Skip to main content
SpringerLink
Log in

Multi-variable Volterra kernels identification using time-delay neural networks: application to unsteady aerodynamic loading

  • Original Paper
  • Published:
Nonlinear Dynamics Aims and scope Submit manuscript

Abstract

In the last decades, the Volterra series theory has been used to construct reduced-order models of nonlinear systems in engineering and applied sciences. For the particular case of weakly nonlinear aerodynamic and aeroelastic systems, the Volterra series theory has been tested as an alternative to the high computing costs of CFD methods. The Volterra series model determination depends on identifying the kernels associated with the respective convolution integrals. The Volterra kernels identification has been tried in many ways, but the majority of them addresses only the direct kernels of single-input, single-output nonlinear systems. However, multiple-input, multiple-output relations are the most typical case for many dynamic systems. In this case, the so-called Volterra cross-kernels represent the internal couplings between multiple inputs. Not many generalizations of the single-input kernel identification methods to multi-input Volterra kernels are available in the literature. This work proposes a methodology for the identification of Volterra direct kernels and cross-kernels, which is based on time-delay neural networks and the relationship between the kernels functions and the internal parameters of the network. Expressions to derive the pth-order Volterra direct kernels and cross-kernels from the internal parameters of a trained time-delay neural network are derived. The method is checked with a two-degree-of-freedom, two-input, one-output nonlinear system to demonstrate its capabilities. The application to a mildly nonlinear unsteady aerodynamic loading due to pitching and heaving motions of an airfoil is also evaluated. The Volterra direct kernels and cross-kernels of up to third order are successfully identified using training datasets computed with CFD simulations of the Euler equations. Comparisons between CFD simulations and Volterra model predictions are presented, thereby ensuring the potential of the method to systematically extract kernels from neural networks.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12

Similar content being viewed by others

Notes

  1. \(k=\left( \frac{\omega b}{U_{\infty }}\right) \), where \(\omega \) is the angular frequency, and \(b=c/2\) is the airfoil semi-chord.

References

  1. Volterra, V.: Theory of Functionals and of Integral and Integro-differential Equations. Dover Books on Mathematics Series. Dover Publications, New York (2005)

    Google Scholar 

  2. Yamanaka, O., Ohmori, H., Sano, A.: Design method of exact model matching control for finite Volterra series systems. Int. J. Control 68(1), 107–124 (1997). https://doi.org/10.1080/002071797223758

    Article  MathSciNet  MATH  Google Scholar 

  3. Silva, W.A.: Identification of nonlinear aeroelastic systems based on the Volterra theory: progress and opportunities. Nonlinear Dyn. 39(1), 25–62 (2005). https://doi.org/10.1007/s11071-005-1907-z

    Article  MathSciNet  MATH  Google Scholar 

  4. Peng, Z.K., Lang, Z.Q., Billings, S.A.: Analysis of locally nonlinear MDOF systems using nonlinear output frequency response functions. J. Vib. Acoust. 131(5), 051007–051007-10 (2009). https://doi.org/10.1115/1.3147139

    Article  Google Scholar 

  5. Wu, T., Kareem, A.: Vortex-induced vibration of bridge decks: Volterra series-based model. J. Eng. Mech. 139(12), 1831–1843 (2013). https://doi.org/10.1061/(ASCE)EM.1943-7889.0000628

    Article  Google Scholar 

  6. Billings, S.A.: Identification of nonlinear systems—a survey. IEE Proc. D Control Theory Appl. 127(6), 272–285 (1980). https://doi.org/10.1049/ip-d:19800047

    Article  MathSciNet  Google Scholar 

  7. Cheng, C.M., Peng, Z.K., Zhang, W.M., Meng, G.: Volterra-series-based nonlinear system modeling and its engineering applications: a state-of-the-art review. Mech. Syst. Signal Process. 87, 340–364 (2017). https://doi.org/10.1016/j.ymssp.2016.10.029

    Article  Google Scholar 

  8. Schetzen, M.: The Volterra and Wiener Theories of Nonlinear Systems. Wiley, New York (1980)

    MATH  Google Scholar 

  9. Franz, M.O., Schölkopf, B.: A unifying view of Wiener and Volterra theory and polynomial kernel regression. Neural Comput. 18(12), 3097–3118 (2006). https://doi.org/10.1162/neco.2006.18.12.3097

    Article  MathSciNet  MATH  Google Scholar 

  10. Vanbeylen, L., Pintelon, R., Schoukens, J.: Blind maximum-likelihood identification of Wiener systems. IEEE Trans. Signal Process. 57(8), 3017–3029 (2009). https://doi.org/10.1109/TSP.2009.2017001

    Article  MathSciNet  MATH  Google Scholar 

  11. Billings, S.A., Fakhouki, S.Y.: Nonlinear system identification using the Hammerstein model. Int. J. Syst. Sci. 10(5), 567–578 (1979). https://doi.org/10.1080/00207727908941603

    Article  Google Scholar 

  12. Hunter, I.W., Korenberg, M.J.: The identification of nonlinear biological systems: Wiener and Hammerstein cascade models. Biol. Cybern. 55, 135–144 (1986). https://doi.org/10.1007/BF00341929

    Article  MathSciNet  MATH  Google Scholar 

  13. Korenberg, M.J., Hunter, I.W.: The identification of nonlinear biological systems: LNL cascade models. Biol. Cybern. 55, 125–134 (1986). https://doi.org/10.1007/BF00341928

    Article  MathSciNet  MATH  Google Scholar 

  14. Sjöberg, J., Schoukens, J.: Initializing Wiener–Hammerstein models based on partitioning of the best linear approximation. Automatica 48(2), 353–359 (2012). https://doi.org/10.1016/j.automatica.2011.07.007

    Article  MathSciNet  MATH  Google Scholar 

  15. Guo, Y., Guo, L.Z., Billings, S.A., Coca, D., Lang, Z.Q.: Volterra series approximation of a class of nonlinear dynamical systems using the Adomian decomposition method. Nonlinear Dyn. 74(1), 359–371 (2013). https://doi.org/10.1007/s11071-013-0975-8

    Article  MathSciNet  MATH  Google Scholar 

  16. Marmarelis, P.Z., Naka, K.: Identification of multi-input biological systems. IEEE Trans. Biomed. Eng. BME–21(2), 88–101 (1974). https://doi.org/10.1109/TBME.1974.324293

    Article  Google Scholar 

  17. Marmarelis, P.Z., Marmarelis, V.: Analysis of Physiological Systems: The White-Noise Approach. Plenum Press, New York (1978)

    Book  MATH  Google Scholar 

  18. Worden, K., Manson, G., Tomlinson, G.R.: A harmonic probing algorithm for the multi-input Volterra series. J. Sound Vib. 201(1), 67–84 (1997). https://doi.org/10.1006/jsvi.1996.0746

    Article  MathSciNet  MATH  Google Scholar 

  19. Chatterjee, A., Vyas, N.A.: Non-linear parameter estimation in multi-degree-of-freedom systems using multi-input Volterra series. Mech. Syst. Signal Process. 18(3), 457–489 (2004). https://doi.org/10.1016/S0888-3270(03)00016-5

    Article  Google Scholar 

  20. Wray, J., Green, G.G.R.: Calculation of the Volterra kernels of non-linear dynamic systems using an artificial neural network. Biol. Cybern. 71(3), 187–195 (1994). https://doi.org/10.1007/BF00202758

    Article  MATH  Google Scholar 

  21. Stegmayer, G., Chiotti, O.: Identification of frequency-domain Volterra model using neural networks. In: Duch, W., Kacprzyk, J., Oja, E., Zadrożny, S. (eds.) Artificial Neural Networks: Formal Models and Their Applications—ICANN 2005, pp. 465–471. Springer, Berlin (2005). https://doi.org/10.1007/11550907_73

    Chapter  Google Scholar 

  22. Stegmayer, G., Chiotti, O.: Volterra NN-based behavioral model for new wireless communications devices. Neural Comput. Appl. 18(3), 283–291 (2009). https://doi.org/10.1007/s00521-008-0180-8

    Article  Google Scholar 

  23. Mišić, J., Marković, V., Marinković, Z.: Volterra kernels extraction from neural networks for amplifier behavioral modeling. In: 2014 X International Symposium on Telecommunications (BIHTEL), pp. 1–6 (2014). https://doi.org/10.1109/BIHTEL.2014.6987646

  24. Silva, W.A.: Application of nonlinear systems theory to transonic unsteady aerodynamic responses. J. Aircr. 30(5), 660–668 (1993). https://doi.org/10.2514/3.46395

    Article  Google Scholar 

  25. Silva, W.A.: Reduced-order models based on linear and nonlinear aerodynamic impulse responses. In: 40th Structures, Structural Dynamics, and Materials Conference and Exhibit, St. Louis, MO, USA (1999). https://doi.org/10.2514/6.1999-1262. AIAA-99-1262

  26. Silva, W.A., Beran, P.S., Cesnik, C.E.S., Guendel, R.E., Kurdila, A., Prazenica, R.J., Librescu, L., Marzocca, P., Raveh, D.E.: Reduced-order modeling: cooperative research and development at the NASA Langley Research Center. In: CEAS/AIAA/AIAE International Forum on Aeroelasticity and Structural Dynamics – IFASD, pp. 1–16, Madrid, Spain (2001). IFASD-2001-008

  27. Marzocca, P., Silva, W.A., Librescu, L.: Nonlinear open-/closed-loop aeroelastic analysis of airfoils via Volterra series. AIAA J. 42(4), 673–686 (2004). https://doi.org/10.2514/1.9552

    Article  Google Scholar 

  28. Lucia, D.J., Beran, P.S., Silva, W.A.: Reduced-order modeling: new approaches for computational physics. Prog. Aerosp. Sci. 40(1), 51–117 (2004). https://doi.org/10.1016/j.paerosci.2003.12.001

    Article  Google Scholar 

  29. Balajewicz, M., Nitzsche, F., Feszty, D.: Application of multi-input Volterra theory to nonlinear multi-degree-of-freedom aerodynamic systems. AIAA J. 48(1), 56–62 (2010). https://doi.org/10.2514/1.38964

    Article  Google Scholar 

  30. Balajewicz, M., Dowell, E.: Reduced-order modeling of flutter and limit-cycle oscillations using the sparse Volterra series. J. Aircr. 49(6), 1803–1812 (2012). https://doi.org/10.2514/1.C031637

    Article  Google Scholar 

  31. de Paula, N.C.G., Marques, F.D., Silva, W.A.: Volterra kernels assessment via time-delay neural networks for nonlinear unsteady aerodynamic loading identification. AIAA J. (2019). https://doi.org/10.2514/1.J057229

    Article  Google Scholar 

  32. Bogacki, P., Shampine, L.F.: A 3 (2) pair of Runge–Kutta formulas. Appl. Math. Lett. 2(4), 321–325 (1989). https://doi.org/10.1016/0893-9659(89)90079-7

    Article  MathSciNet  MATH  Google Scholar 

  33. Hagan, M.T., Demuth, H.B., Beale, M.H., de Jess, O.: Neural Network Design, 2nd edn. Martin Hagan (2014). ISBN 0971732116, 9780971732117

  34. Camilo, E., Marques, F.D., Azevedo, J.L.F.: Hopf bifurcation analysis of typical sections with structural nonlinearities in transonic flow. Aerosp. Sci. Technol. 30(1), 163–174 (2013). https://doi.org/10.1016/j.ast.2013.07.013

    Article  Google Scholar 

  35. Raveh, D.E.: Reduced-order models for nonlinear unsteady aerodynamics. AIAA J. 39(8), 1417–1429 (2001). https://doi.org/10.2514/2.1473

    Article  Google Scholar 

  36. Landon, R.H., Davis, S.S.: Compendium of unsteady aerodynamic measurements. Technical Report R-702, AGARD—Advisory Group for Aerospace Research and Development (NATO) (1982)

Download references

Acknowledgements

The authors acknowledge the financial support of: Grant 2017/02926-9, São Paulo Research Foundation (FAPESP) and Grants 307658/2016-3 and 131493/2016-7, National Council for Scientific and Technological Development (CNPq).

Author information

Authors and Affiliations

  1. Department of Mechanical Engineering, Universidade de São Paulo, Escola de Engenharia de São Carlos, Av. Trabalhador Sancarlense, 400, São Carlos, Brazil

    N. C. G. de Paula & F. D. Marques

Authors
  1. N. C. G. de Paula
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. F. D. Marques
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to F. D. Marques.

Ethics declarations

Conflict of interest

The authors declare that there is no conflict of interest regarding the publication of this paper.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

de Paula, N.C.G., Marques, F.D. Multi-variable Volterra kernels identification using time-delay neural networks: application to unsteady aerodynamic loading. Nonlinear Dyn 97, 767–780 (2019). https://doi.org/10.1007/s11071-019-05011-8

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11071-019-05011-8

Keywords

Search

Navigation