Skip to main content Accessibility help

Super-resolution reconstruction of turbulent flows with machine learning

  • Kai Fukami (a1) (a2) (a3), Koji Fukagata (a1) and Kunihiko Taira (a2) (a3)


We use machine learning to perform super-resolution analysis of grossly under-resolved turbulent flow field data to reconstruct the high-resolution flow field. Two machine learning models are developed, namely, the convolutional neural network (CNN) and the hybrid downsampled skip-connection/multi-scale (DSC/MS) models. These machine learning models are applied to a two-dimensional cylinder wake as a preliminary test and show remarkable ability to reconstruct laminar flow from low-resolution flow field data. We further assess the performance of these models for two-dimensional homogeneous turbulence. The CNN and DSC/MS models are found to reconstruct turbulent flows from extremely coarse flow field images with remarkable accuracy. For the turbulent flow problem, the machine-leaning-based super-resolution analysis can greatly enhance the spatial resolution with as little as 50 training snapshot data, holding great potential to reveal subgrid-scale physics of complex turbulent flows. With the growing availability of flow field data from high-fidelity simulations and experiments, the present approach motivates the development of effective super-resolution models for a variety of fluid flows.


Corresponding author

Email address for correspondence:


Hide All
Colonius, T. & Taira, K. 2008 A fast immersed boundary method using a nullspace approach and multi-domain far-field boundary conditions. Comput. Meth. Appl. Mech. Engng 197, 21312146.
Dong, C., Loy, C. C., He, K. & Tang, X. 2016 Image super-resolution using deep convolutional networks. IEEE Trans. Pattern Anal. Mach. Intell. 38 (2), 295307.
Du, X., Qu, X., He, Y. & Guo, D. 2018 Single image super-resolution based on multi-scale competitive convolutional neural network. Sensors 18 (789), 117.
Duraisamy, K., Iaccarino, G. & Xiao, H. 2019 Turbulence modeling in the age of data. Annu. Rev. Fluid. Mech. 51, 357377.
Fukami, K., Kawai, K. & Fukagata, K. 2018 A synthetic turbulent inflow generator using machine learning. Phys. Rev. Fluids (submitted), arXiv:1806.08903.
He, K., Zhang, X., Ren, S. & Sun, J. 2016 Deep residual learning for image recognition. In Proceedings of Computer Vision and Pattern Recognition, pp. 770–778.
Hou, W., Darakananda, D. & Eldredge, J. D.2019 Machine learning based detection of flow disturbances using surface pressure measurements. AIAA Paper 2019–1148.
Keys, R. 1981 Cubic convolution interpolation for digital image processing. IEEE Trans. Acoust. Speech Signal Process. 29 (6), 11531160.
Kingma, D. & Ba, J.2014 A method for stochastic optimization. arXiv:1412.6980.
Koizumi, H., Tsutsumi, S. & Shima, E.2010 Feedback control of Karman vortex shedding from a cylinder using deep reinforcement learning. AIAA Paper 2018–3691.
Kutz, J. N. 2016 Deep learning in fluid dynamics. J. Fluid Mech. 814, 14.
Le, Q., Ngiam, J., Chen, Z., Chia, D. & Koh, P. 2010 Tiled convolutional neural networks. Adv. Neural Inform. Proc. Syst. 23, 12791287.
Lecun, Y., Bottou, L., Bengio, Y. & Haffner, P. 1998 Gradient-based learning applied to document recognition. Proc. IEEE 86 (11), 22782324.
Ling, J., Kurzawski, A. & Templeton, J. 2016 Reynolds averaged turbulence modelling using deep neural networks with embedded invariance. J. Fluid Mech. 807, 155166.
Leoni, P. C. D., Mazzino, A. & Biferale, L. 2018 Inferring flow parameters and turbulent configuration with physics-informed data assimilation and spectral nudging. Phys. Rev. Fluids 3, 104604.
Maulik, R. & San, O. 2017 A neural network approach for the blind deconvolution of turbulent flows. J. Fluid Mech. 831, 151181.
Nair, V. & Hinton, G. E. 2010 Rectified linear units improve restricted Boltzmann machines. In Proc. 27th International Conference on Machine Learning.
Prechelt, L. 1998 Automatic early stopping using cross validation: quantifying the criteria. Neural Networks 11 (4), 761767.
Romano, Y., Ishidoro, J. & Milanfar, P. 2017 RAISR: rapid and accurate image super resolution. IEEE Trans. Comput. Imaging 3 (1), 110125.
Salehipour, H. & Peltier, W. R. 2019 Deep learning of mixing by two ‘atoms’ of stratified turbulence. J. Fluid Mech. 861 (R4), 114.
San, O. & Maulik, R. 2018 Extreme learning machine for reduced order modeling of turbulent geophysical flows. Phys. Rev. E 97, 042322.
Shanker, M., Hu, M. Y. & Hung, M. S. 1996 Effect of data standardization on neural network training. Omega 24 (4), 385397.
Taira, K. & Colonius, T. 2007 The immersed boundary method: a projection approach. J. Comput. Phys. 225 (2), 21182137.
Taira, K., Nair, A. G. & Brunton, S. L. 2016 Network structure of two-dimensional decaying isotropic turbulence. J. Fluid Mech. 795 (R2), 111.
Zhang, Y., Sung, W. & Mavris, D.2018 Application of convolutional neural network to predict airfoil lift coefficient. AIAA Paper 2018–1903.
MathJax is a JavaScript display engine for mathematics. For more information see

JFM classification

Super-resolution reconstruction of turbulent flows with machine learning

  • Kai Fukami (a1) (a2) (a3), Koji Fukagata (a1) and Kunihiko Taira (a2) (a3)


Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed.