Advertisement

Nonlinear Dynamics

, Volume 96, Issue 3, pp 2157–2177 | Cite as

Deep neural network for unsteady aerodynamic and aeroelastic modeling across multiple Mach numbers

  • Kai Li
  • Jiaqing KouEmail author
  • Weiwei Zhang
Original Paper
  • 122 Downloads

Abstract

Aerodynamic reduced-order model (ROM) is a useful tool to predict nonlinear unsteady aerodynamics with reasonable accuracy and very low computational cost. The efficacy of this method has been validated by many recent studies. However, the generalization capability of aerodynamic ROMs with respect to different flow conditions and different aeroelastic parameters should be further improved. In order to enhance the predicting capability of ROM for varying operating conditions, this paper presents an unsteady aerodynamic model based on long short-term memory (LSTM) network from deep learning theory for large training dataset and sampling space. This type of network has attractive potential in modeling temporal sequence data, which is well suited for capturing the time-delayed effects of unsteady aerodynamics. Different from traditional reduced-order models, the current model based on LSTM network does not require the selection of delay orders. The performance of the proposed model is evaluated by a NACA 64A010 airfoil pitching and plunging in the transonic flow across multiple Mach numbers. It is demonstrated that the model can accurately capture the dynamic characteristics of aerodynamic and aeroelastic systems for varying flow and structural parameters.

Keywords

Unsteady aerodynamic Transonic flow Long short-term memory network Limit-cycle oscillation Reduced-order model 

Notes

Acknowledgements

This work was supported by the National Natural Science Foundation of China (No. 11572252), the National Science Fund for Excellent Young Scholars (No. 11622220) and “111” project of China (No. B17037) and ATCFD project (2015-F-016). We thank Professor Bernd Noack of LIMSI-CNRS for his valuable discussions and suggestions on this work.

Compliance with ethical standards

Conflict of interest

The authors declare that there is no conflict of interest in preparing this article.

References

  1. 1.
    Chen, G., Sun, J., Li, Y.: Active flutter suppression control law design method based on balanced proper orthogonal decomposition reduced order model. Nonlinear Dyn. 70, 1–12 (2012)MathSciNetGoogle Scholar
  2. 2.
    Da Ronch, A., Badcock, K.J., Wang, Y., Wynn, A., Palacios, R.: Nonlinear model reduction for flexible aircraft control design. In: AIAA Atmospheric Flight Mechanic Conference, AIAA- 2012-4404. American Institute of Aeronautics and Astronautics, Minneapolis, Minnesota (2012)Google Scholar
  3. 3.
    Brunton, S., Noack, B.: Closed-loop turbulence control: progress and challenges. Appl. Mech. Rev. 67(5), 050801 (2015)Google Scholar
  4. 4.
    Wang, Z., Zhang, W., Wu, X., Chen, K.: A novel unsteady aerodynamic reduced-order modeling method for transonic aeroelastic optimization. J. Fluid. Struct. 82, 308–328 (2018)Google Scholar
  5. 5.
    Zhang, W., Li, X., Ye, Z., Jiang, Y.: Mechanism of frequency lock-in in vortex-induced vibrations at low Reynolds numbers. J. Fluid Mech. 783, 72–102 (2015)MathSciNetzbMATHGoogle Scholar
  6. 6.
    Gao, C., Zhang, W., Li, X., Liu, Y., Quan, J., Ye, Z., Jiang, Y.: Mechanism of frequency lock-in in transonic buffeting flow. J. Fluid Mech. 818, 528–561 (2017)MathSciNetzbMATHGoogle Scholar
  7. 7.
    Hesse, H., Palacios, R.: Reduced-order aeroelastic models of maneuvering flexible aircraft. AIAA J. 52(8), 1717–1732 (2014)Google Scholar
  8. 8.
    Lucia, D.J., Beran, P.S., Silva, W.A.: Reduced-order modeling: new approaches for computational physics. Prog. Aerosp. Sci. 40(1–2), 51–117 (2004)Google Scholar
  9. 9.
    Ghoreyshi, M., Jirasek, A., Cummings, R.M.: Reduced order unsteady aerodynamic modeling for stability and control analysis using computational fluid dynamics. Prog. Aerosp. Sci. 71, 167–217 (2014)Google Scholar
  10. 10.
    Kou, J., Zhang, W.: A hybrid reduced-order framework for complex aeroelastic simulations. Aerosp. Sci. Technol. 84, 880–894 (2019)Google Scholar
  11. 11.
    Willcox, K., Peraire, J.: Balanced model reduction via the proper orthogonal decomposition. AIAA J. 40(11), 2323–2330 (2002)Google Scholar
  12. 12.
    Schmid, P.J.: Dynamic mode decomposition of numerical and experimental data. J. Fluid Mech. 656, 5–28 (2010)MathSciNetzbMATHGoogle Scholar
  13. 13.
    Brewick, P.T., Masri, S.F.: An evaluation of data-driven identification strategies for complex nonlinear dynamic systems. Nonlinear Dyn. 85, 1297–1318 (2016)MathSciNetGoogle Scholar
  14. 14.
    He, S., Yang, Z., Gu, Y.: Nonlinear dynamics of an aeroelastic airfoil with free-play in transonic flow. Nonlinear Dyn. 87, 2099–2125 (2017)Google Scholar
  15. 15.
    Zhang, W., Ye, Z.: Reduced-order-model-based flutter analysis at high angle of attack. J. Aircr. 44(6), 2086–2089 (2007)Google Scholar
  16. 16.
    Zhang, W., Wang, B., Ye, Z., Quan, J.: Efficient method for limit cycle flutter analysis based on nonlinear aerodynamic reduced-order models. AIAA J. 50(5), 1019–1028 (2012)Google Scholar
  17. 17.
    Da Ronch, A., Ghoreyshi, M., Badcock, K.J.: On the generation of flight dynamics aerodynamic tables by computational fluid dynamics. Prog. Aerosp. Sci. 47(8), 597–620 (2011)Google Scholar
  18. 18.
    Kou, J., Zhang, W., Yin, M.: Novel Wiener models with a time-delayed nonlinear block and their identification. Nonlinear Dyn. 85, 2389–2404 (2016)Google Scholar
  19. 19.
    Duriez, Thomas, Brunton, Steven L., Noack, Bernd R.: Machine Learning Control-Taming Nonlinear Dynamics and Turbulence. Springer, Switzerland (2017)zbMATHGoogle Scholar
  20. 20.
    Tracey, B.D., Duraisamy, K., Alonso, J.J.: A machine learning strategy to assist turbulence model development. In: 53rd AIAA Aerospace Sciences Meeting, AIAA SciTech, Kissimmee, FL (2015)Google Scholar
  21. 21.
    Mannarino, A., Mantegazza, P.: Nonlinear aeroelastic reduced order modeling by recurrent neural networks. J. Fluid. Struct. 48, 103–121 (2014)Google Scholar
  22. 22.
    Amsallem, D., Farhat, C.: Interpolation method for adapting reduced-order models and application to aeroelasticity. AIAA J. 46(7), 1803–1813 (2008)Google Scholar
  23. 23.
    Skujins, T., Cesnik, C.E.S.: Reduced-order modeling of unsteady aerodynamics across multiple mach regimes. J. Aircr. 51(6), 1681–1704 (2014)Google Scholar
  24. 24.
    Winter, M., Breitsamter, C.: Neurofuzzy-model-based unsteady aerodynamic computations across varying freestream conditions. AIAA J. 54(9), 2705–2720 (2016)Google Scholar
  25. 25.
    Liu, H., Huang, R., Zhao, Y., Hu, H.: Reduced-order modeling of unsteady aerodynamics for an elastic wing with control surfaces. J. Aerosp. Eng. 30(3), 04016083 (2016)Google Scholar
  26. 26.
    Kou, J., Zhang, W.: Reduced-order modeling for nonlinear aeroelasticity with varying Mach numbers. J. Aerosp. Eng. 31(6), 04018105 (2018)Google Scholar
  27. 27.
    Liu, H., Hu, H., Zhao, Y., et al.: Efficient reduced-order modeling of unsteady aerodynamics robust to flight parameter variations. J. Fluids Struct. 49(8), 728–741 (2014)Google Scholar
  28. 28.
    Zhang, W., Kou, J., Wang, Z.: Nonlinear aerodynamic reduced-order model for limit-cycle oscillation and flutter. AIAA J. 54(10), 3302–3310 (2016)Google Scholar
  29. 29.
    Schmidhuber, J.: Deep learning in neural networks: an overview. Neural Netw. 61, 85–117 (2015)Google Scholar
  30. 30.
    Sainath, T.N., Mohamed, A.R., Kingsbury, B., Ramabhadran, B.: Deep convolutional neural networks for LVCSR. In: 2013 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 8614–8618 (2013)Google Scholar
  31. 31.
    Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. In: 2012 International Conference on Neural Information Processing Systems, pp. 1097–1105 (2012)Google Scholar
  32. 32.
    Xiong, H., Alipanahi, B., et al.: The human splicing code reveals new insights into the genetic determinants of disease. Science 347(6218), 1254806 (2015)Google Scholar
  33. 33.
    Jin, X., Shao, J., Zhang, X., An, W., Malekian, R.: Modeling of nonlinear system based on deep learning framework. Nonlinear Dyn. 84, 1327–1340 (2016)MathSciNetGoogle Scholar
  34. 34.
    Rosa, E.D.L., Yu, W.: Randomized algorithms for nonlinear system identification with deep learning modification. Inf. Sci. 364–365, 197–212 (2016)Google Scholar
  35. 35.
    Ling, J., Kurzawski, A., Templeton, J.: Reynolds averaged turbulence modeling using deep learning neural networks with embedded invariance. J. Fluid Mech. 807, 155–166 (2016)MathSciNetzbMATHGoogle Scholar
  36. 36.
    Efrati, A.: Apples machines can learn too. The Information. https://www.theinformation.com/articles/apples-machines-can-learn-too (2016). Accessed 13 June 2016
  37. 37.
    Vogels, W.: Bringing the magic of amazon ai and alexa to apps on aws. All Things Distributed. https://www.allthingsdistributed.com (2016). Accessed 30 Nov 2016
  38. 38.
    Zhao, Z., Chen, W., Wu, X., et al.: LSTM network: a deep learning approach for short-term traffic forecast. IET Intell. Transp. Syst. 11(2), 68–75 (2017)Google Scholar
  39. 39.
    Fu, R., Zhang, Z., Li L.: Using LSTM and GRU neural network methods for traffic flow prediction. In: 2016 31st Youth Academic Annual Conference of Chinese Association Automation, pp. 324–328 (2016)Google Scholar
  40. 40.
    Wang, Z., Xiao, D., Fang, F., Govindan, R., Pain, C.C., Guo, Y.: Model identification of reduced order fluid dynamics systems using deep learning. Int. J. Numer. Meth. Fluids. 86(4), 255–268 (2018)MathSciNetGoogle Scholar
  41. 41.
    Mohan, A.T., Gaitonde, D.V.: A Deep Learning based Approach to Reduced Order Modeling for Turbulent Flow Control using LSTM Neural Networks (2018). arXiv:1804.09269v1 [physics.comp-ph]
  42. 42.
    LeCun, Y., Bengio, Y., Hinton, G.E.: Deep learning. Nature 521(7553), 436–444 (2015)Google Scholar
  43. 43.
    Hochreiter, S., Schmidhuber, J.: Long short-term memory. Neural Comput. 9(8), 1735–1780 (1997)Google Scholar
  44. 44.
    Pascanu. R., Mikolov, T., Bengio, Y.: On the difficulty of training recurrent neural networks. In: International Conference on International Conference on Machine Learning. JMLR.org, III-1310 (2013)Google Scholar
  45. 45.
    Chollet, F.: Keras (2015). https://github.com/fchollet/keras
  46. 46.
    Kou, J., Zhang, W.: Multi-kernel neural networks for nonlinear unsteady aerodynamic reduced-order modeling. Aerosp. Sci. Technol. 67, 309–326 (2017)Google Scholar
  47. 47.
    Kou, J., Zhang, W.: An approach to enhance the generalization capability of nonlinear aerodynamic reduced-order models. Aerosp. Sci. Technol. 49(1), 197–208 (2016)Google Scholar
  48. 48.
    Goodfellow, Ian, Bengio, Yoshua, Courville, Aaron: Deep Learning, pp. 246–252. MIT Press, Cambridge (2016)zbMATHGoogle Scholar
  49. 49.
    Jiang, Y.: Numerical Solution of Navier–Stokes Equations on Generalized Mesh and Its Applications. Ph.D. Dissertation, Northwestern Polytechnical Univ., Xi’an, China (2013)Google Scholar
  50. 50.
    Boer, A., Schoot, M.S., Faculty, H.B.: Mesh deformation based on radial basis function interpolation. Comput. Struct. 85(11–14), 784–795 (2007)Google Scholar
  51. 51.
    Zhang, W., Gao, C., Liu, Y., Ye, Z., Jiang, Y.: The interaction between flutter and buffet in transonic flows. Nonlinear Dyn. 82(4), 1851–1865 (2015)Google Scholar
  52. 52.
    Gao, C., Zhang, W., Liu, Y., Ye, Z., Jiang, Y.: Numerical study on the correlation of transonic single-degree-of-freedom flutter and buffet. Sci. China Phys. Mech. Astron. 58(8), 084701 (2015)Google Scholar
  53. 53.
    Zhang, W., Jiang, Y., Ye, Z.: Two better loosely coupled solution algorithms of CFD based aeroelastic simulation. Eng. Appl. Comput. Fluid Mech. 1(4), 253–262 (2007)Google Scholar
  54. 54.
    Graves, A., Mohamed, A.R., Hinton G.E.: Speech recognition with deep recurrent neural networks. In: 2013 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 6645–6649 (2013)Google Scholar
  55. 55.
    Pascanu, R., Gulcehre, C., Cho, K., Bengio, Y.: How to construct deep recurrrent neural networks (2013). arXiv: 1312.6026 [cs.NE]
  56. 56.
    Kingma, D., Adam B.J.: A method for stochastic optimization (2014). arXiv: 1412.6980 [cs.LG]
  57. 57.
    Thomas, J.P., Dowell, E.H., Hall, K.C.: Nonlinear inviscid aerodynamic effects on transonic divergence, flutter and limit cycle oscillations. AIAA J. 40(4), 638–646 (2002)Google Scholar

Copyright information

© Springer Nature B.V. 2019

Authors and Affiliations

  1. 1.School of AeronauticsNorthwestern Polytechnical UniversityXi’anChina

Personalised recommendations