thermometry. Opt. Lett. 44 (2), 227–230.Google Scholar
Floyd, J., Geipel, P. & Kempf, A. M. 2011 Computed tomography of chemiluminescence (ctc): Instantaneous 3D measurements and phantom studies of a turbulent opposed jet flame. Combust. Flame 158 (2), 376–391.Google Scholar
Halls, B. R., Gord, J. R., Meyer, T. R., Thul, D. J., Slipchenko, M. & Roy, S. 2017a 20-kHz-rate three-dimensional tomographic imaging of the concentration field in a turbulent jet. Proc. Combust. Inst. 36 (3), 4611–4618.Google Scholar
Halls, B. R., Hsu, P. S., Jiang, N., Legge, E. S., Felver, J. J., Slipchenko, M. N., Roy, S., Meyer, T. R. & Gord, J. R. 2017b kHz-rate four-dimensional fluorescence tomography using an ultraviolet-tunable narrowband burst-mode optical parametric oscillator. Optica 4 (8), 897–902.Google Scholar
Halls, B. R., Hsu, P. S., Roy, S., Meyer, T. R. & Gord, J. R. 2018 Two-color volumetric laser-induced fluorescence for 3D OH and temperature fields in turbulent reacting flows. Opt. Lett. 43 (12), 2961–2964.Google Scholar
Halls, B. R., Jiang, N., Meyer, T. R., Roy, S., Slipchenko, M. N. & Gord, J. R. 2017c 4D spatiotemporal evolution of combustion intermediates in turbulent flames using burst-mode volumetric laser-induced fluorescence. Opt. Lett. 42 (14), 2830–2833.Google Scholar
Hasegawa, T., Nakamichi, R. & Nishiki, S. 2002 Mechanism of flame evolution along a fine vortex. Combust. Theor. Model. 6 (3), 413–424.Google Scholar
Hochreiter, S. & Schmidhuber, J. 1997 Long short-term memory. Neural Comput. 9 (8), 1735–1780.Google Scholar
Hult, J., Burns, I. S. & Kaminski, C. F. 2005 Two-line atomic fluorescence flame thermometry using diode lasers. Proc. Combust. Inst. 30 (1), 1535–1543.Google Scholar
Kaminski, C. F., Bai, X. S., Hult, J., Dreizler, A., Lindenmaier, S. & Fuchs, L. 2000 Flame growth and wrinkling in a turbulent flow. Appl. Phys. B 71 (5), 711–716.Google Scholar
Kashinath, K., Waugh, I. C. & Juniper, M. P. 2014 Nonlinear self-excited thermoacoustic oscillations of a ducted premixed flame: bifurcations and routes to chaos. J. Fluid Mech. 761, 399–430.Google Scholar
Klein, M., Chakraborty, N. & Ketterl, S. 2017 A comparison of strategies for direct numerical simulation of turbulence chemistry interaction in generic planar turbulent premixed flames. Flow Turbul. Combust. 99 (3), 955–971.Google Scholar
Kutz, J. N. 2017 Deep learning in fluid dynamics. J. Fluid Mech. 814, 1–4.Google Scholar
LeCun, Y., Bengio, Y. & Hinton, G. 2015 Deep learning. Nature 521 (7553), 436–444.Google Scholar
Li, N., Lu, G., Li, X. & Yan, Y. 2016 Prediction of NOx emissions from a biomass fired combustion process based on flame radical imaging and deep learning techniques. Combust. Sci. Technol. 188 (2), 233–246.Google Scholar
Litjens, G., Kooi, T., Bejnordi, B. E., Setio, A. A. A., Ciompi, F., Ghafoorian, M., van der Laak, J. A. W. M., van Ginneken, B. & Sánchez, C. I. 2017 A survey on deep learning in medical image analysis. Med. Image Anal. 42, 60–88.Google Scholar
Ma, L., Lei, Q., Wu, Y., Xu, W., Ombrello, T. M. & Carter, C. D. 2016 From ignition to stable combustion in a cavity flameholder studied via 3D tomographic chemiluminescence at 20 kHz. Combust. Flame 165, 1–10.Google Scholar
Mohan, A. T. & Gaitonde, D. V.2018 A deep learning based approach to reduced order modeling for turbulent flow control using LSTM neural networks. arXiv:1804.09269.Google Scholar
Nelson, D. M. Q., Pereira, A. C. M. & Oliveira, R. A. D. 2017 Stock market’s price movement prediction with LSTM neural networks. In Intl Joint Conference on Neural Networks, pp. 1419–1426. IEEE.Google Scholar
Noack, B. R., Morzyski, M. & Tadmor, G. 2011 Reduced-Order Modelling for Flow Control. Springer.Google Scholar
Qing, X. & Niu, Y. 2018 Hourly day-ahead solar irradiance prediction using weather forecasts by LSTM. Energy 148, 461–468.Google Scholar
Ruan, C., Yu, T., Chen, F., Wang, S., Cai, W. & Lu, X. 2019 Experimental characterization of the spatiotemporal dynamics of a turbulent flame in a gas turbine model combustor using computed tomography of chemiluminescence. Energy 170, 744–751.Google Scholar
Shi, L. L., Liu, Y. Z. & Yu, J. 2010 PIV measurement of separated flow over a blunt plate with different chord-to-thickness ratios. J. Fluids Struct. 26 (4), 644–657.Google Scholar
Sohn, C. H., Kim, J. S., Chung, S. H. & Maruta, K. 2000 Nonlinear evolution of diffusion flame oscillations triggered by radiative heat loss. Combust. Flame 123 (1), 95–106.Google Scholar
Tóth, P., Garami, A. & Csordás, B. 2017 Image-based deep neural network prediction of the heat output of a step-grate biomass boiler. Appl. Energy 200, 155–169.Google Scholar
Wang, Z., Song, C. & Chen, T. 2017 Deep learning based monitoring of furnace combustion state and measurement of heat release rate. Energy 131, 106–112.Google Scholar
Wang, Z., Xiao, D., Fang, F., Govindan, R., Pain, C. C. & Guo, Y. 2018 Model identification of reduced order fluid dynamics systems using deep learning. Intl J. Numer. Meth. Fluids 86 (4), 255–268.Google Scholar
Xu, Y., Pei, J. & Lai, L. 2017 Deep learning based regression and multi-class models for acute oral toxicity prediction with automatic chemical feature extraction. J. Chem. Inform. Model 57 (11), 2672–2685.Google Scholar
Yi, L., Yu, F. & Chen, J. 2017 Flame images for oxygen content prediction of combustion systems using DBN. Energy Fuels 31 (8), 8776–8783.Google Scholar
You, Y., Zhang, Z., Hsieh, C. J., Demmel, J. & Keutzer, K. 2017 ImageNet training in minutes. In 47th International Conference on Parallel Processing. ACM.Google Scholar
Yu, T., Ruan, C., Liu, H., Cai, W. & Lu, X. 2018 Time-resolved measurements of a swirl flame at 4 kHz via computed tomography of chemiluminescence. Appl. Opt. 57 (21), 5962–5969.Google Scholar

Huang et al. supplementary movie 1

The evolution of Flame #1

Download Huang et al. supplementary movie 1(Video)
Video 869 KB

Huang et al. supplementary movie 2

The evolution of Flame #2

Download Huang et al. supplementary movie 2(Video)
Video 830.5 KB
Supplementary material: File

Huang et al. supplementary material

Supplementary material

Download Huang et al. supplementary material(File)
File 65.4 KB