Abstract
The accurate reconstruction of global flow fields from sparse measurements has been a longstanding challenge in which the quantity and positioning of measurements play a critical role. To address this issue, we propose a global flow field reconstruction method based on a mode decomposition autoencoder, which maintains interpretability while effectively handling arbitrary quantities and positioning of sensors, ensuring high accuracy in the reconstruction of flow fields and other modal data. An autoencoder is trained on global flow fields to capture the nonlinear modes of the flow. The backpropagation capability of the deep network is leveraged to transform the flow field reconstruction problem into an interpretable optimization problem, which is solved to obtain the complete flow field. In experiments carried out on a stable ocean surface temperature dataset and an unstable multi-cylinder airflow dataset, the proposed method consistently achieved high accuracy across various flow fields, surpassing the performance of current approaches.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Dong, W., Chen, X., Yang, Q.: Data-driven scenario generation of renewable energy production based on controllable generative adversarial networks with interpretability. Appl. Energy 308, 118387 (2022)
Erichson, N.B., Mathelin, L., Yao, Z., Brunton, S.L., Mahoney, M.W., Kutz, J.N.: Shallow neural networks for fluid flow reconstruction with limited sensors. Proc. R. Soc. A: Math. Phys. Eng. Sci. 476(2238), 20200097 (2020)
Everson, R., Sirovich, L.: Optics InfoBase: Journal of the Optical Society of America A - Karhunen-Loève procedure for Gappy data. JOSA A 12(8), 2–9 (1995)
Fresca, S., Manzoni, A.: POD-DL-ROM: enhancing deep learning-based reduced order models for nonlinear parametrized PDEs by proper orthogonal decomposition. Comput. Methods Appl. Mech. Eng. 388, 114181 (2022)
Fukami, K., Maulik, R., Ramachandra, N., Fukagata, K., Taira, K.: Global field reconstruction from sparse sensors with Voronoi tessellation-assisted deep learning. Nat. Mach. Intell. 3(11), 945–951 (2021)
Giannopoulos, A., Aider, J.L.: Data-driven order reduction and velocity field reconstruction using neural networks: the case of a turbulent boundary layer. Phys. Fluids 32(9), 095117 (2020)
Hasegawa, K., Fukami, K., Murata, T., Fukagata, K.: CNN-LSTM based reduced order modeling of two-dimensional unsteady flows around a circular cylinder at different Reynolds numbers. Fluid Dyn. Res. 52(6), 065501 (2020)
Ishii, M., Shouji, A., Sugimoto, S., Matsumoto, T.: Objective analyses of sea-surface temperature and marine meteorological variables for the 20th century using ICOADS and the Kobe Collection. Int. J. Climatol. 25(7), 865–879 (2005)
Kazemi, A., Stoddard, M., Amini, A.A.: Reduced-order modeling of 4D flow MRI and CFD in stenotic flow using Proper Orthogonal Decomposition (POD) and Dynamic Mode Decomposition (DMD). In: Medical Imaging 2022: Biomedical Applications in Molecular, Structural, and Functional Imaging, vol. 12036, pp. 509–519. SPIE, April 2022
Kingma, D.P., Ba, J.: Adam: A Method for Stochastic Optimization, January 2017
Lario, A., Maulik, R., Schmidt, O.T., Rozza, G., Mengaldo, G.: Neural-network learning of SPOD latent dynamics. J. Comput. Phys. 468, 111475 (2022)
Lavrinov, V.V., Lavrinova, L.N.: Reconstruction of wavefront distorted by atmospheric turbulence using a Shack-Hartman sensor. Comput. Opt. 43, 586–595 (2019)
Leinonen, J., Nerini, D., Berne, A.: Stochastic super-resolution for downscaling time-evolving atmospheric fields with a generative adversarial network. IEEE Trans. Geosci. Remote Sens. 59(9), 7211–7223 (2021)
Li, C.Y., Tse, T.K.T., Hu, G.: Dynamic Mode Decomposition on pressure flow field analysis: flow field reconstruction, accuracy, and practical significance. J. Wind Eng. Ind. Aerodyn. 205, 104278 (2020)
Liu, N., et al.: Meshless surface wind speed field reconstruction based on machine learning. Adv. Atmos. Sci. 39(10), 1721–1733 (2022)
Liu, T., Li, Y., Jing, Q., Xie, Y., Zhang, D.: Supervised learning method for the physical field reconstruction in a nanofluid heat transfer problem. Int. J. Heat Mass Transf. 165, 120684 (2021)
Manfren, M., James, P.A., Tronchin, L.: Data-driven building energy modelling - an analysis of the potential for generalisation through interpretable machine learning. Renew. Sustain. Energy Rev. 167, 112686 (2022)
Maulik, R., Egele, R., Lusch, B., Balaprakash, P.: Recurrent neural network architecture search for geophysical emulation. In: SC20: International Conference for High Performance Computing, Networking, Storage and Analysis, pp. 1–14, November 2020
Murata, T., Fukami, K., Fukagata, K.: Nonlinear mode decomposition with convolutional neural networks for fluid dynamics. J. Fluid Mech. 882, A13 (2020)
Pascarella, G., Fossati, M., Barrenechea, G.: Impact of POD modes energy redistribution on flow reconstruction for unsteady flows of impulsively started airfoils and wings. Int. J. Comput. Fluid Dyn. 34(2), 108–118 (2020)
Sun, S., Liu, S., Chen, M., Guo, H.: An optimized sensing arrangement in wind field reconstruction using CFD and POD. IEEE Trans. Sustain. Energy 11(4), 2449–2456 (2020)
Wang, W., Wainwright, M.J., Ramchandran, K.: Information-theoretic limits on sparse signal recovery: dense versus sparse measurement matrices. IEEE Trans. Inf. Theory 56(6), 2967–2979 (2010)
Wang, Z., Gong, K., Fan, W., Li, C., Qian, W.: Prediction of swirling flow field in combustor based on deep learning. Acta Astronaut. 201, 302–316 (2022)
Yu, J., Hesthaven, J.S.: Flowfield reconstruction method using artificial neural network. AIAA J. 57(2), 482–498 (2019)
Acknowledgements
This research was supported by the National Key Research and Development Program of China (No. 2020YFB1709500).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Qiu, J., Yuan, W., Hu, X., Zhang, J., Chi, X. (2024). Sparse Reconstruction Method for Flow Fields Based on Mode Decomposition Autoencoder. In: Liu, F., Sadanandan, A.A., Pham, D.N., Mursanto, P., Lukose, D. (eds) PRICAI 2023: Trends in Artificial Intelligence. PRICAI 2023. Lecture Notes in Computer Science(), vol 14325. Springer, Singapore. https://doi.org/10.1007/978-981-99-7019-3_26
Download citation
DOI: https://doi.org/10.1007/978-981-99-7019-3_26
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-99-7018-6
Online ISBN: 978-981-99-7019-3
eBook Packages: Computer ScienceComputer Science (R0)