Abstract
High-speed planar imaging of key combustion species, like hydroxyl radicals (OH), is crucial for understanding the complex chemistry–turbulence interactions in turbulent flames. However, conducting high-speed (kHz) diagnostics is challenging due to the requirements on advanced optical system, including both fast lasers and cameras. In this paper, we report a computational imaging method to artificially achieve higher diagnosing rates based on experimental data at relatively low rates. Sequencies of planar laser-induced fluorescence (PLIF) of OH recorded at 100 kHz in a turbulent flame were first down sampled to 50 kHz, 33.3 kHz and 20 kHz, respectively, and then used as a data source to train several networks. The accuracies of the models were assessed by comparing the predicted images with those from laser measurements. It was found that, among the models tested, convolutional long short-term memory network (CONV-LSTM) can provide the best predictions and is reliable in predicting consecutive images with higher repetition rates. The model can also generate consecutive OH-PLIF images at 200 kHz based on the 100 kHz experimental data. This work sheds light on the hybrid of deep learning-based computational methods with conventional high-speed laser diagnosing techniques, which can potentially increase the temporal resolution of planar optical measurement in turbulent flows, and significantly reduce the computational timing.
Similar content being viewed by others
References
K. Kohse-Höinghaus et al., Combustion at the focus: laser diagnostics and control. Proc. Combust. Inst. 30(1), 89–123 (2005)
V. Sick, High speed imaging in fundamental and applied combustion research. Proc. Combust. Inst. 34(2), 3509–3530 (2013)
R.K. Hanson, J.M. Seitzman, P.H. Paul, Planar laser-fluorescence imaging of combustion gases. Appl. Phys. B 50(6), 441–454 (1990)
G. Grünefeld, M. Schütte, P. Andresen, Simultaneous multiple-line Raman/Rayleigh/LIF measurements in combustion. Appl. Phys. B 70(2), 309–313 (2000)
U. Retzer et al., Burst-mode OH/CH2O planar laser-induced fluorescence imaging of the heat release zone in an unsteady flame. Opt. Express 26(14), 18105–18114 (2018)
M.N. Slipchenko, T.R. Meyer, S. Roy, Advances in burst-mode laser diagnostics for reacting and nonreacting flows. Proc. Combust. Inst. 38(1), 1533–1560 (2021)
M.E. Smyser et al., Compact burst-mode Nd:YAG laser for kHz–MHz bandwidth velocity and species measurements. Opt. Lett. 43(4), 735–738 (2018)
C. Yang, H. Tang, G. Magnotti, High-speed 1D Raman analyzer for temperature and major species measurements in a combustion environment. Opt. Lett. 45(10), 2817–2820 (2020)
S. Roy et al., 100-ps-pulse-duration, 100-J burst-mode laser for kHz–MHz flow diagnostics. Opt. Lett. 39(22), 6462–6465 (2014)
M.N. Slipchenko et al., Quasi-continuous burst-mode laser for high-speed planar imaging. Opt. Lett. 37(8), 1346–1348 (2012)
C. Fu et al., Experimental investigation on an acoustically forced flame with simultaneous high-speed LII and stereo PIV at 20 kHz. Appl. Opt. 58(10), C104–C111 (2019)
J.B. Michael et al., 100 kHz thousand-frame burst-mode planar imaging in turbulent flames. Opt Lett 39(4), 739–742 (2014)
T.A. McManus et al., Spatio-temporal characteristics of temperature fluctuations in turbulent non-premixed jet flames. Proc. Combust. Inst. 35(2), 1191–1198 (2015)
R.A. Patton et al., Multi-kHz temperature imaging in turbulent non-premixed flames using planar Rayleigh scattering. Appl. Phys. B 108(2), 377–392 (2012)
S. Roy et al., 100-kHz-rate gas-phase thermometry using 100-ps pulses from a burst-mode laser. Opt Lett 40(21), 5125–5128 (2015)
J.D. Miller et al., Spatiotemporal analysis of turbulent jets enabled by 100-kHz, 100-ms burst-mode particle image velocimetry. Exp. Fluids 57(12), 192 (2016)
J.J. Philo, M.D. Frederick, C.D. Slabaugh, 100 kHz PIV in a liquid-fueled gas turbine swirl combustor at 1 MPa. Proc. Combust. Inst. 38(1), 1571–1578 (2021)
B. Peterson et al., An experimental study of the detailed flame transport in a SI engine using simultaneous dual-plane OH-LIF and stereoscopic PIV. Combust. Flame 202, 16–32 (2019)
I. Boxx et al., High-speed laser diagnostics for the study of flame dynamics in a lean premixed gas turbine model combustor. Exp. Fluids 52(3), 555–567 (2012)
R. Wellander, M. Richter, M. Aldén, Time-resolved (kHz) 3D imaging of OH PLIF in a flame. Exp. Fluids 55(6), 1764 (2014)
S.D. Hammack et al., CH PLIF and PIV implementation using C-X (0,0) and intra-vibrational band filtered detection. Appl. Phys. B (2018). https://doi.org/10.1007/s00340-017-6883-8
J. Sjöholm et al., Ultra-high-speed pumping of an optical parametric oscillator (OPO) for high-speed laser-induced fluorescence measurements. Measur Sci Technol 20(2), 025306 (2009)
J. Sjöholm et al., Simultaneous visualization of OH, CH, CH2O and toluene PLIF in a methane jet flame with varying degrees of turbulence. Proc. Combust. Inst. 34(1), 1475–1482 (2013)
Z. Wang et al., Ultra-high-speed PLIF imaging for simultaneous visualization of multiple species in turbulent flames. Opt. Express 25(24), 30214–30228 (2017)
J.D. Miller et al., Ultrahigh-frame-rate OH fluorescence imaging in turbulent flames using a burst-mode optical parametric oscillator. Opt. Lett. 34(9), 1309–1311 (2009)
W. Zhang et al., Generating planar distributions of soot particles from luminosity images in turbulent flames using deep learning. Appl. Phys. B (2021). https://doi.org/10.1007/s00340-020-07571-9
C.-S. Liu, R.-C. Song, S.-J. Fu, Design of a laser-based autofocusing microscope for a sample with a transparent boundary layer. Appl. Phys. B 125(11), 199 (2019)
W. Zhang et al., 100 kHz CH2O imaging realized by lower speed planar laser-induced fluorescence and deep learning. Opt. Express 29(19), 30857–30877 (2021)
T. Li, Z. Zhang, H. Chen, Predicting the combustion state of rotary kilns using a Convolutional Recurrent Neural Network. J. Process Control 84, 207–214 (2019)
Z.C. Lipton, J. Berkowitz, C. Elkan, A critical review of recurrent neural networks for sequence learning.. 2015. arXiv e-prints. arXiv:1506.00019
S. Hochreiter, J. Schmidhuber, Long short-term memory. Neural Comput. 9(8), 1735–1780 (1997)
J. Chung, et al., Empirical evaluation of gated recurrent neural networks on sequence modeling., 2014. arXiv e-prints. arXiv:1412.3555.
A. Krizhevsky, I. Sutskever, G.E. Hinton, ImageNet classification with deep convolutional neural networks. Commun. ACM 60(6), 84–90 (2017)
Z. Leihong et al., Restoration of single pixel imaging in atmospheric turbulence by fourier filter and CGAN. Appl. Phys. B (2021). https://doi.org/10.1007/s00340-021-07596-8
J. Li et al., Object identification in computational ghost imaging based on deep learning. Appl. Phys. B (2020). https://doi.org/10.1007/s00340-020-07514-4
S. Hong, et al. PSIque: Next sequence prediction of satellite images using a convolutional sequence-to-sequence network, 2017. .arXiv:1711.10644
X. Shi, et al., Convolutional LSTM network: a machine learning approach for precipitation nowcasting, 2015. arXiv:1506.04214
S. Kim, et al., DeepRain: ConvLSTM network for precipitation prediction using multichannel radar data, 2017. arXiv:1711.02316
C. Finn, I. Goodfellow, S. Levine, Unsupervised learning for physical interaction through video prediction, 2016. arXiv:1605.07157
W. Lotter, G. Kreiman, D. Cox, Deep predictive coding networks for video prediction and unsupervised learning, 2016. arXiv:1605.08104
V. Patraucean, A. Handa, R. Cipolla, Spatio-temporal video autoencoder with differentiable memory, 2015. arXiv:1511.06309
Z. Wang et al., Investigation of OH and CH2O distributions at ultra-high repetition rates by planar laser induced fluorescence imaging in highly turbulent jet flames. Fuel 234, 1528–1540 (2018)
R. Hanson, Combustion diagnostics: planar imaging techniques. Symp (Int) Combust 21, 1677–1691 (1988)
J. Rosell et al., Multi-species PLIF study of the structures of turbulent premixed methane/air jet flames in the flamelet and thin-reaction zones regimes. Combust. Flame 182, 324–338 (2017)
B. Zhou et al., Distributed reactions in highly turbulent premixed methane/air flames: Part I Flame structure characterization. Combust. Flame 162(7), 2937–2953 (2015)
B. Zhou et al., Simultaneous multi-species and temperature visualization of premixed flames in the distributed reaction zone regime. Proc. Combust. Inst. 35(2), 1409–1416 (2015)
D.-A. Clevert, T. Unterthiner, S. Hochreiter, Fast and accurate deep network learning by exponential linear units (elus) 2015. arXiv preprint arXiv:1511.07289
Klambauer, G., et al. Self-normalizing neural networks. In 31st Annual Conference on Neural Information Processing Systems, NIPS 2017, December 4, 2017 - December 9, 2017. 2017. Long Beach, CA, United states: Neural information processing systems foundation.
V. Nair, G.E. Hinton, Rectified linear units improve Restricted Boltzmann machines. In 27th International Conference on Machine Learning, ICML 2010, June 21, 2010 - June 25, 2010. 2010. Haifa, Israel: Unavailable.
I. Sutskever, O. Vinyals, Q.V. Le, Sequence to sequence learning with neural networks. In 28th Annual Conference on Neural Information Processing Systems 2014, NIPS 2014, December 8, 2014–December 13, 2014. 2014. Montreal, QC, Canada: Neural information processing systems foundation.
N. Srivastava, E. Mansimov, R. Salakhutdinov. Unsupervised learning of video representations using LSTMs. In 32nd International Conference on Machine Learning, ICML 2015, July 6, 2015–July 11, 2015. 2015. Lile, France: International Machine Learning Society (IMLS).
D.E. Rumelhart, G.E. Hinton, R.J. Williams, Learning representations by back-propagating errors. Nature 323(6088), 533–536 (1986)
Y. LeCun, Y. Bengio, G. Hinton, Deep learning. Nature 521(7553), 436–444 (2015)
D.P. Kingma, J. Ba, Adam: A Method for Stochastic Optimization, 2014. arXiv:1412.6980
W. Zhou et al., Image quality assessment: from error visibility to structural similarity. IEEE Trans. Image Process. 13(4), 600–612 (2004)
J. Redmon, et al. You Only Look Once: Unified, Real-Time Object Detection. In 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2016.
H. Belmabrouk, M. Michard, Taylor length scale measurement by laser Doppler velocimetry. Exp. Fluids 25(1), 69–76 (1998)
J.I. Liang, J. Piper, J.Y. Tang, Erosion and dilation of binary images by arbitrary structuring elements using interval coding. Pattern Recogn. Lett. 9(3), 201–209 (1989)
W. Lorensen, H. Cline, Marching cubes: a high resolution 3d surface construction algorithm. ACM SIGGRAPH Comput. Graph. 21, 163 (1987)
W. Kesheng, O. Ekow, S. Arie. Optimizing connected component labeling algorithms. In Proc. SPIE. 2005.
Acknowledgements
This research was financially supported by the National Natural Science Foundation of China under Grant No. 52006137, as well as Shanghai Sailing Program, under Grant No. 19YF1423400.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Guo, H., Zhang, W., Nie, X. et al. High-speed planar imaging of OH radicals in turbulent flames assisted by deep learning. Appl. Phys. B 128, 52 (2022). https://doi.org/10.1007/s00340-021-07742-2
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s00340-021-07742-2