A Deep Learning Approach to Identifying Shock Locations in Turbulent Combustion Tensor Fields

  • Mathew Monfort
  • Timothy Luciani
  • Jonathan Komperda
  • Brian Ziebart
  • Farzad Mashayek
  • G. Elisabeta MaraiEmail author
Conference paper
Part of the Mathematics and Visualization book series (MATHVISUAL)


We introduce a deep learning approach for the identification of shock locations in large scale tensor field datasets. Such datasets are typically generated by turbulent combustion simulations. In this proof of concept approach, we use deep learning to learn mappings from strain tensors to Schlieren images which serve as labels. The use of neural networks allows for the Schlieren values to be approximated more efficiently than calculating the values from the density gradient. In addition, we show that this approach can be used to predict the Schlieren values for both two-dimensional and three-dimensional tensor fields, potentially allowing for anomaly detection in tensor flows. Results on two shock example datasets show that this approach can assist in the extraction of features from reacting flow tensor fields.



This work was partially supported by the National Science Foundation through award NSF CAREER IIS-1541277. We gratefully acknowledge the NSF Graduate Research Fellowship program for supporting Tim. We thank Adrian Maries and Shiwangi Singh for the initial literature search, and the Electronic Visualization Lab for their feedback and support. We further thank the Peyman Lab for the original motivation behind this line of work, and the Dagstuhl 16142 and 14082 seminars run by the Leibniz Center for Informatics for the many useful discussions regarding tensor field analysis and visualization .


  1. 1.
    Abbassi, H., Komperda, J., Mashayek, F., Jacobs, G.: Application of entropy viscosity method for supersonic flow simulation using discontinuous spectral element method. AIAA Paper 2012-1115 (2012)Google Scholar
  2. 2.
    Abbassi, H., Mashayek, F., Jacobs, G.: Entropy viscosity approach for compressible turbulent simulations using discontinuous spectral element method. AIAA Paper 2014-0947 (2014)Google Scholar
  3. 3.
    Abbassi, H., Mashayek, F., Jacobs, G.: Shock capturing with entropy-based artificial viscosity for staggered grid discontinuous spectral element method. Comput. Fluids 98, 152–163 (2014)MathSciNetCrossRefGoogle Scholar
  4. 4.
    Banaeizadeh, A., Li, Z., Jaberi, F.: Compressible scalar filtered mass density function model for high-speed turbulent flows. AIAA J. 49(10), 2130–2143 (2011)CrossRefGoogle Scholar
  5. 5.
    Bengio, Y., Lecun, Y.: Scaling Learning Algorithms Toward AI, pp. 321–359. MIT Press, Cambridge (2007)Google Scholar
  6. 6.
    Berglund, M., Fureby, C.: Les of supersonic combustion in a scramjet engine model. In: Proceedings of the Combustion Institute, vol. 31, pp. 2497–2504. The Combustion Institute, Pittsburgh, PA (2007)Google Scholar
  7. 7.
    Bergstra, J., Bengio, Y.: Random search for hyper-parameter optimization. J. Mach. Learn. Res. 13, 281–305 (2012)MathSciNetzbMATHGoogle Scholar
  8. 8.
    Bermejo-Moreno, I.: Subgrid-scale modeling of shock-turbulence interaction for large-eddy simulation. Annual Research Briefs, Center for Turbulence Research, Stanford University, Stanford, CA (2009)Google Scholar
  9. 9.
    Bourlard, H., Kamp, Y.: Auto-association by multilayer perceptrons and singular value decomposition. Biol. Cybernet. 59(4), 291–294 (1988)MathSciNetCrossRefzbMATHGoogle Scholar
  10. 10.
    Caban, J., Joshi, A., Rheingans, P.: Texture-based feature tracking for effective time-varying data visualization. IEEE Trans. Vis. Comput. Graph. 13(6), 1472–1479 (2007)CrossRefGoogle Scholar
  11. 11.
    Clevert, D., Unterthiner, T., Hochreiter, S.: Fast and accurate deep network learning by exponential linear units (ELUs). CoRR abs/1511.07289 (2015)Google Scholar
  12. 12.
    Collobert, R., Kavukcuoglu, K., Farabet, C.: Torch7: A matlab-like environment for machine learning. In: BigLearn, NIPS Workshop (2011)Google Scholar
  13. 13.
    Dimitrov, L.I.: Pseudo-colored visualization of EEG-activities on the human cortex using MRI-based volume rendering and Delaunay interpolation. Medical Imaging, 460–469 (1995)Google Scholar
  14. 14.
    Ducros, F., Ferrand, V., Nicoud, F., Weber, C., Darracq, D., Gacherieu, C., Poinsot, T.: Large-eddy simulation of the shock/turbulence interaction. J. Comput. Phys. 152(2), 517–549 (1999)CrossRefzbMATHGoogle Scholar
  15. 15.
    Edwards, J.R., Potturi, A., Fulton, J.A.: Large-eddy / reynolds-averaged navier-stokes simulations of scramjet combustor flow fields. AIAA Paper 2012–4262 (2012)Google Scholar
  16. 16.
    Fulton, J.A., Edwards, J.R., Hassan, H.A., McDaniel, J.C., Goyne, C.P., Rockwell, R.D., Cutler, A.D., Johansen, C.T., Danehy, P.M.: Large-eddy/reynolds-averaged Navier-Stokes simulations of reactive flow in dual-mode scramjet combustor. J. Propuls. Power 30(3), 558–575 (2013)CrossRefGoogle Scholar
  17. 17.
    Ghiasi, Z., Komperda, J., Li, D., Mashayek, F.: Simulation of supersonic turbulent non-reactive flow in ramp-cavity combustor using a discontinuous spectral element method. AIAA Paper 2016-0617 (2016)Google Scholar
  18. 18.
    Graves, A., Liwicki, M., Fernández, S., Bertolami, R., Bunke, H., Schmidhuber, J.: A novel connectionist system for unconstrained handwriting recognition. IEEE Trans. Pattern Anal. Mach. Intell. 31(5), 855–868 (2009)CrossRefGoogle Scholar
  19. 19.
    Hadjadj, A., Kudryavtsev, A.: Computation and flow visualization in high-speed aerodynamics. J. Turbul. 6, 16 (2005)CrossRefGoogle Scholar
  20. 20.
    Hebb, D.O.: The Organization of Behavior. Wiley, New York (1949)Google Scholar
  21. 21.
    Hinton, G.E., Salakhutdinov, R.R.: Reducing the dimensionality of data with neural networks. Science 313(5786), 504–507 (2006)MathSciNetCrossRefzbMATHGoogle Scholar
  22. 22.
    Ioffe, S., Szegedy, C.: Batch normalization: accelerating deep network training by reducing internal covariate shift. CoRR abs/1502.03167 (2015)Google Scholar
  23. 23.
    Irannejad, A., Jaberi, F., Komperda, J., Mashayek, F.: Large eddy simulation of supersonic turbulent combustion with FMDF. AIAA Paper, 1188, p. 2014 (2014)Google Scholar
  24. 24.
    Ji, G., Shen, H.-W., Wenger, R.: Volume tracking using higher dimensional isosurfacing. In: Proceedings of the 14th IEEE Visualization 2003 (VIS’03), Washington, DC, p. 28. IEEE Computer Society (2003)Google Scholar
  25. 25.
    Ju, Y., Shimano, A., Inoue, O.: Vorticity generation and flame distortion induced by shock flame interaction. In: Proceedings of the Combustion Institute, vol. 27, pp. 735–741. The Combustion Institute, Boulder, CO (1998)Google Scholar
  26. 26.
    Khokhlov, A.M., Oran, E.S., Thomas, G.O.: Numerical simulation of deflagration-to-detonation transition: the role of shock–flame interactions in turbulent flames. Combust. Flame 117(1–2), 323–339 (1999)CrossRefGoogle Scholar
  27. 27.
    Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. In: Advances in Neural Information Processing Systems (2012)Google Scholar
  28. 28.
    Larsson, J., Vicquelin, R., Bermejo-Moreno, I.: Large eddy simulations of the HyShot II scramjet. Annual Research Briefs, Center for Turbulence Research, Stanford University, Stanford, CA (2011)Google Scholar
  29. 29.
    LeCun, Y., Bengio, Y.: Convolutional networks for images, speech, and time series. In: Arbib, M.A. (ed.) The Handbook of Brain Theory and Neural Networks, pp. 255–258. MIT Press, Cambridge, MA (1998)Google Scholar
  30. 30.
    Lee, V.W., Kim, C., Chhugani, J., Deisher, M., Kim, D., Nguyen, A.D., Satish, N., Smelyanskiy, M., Chennupaty, S., Hammarlund, P., et al.: Debunking the 100× GPU vs. CPU myth: an evaluation of throughput computing on CPU and GPU. ACM SIGARCH Comput. Archit. News 38(3), 451–460 (2010)Google Scholar
  31. 31.
    Li, Z., Banaeizadeh, A., Rezaeiravesh, S., Jaberi, F.: Advanced modeling of high speed turbulent reacting flows. AIAA Paper 2012-116 (2012)Google Scholar
  32. 32.
    Lovely, D., Haimes, H.: Shock detection from computational fluid dynamics results. In: 14th Computational Fluid Dynamics Conference, Cambridge, MA, pp. 255–258. MIT Press, Cambridge (1999)Google Scholar
  33. 33.
    Ma, K.-L., Rosendale, J.V., Vermeer, W.: 3d shock wave visualization on unstructured grids. In: VVS, p. 87 (1996)Google Scholar
  34. 34.
    Marai, G.E., Luciani, T., Maries, A., Yilmaz, S.L., Nik, M.B.: Visual descriptors for dense tensor fields in computational turbulent combustion: a case study. In: Visualization and Data Analysis, pp. 1–11. Ingenta, Oxford (2016)Google Scholar
  35. 35.
    Maries, A., Luciani, T., Pisciuneri, P.H., Nik, M.B., Yilmaz, S.L., Givi, P., Marai, G.E.: A clustering method for identifying regions of interest in turbulent combustion tensor fields. In: Visualization and Processing of Higher Order Descriptors for Multi-Valued Data, pp. 323–338. Springer, Cham (2015)Google Scholar
  36. 36.
    Matsugu, M., Mori, K., Mitari, Y., Kaneda, Y.: Subject independent facial expression recognition with robust face detection using a convolutional neural network. Neural Netw. 16(5–6), 555–559 (2003)CrossRefGoogle Scholar
  37. 37.
    Maturana, D., Scherer, S.: Voxnet: a 3d convolutional neural network for real-time object recognition. In: 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 922–928. IEEE, New York (2015)Google Scholar
  38. 38.
    McCulloch, W.S., Pitts, W.: A logical calculus of the ideas immanent in nervous activity. Bull. Math. Biophys. 5(4), 115–133 (1943)MathSciNetCrossRefzbMATHGoogle Scholar
  39. 39.
    Milletari, F., Navab, N., Ahmadi, S.: V-net: fully convolutional neural networks for volumetric medical image segmentation. CoRR abs/1606.04797 (2016)Google Scholar
  40. 40.
    Mynski, M.L., Papert, S.A.: Perceptrons: An Introduction to Computational Geometry. MIT Press, Cambridge, MA (1969)zbMATHGoogle Scholar
  41. 41.
    Noh, H., Hong, S., Han, B.: Learning deconvolution network for semantic segmentation. CoRR abs/1505.04366 (2015)Google Scholar
  42. 42.
    Orbach, J.: Principles of neurodynamics. perceptrons and the theory of brain mechanisms. Arch. Gen. Psychiatr. 7(3), 218–219 (1962)Google Scholar
  43. 43.
    Ozer, S., Wei, J., Silver, D., Martin, P.: Group dynamics in scientific visualization. in large data analysis and visualization (LDAV). In: 2012 IEEE Symposium on Large Data Analysis and Visualization (LDAV), October 2012, pp. 97–104Google Scholar
  44. 44.
    Pagendarm, H.-G., Walter, B.: Feature detection from vector quantities in a numerically simulated hypersonic flow field in combination with experimental flow visualization. In: Proceedings of the Conference on Visualization ’94, VIS ’94, pp. 117–123, Los Alamitos, CA. IEEE Computer Society Press (1994)Google Scholar
  45. 45.
    Pascanu, R., Mikolov, T., Bengio, Y.: Understanding the exploding gradient problem. CoRR abs/1211.5063 (2012)Google Scholar
  46. 46.
    Peterson, D.M., Hagenmaier, M., Carter, C.D., Tuttle, S.G.: Hybrid reynolds-averaged and large-eddy simulations of a supersonic cavity flameholder. AIAA Paper 2013–2483 (2013)Google Scholar
  47. 47.
    Qi, C.R., Su, H., Nießner, M., Dai, A., Yan, M., Guibas, L.J.: Volumetric and multi-view cnns for object classification on 3d data. CoRR abs/1604.03265 (2016)Google Scholar
  48. 48.
    Roy, C.J., Edwards, J.R.: Numerical simulation of a three-dimensional flame/shock wave interaction. AIAA J. 38(5), 745–754 (2000)CrossRefGoogle Scholar
  49. 49.
    Rumelhart, D.E., Hinton, G.E., Williams, R.J.: Learning representations by back-propagating errors. In: Anderson, J.A., Rosenfeld, E. (eds.) Neurocomputing: Foundations of Research, pp. 696–699. MIT Press, Cambridge, MA, (1988)Google Scholar
  50. 50.
    Samtaney, R., Silver, D., Zabusky, N., Cao, J.: Visualizing features and tracking their evolution. IEEE Comput. 27(7), 20–27 (1994)CrossRefGoogle Scholar
  51. 51.
    Silver, D.: Object-oriented visualization. IEEE Comput. Graph. Appl. 15(3), 54–62 (1)Google Scholar
  52. 52.
    Silver, D., Wang, X.: Tracking and visualizing turbulent 3d features. IEEE Trans. Vis. Comput. Graph. 3(2), 129–141 (1997)CrossRefGoogle Scholar
  53. 53.
    Springenberg, J.T., Dosovitskiy, A., Brox, T., Riedmiller, M.A.: Striving for simplicity: the all convolutional net. CoRR abs/1412.6806 (2014)Google Scholar
  54. 54.
    Springenberg, J.T., Dosovitskiy, A., Brox, T., Riedmiller, M.A.: Striving for simplicity: the all convolutional net. CoRR abs/1412.6806 (2014)Google Scholar
  55. 55.
    Srinivas, S., Sarvadevabhatla, R.K., Mopuri, K.R., Prabhu, N., Kruthiventi, S.S.S., Babu, R.V.: A taxonomy of deep convolutional neural nets for computer vision. Front. Robot. AI 2, 36 (2016)CrossRefGoogle Scholar
  56. 56.
    Tompson, J., Goroshin, R., Jain, A., LeCun, Y., Bregler, C.: Efficient object localization using convolutional networks. CoRR abs/1411.4280 (2014)Google Scholar
  57. 57.
    Toro, E.F.: Riemann Solvers and Numerical Methods for Fluid Dynamics, 3rd edn. Springer, Berlin (2009)CrossRefzbMATHGoogle Scholar
  58. 58.
    Tzeng, F.-Y., Ma, K.-L.: Intelligent feature extraction and tracking for visualizing large-scale 4d flow simulations. In: Proceedings of the 2005 ACM/IEEE Conference on Supercomputing, SC ’05, Washington, DC, p. 6. IEEE Computer Society (2005)Google Scholar
  59. 59.
    Zeiler, M.D.: ADADELTA: an adaptive learning rate method. CoRR abs/1212.5701 (2012)Google Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  • Mathew Monfort
    • 1
  • Timothy Luciani
    • 1
  • Jonathan Komperda
    • 1
  • Brian Ziebart
    • 1
  • Farzad Mashayek
    • 1
  • G. Elisabeta Marai
    • 1
    Email author
  1. 1.University of Illinois at ChicagoChicagoUSA

Personalised recommendations