Nonlinear Dynamics

, Volume 85, Issue 2, pp 1297–1318 | Cite as

An evaluation of data-driven identification strategies for complex nonlinear dynamic systems

  • Patrick T. Brewick
  • Sami F. Masri
Original Paper


The development of suitable mathematical models on the basis of dynamic measurements from dispersed structural systems that may be undergoing significant nonlinear behavior is an important and very challenging problem in the field of Applied Mechanics that has drawn the attention of numerous investigators and motivated the development of many approaches for extracting reduced-order, reduced-complexity models from such systems. However, even though numerous nonlinear system identification techniques that are focused on the class of problems encountered in the structural dynamics field have been developed over the past decades, there are no systematic studies available that rigorously compare the performance and fidelity of such methods under similar operating conditions, and when encountering challenging nonlinear phenomena (such as hysteresis) that are present in physical systems, at different scales. This paper explores a variety of data-driven identification techniques for complex nonlinear systems and provides a much needed critical comparison of the accuracy and performance of each method. The Volterra/Wiener neural network (VWNN), a more recent development in nonlinear identification, is featured and compared against several existing methods, including polynomial-based nonlinear estimators and other artificial neural network systems. A representative three degree-of-freedom structure with nonlinear restoring force elements is used as the primary means of comparison for the different methods, and a variety of nonlinear models were investigated, including bilinear hysteresis, polynomial stiffness, and Bouc–Wen hysteresis. Performance comparisons were based on the ability to estimate the acceleration responses for both training and testing simulations. The results showed that, in general, the VWNN provided better accuracy in its estimates for each model. The VWNN also performed best when evaluated for scenarios in which numerical integration is required to find velocity and displacement information from measured accelerations or sensor noise is present in the measured responses.


Nonlinear identification Data-driven methods Neural networks Bouc–Wen Hysteresis 



The first author would like to acknowledge the support of the Viterbi Postdoctoral Fellowship from the University of Southern California. The assistance of Dr. Armen Derkevorkian in lending his expertise of artificial neural networks and generously granting permission to modify figures from his own works is gratefully acknowledged.


  1. 1.
    Andronikou, A., Bekey, G.A.: Identification of hysteretic systems. In: 18th IEE Conference on Decision and Control (1984)Google Scholar
  2. 2.
    Antoni, J.: Blind separation of vibration components: principles and demonstrations. Mech. Syst. Signal Process. 19, 1166–1180 (2005)CrossRefGoogle Scholar
  3. 3.
    Baber, T., Noori, M.: Random vibration of degrading, pinching systems. ASCE J. Eng. Mech. 111(8), 1010–1026 (1985)CrossRefGoogle Scholar
  4. 4.
    Baber, T., Noori, M.: Modeling general hysteresis behavior and random vibration application. J. Vib. Acoust. Stress Reliab. Des. 108(4), 411–420 (1986)CrossRefGoogle Scholar
  5. 5.
    Baber, T., Wen, Y.: Stochastic equivalent linearization for hysteretic, degrading, multistory structures. Technical Report, Civil Engeering Studies, SRS No. 471, University of Illinois, Urbana, Illinois (1979)Google Scholar
  6. 6.
    Baber, T., Wen, Y.: Random vibration of hysteretic degrading systems. J. Eng. Mech. Div. ASCE 107, 1069–1087 (1981)Google Scholar
  7. 7.
    Benedettini, F., Capecchi, D., Vestroni, F.: Identification of hysteretic oscillators under earthquake loading by nonparametric models. ASCE J. Eng. Mech. 121, 606–612 (1995)CrossRefGoogle Scholar
  8. 8.
    Bouc, R.: Forced vibration of mechanical systems with hysteresis. In: 4th Conference on Nonlinear Oscillation, Prague, Czechoslovakia (1967)Google Scholar
  9. 9.
    Brincker, R., Andersen, P., Zhang, L.: Modal identification from ambient responses using frequency domain decomposition. In: Proceedings of the 18th International Modal Analysis Conference (IMAC), San Antonio, TX, pp. 625–630 (2000)Google Scholar
  10. 10.
    Caughey, T.: Random excitation of a system with bilinear hysteresis. ASME J. Appl. Mech. 27, 649–652 (1960)MathSciNetCrossRefGoogle Scholar
  11. 11.
    Chassiakos, A., Masri, S., Smyth, A., Anderson, J.: Adaptive methods for identification of hysteretic structures. In: American Control Conference, ACC95, Seattle, WA (1995)Google Scholar
  12. 12.
    Chassiakos, A., Masri, S., Smyth, A., Caughey, T.: On-line identification of hysteretic systems. ASME J. Appl. Mech. 65, 194–203 (1998)CrossRefGoogle Scholar
  13. 13.
    Chatzi, E.N., Smyth, A.W.: The unscented Kalman filter and particle filter methods for nonlinear structural system identification with non-collocated heterogeneous sensing. Struct. Control Health Monit. 16, 99–123 (2009)CrossRefGoogle Scholar
  14. 14.
    Chatzi, E.N., Smyth, A.W., Masri, S.F.: Experimental application of on-line parametric identification for nonlinear hysteretic systems with model uncertainty. Struct. Saf. 32(5), 326–337 (2010)CrossRefGoogle Scholar
  15. 15.
    Derkevorkian, A., Hernandez-Garcia, M., Yun, H.B., Masri, S.F., Li, P.: Nonlinear data-driven computational models for response prediction and change detection. Struct. Control Health Monit. 22(2), 273–288 (2015)CrossRefGoogle Scholar
  16. 16.
    Derkevorkian, A., Masri, S.F., Fujino, Y., Siringoringo, D.M.: Development and validation of nonlinear computational models of dispersed structures under strong earthquake excitation. Earthq. Eng. Struct. Dyn. 43, 1089–1105 (2014)CrossRefGoogle Scholar
  17. 17.
    Dominguez, A., Sedaghati, R., Stiharu, I.: A new dynamic hysteresis model for magnetorheological dampers. Smart Mater. Struct. 15, 1179–1189 (2006)CrossRefGoogle Scholar
  18. 18.
    Erlicher, S., Point, N.: Thermodynamic admissibility of Bouc–Wen type hysteresis models. C. R. Mec. 332, 51–57 (2004)CrossRefMATHGoogle Scholar
  19. 19.
    Ewins, D.J.: Modal Testing: Theory, Practice and Application, 2nd edn. Research Studies Press LTD, Baldock (2000)Google Scholar
  20. 20.
    Ghanem, R., Romeo, F.: Wavelet-based approach for model and parameter identification of non-linear systems. Int. J. Non-Linear Mech. 36, 835–859 (2001)CrossRefMATHGoogle Scholar
  21. 21.
    Hoshiya, M., Saito, E.: Structural identification by extended Kalman filter. ASCE J. Eng. Mech. 110(12), 1757–1772 (1984)CrossRefGoogle Scholar
  22. 22.
    Ibrahim, S.: Efficient random decrement computation for identification of ambient responses. In: Proceedings of the 19th IMAC, Orlando, FL, pp. 678–703 (2001)Google Scholar
  23. 23.
    Ismail, M., Ikhouane, F., Rodellar, J.: The hysteresis Bouc–Wen model, a survey. Arch. Comput. Methods Eng. 16, 161–188 (2009)CrossRefMATHGoogle Scholar
  24. 24.
    Iwan, W.: A distributed-element model for hysteresis and its steady-state dynamic response. ASME J. Appl. Mech. 33, 893–900 (1966)CrossRefGoogle Scholar
  25. 25.
    Iwan, W., Cifuentes, A.: A model for system identification of degrading structures. Earthq. Eng. Struct. Dyn. 14, 877–890 (1986)CrossRefGoogle Scholar
  26. 26.
    Iwan, W., Lutes, L.: Response of the bilinear hysteretic system to stationary random excitation. J. Acoust. Soc. Am. 43, 545–552 (1968)CrossRefGoogle Scholar
  27. 27.
    James, G., Carne, T., Lauffer, J.: The natural excitation technique (NExT) for modal parameter extraction from operating structures. Int. J. Anal. Exp. Modal Anal. 10(4), 260–277 (1995)Google Scholar
  28. 28.
    Jayakumar, P., Beck, J.L.: System identification using nonlinear structural models. In: Structural Safety Evaluation Based on System Identification Approaches, pp. 82–102. Springer Fachmedien, Wiesbaden (1988)Google Scholar
  29. 29.
    Jennings, P.: Periodic response of a general yielding structure. J. Eng. Mech. Div. ASCE 90, 131–166 (1964)Google Scholar
  30. 30.
    Kerschen, G., Poncelet, F., Golinval, J.C.: Physical interpretation of independent component analysis in structural dynamics. Mech. Syst. Signal Process. 21, 1561–1575 (2007)CrossRefGoogle Scholar
  31. 31.
    Kerschen, G., Worden, K., Vakakis, A., Golinval, J.C.: Past, present and future of nonlinear system identification in structural dynamics. Mech. Syst. Signal Process. 20, 505–592 (2006)CrossRefGoogle Scholar
  32. 32.
    Kosmatopoulos, E., Polycarpou, M., Christodoulou, M., Ioannou, P.: High-order neural network structures for identification of dynamical systems. IEEE Trans. Neural Netw. 6, 422–431 (1995)CrossRefGoogle Scholar
  33. 33.
    Kosmatopoulos, E., Smyth, A., Masri, S., Chassiakos, A.: Robust adaptive neural estimation of restoring forces in nonlinear structures. ASME J. Appl. Mech. 68, 880–893 (2001)CrossRefMATHGoogle Scholar
  34. 34.
    Li, S., Suzuki, Y., Noori, M.: Identification of hysteretic systems with slip using bootstrap filter. Mech. Syst. Signal Process. 18, 781–795 (2004)CrossRefGoogle Scholar
  35. 35.
    Lin, J.W.: Adaptive identification of structural systems by training artificial neural networks. Adv. Inf. Sci. Serv. Sci. 4, 10–17 (2012)Google Scholar
  36. 36.
    Lin, J.W., Betti, R.: On-line identification and damage detection in non-linear structural systems using a variable forgetting factor approach. Earthq. Eng. Struct. Dyn. 33, 419–444 (2004)CrossRefGoogle Scholar
  37. 37.
    Lin, J.W., Wu, T.H.: Modeling and assessment of VWNN for signal processing of structural systems. Struct. Eng. Mech. 45, 53–67 (2013)MathSciNetCrossRefGoogle Scholar
  38. 38.
    Loh, C., Chung, S.: A three-stage identification approach for hysteretic systems. Earthq. Eng. Struct. Dyn. 22, 129–150 (1993)CrossRefGoogle Scholar
  39. 39.
    Ma, F., Zhang, H., Bockstedte, A., Foliente, G., Paevere, P.: Parameter analysis of the differential model of hysteresis. ASME J. Appl. Mech. 71, 342–349 (2004)CrossRefMATHGoogle Scholar
  40. 40.
    Masri, S.: Forced vibration of the damped bilinear hysteretic oscillator. J. Acoust. Soc. Am. 57, 106–113 (1975)CrossRefMATHGoogle Scholar
  41. 41.
    Masri, S., Caffrey, J., Caughey, T., Smyth, A., Chassiakos, A.: A general data-based approach for developing reduced-order models of nonlinear MDOF systems. Nonlinear Dyn. 39, 95–112 (2005)CrossRefMATHGoogle Scholar
  42. 42.
    Masri, S., Caughey, T.: A nonparametric identification technique for nonlinear dynamic problems. ASME J. Appl. Mech. 46, 433–447 (1979)CrossRefMATHGoogle Scholar
  43. 43.
    Masri, S., Miller, R., Saud, A., Caughey, T.: Identification of nonlinear vibrating structures; part I: formulation. ASME J. Appl. Mech. 109, 918–922 (1987)CrossRefMATHGoogle Scholar
  44. 44.
    Masri, S., Miller, R., Saud, A., Caughey, T.: Identification of nonlinear vibrating structures; part II: applications. ASME J. Appl. Mech. 109, 923–929 (1987)CrossRefMATHGoogle Scholar
  45. 45.
    Masri, S., Miller, R., Traina, M., Caughey, T.: Development of bearing friction models from experimental measurements. J. Sound Vib. 148, 455–475 (1991)CrossRefGoogle Scholar
  46. 46.
    Masri, S., Tasbihgoo, F., Caffrey, J., Smyth, A., Chassiakos, A.: Data-based model-free representation of complex hysteretic MDOF systems. Struct. Control Health Monit. 13, 365–387 (2006)CrossRefMATHGoogle Scholar
  47. 47.
    MATLAB and Neural Network Toolbox: Release 2014b. TheMathWorks, Inc., Natick, MA (2014)Google Scholar
  48. 48.
    Peeters, B., De Roeck, G.: Stochastic system identification for operational modal analysis: a review. J. Dyn. Syst. Meas. Control 123, 659–667 (2001)CrossRefGoogle Scholar
  49. 49.
    Pei, J., Smyth, A., Kosmatopoulos, E.: Analysis and modification of Volterra/Wiener neural networks for the adaptive identification of non-linear hysteretic dynamic systems. J. Sound Vib. 275, 693–718 (2004)CrossRefGoogle Scholar
  50. 50.
    Peng, C., Iwan, W.: An identification methodology for a class of hysteretic structures. Earthq. Eng. Struct. Dyn. 21, 695–712 (1992)CrossRefGoogle Scholar
  51. 51.
    Pi, Y., Mickleborough, N.: Modal identification of vibrating structures using ARMA model. ASCE J. Eng. Mech. 115, 2232–2250 (1989)CrossRefMATHGoogle Scholar
  52. 52.
    Smyth, A., Masri, S., Chassiakos, A., Caughey, T.: On-line parametric identification of MDOF nonlinear hysteretic systems. ASCE J. Eng. Mech. 125, 133–142 (1999)CrossRefGoogle Scholar
  53. 53.
    Smyth, A., Masri, S., Kosmatopoulos, E., Chassiakos, A., Caughey, T.: Development of adaptive modeling techniques for non-linear hysteretic systems. Int. J. Non-linear Mech. 37, 1435–1451 (2002)CrossRefMATHGoogle Scholar
  54. 54.
    Smyth, A.W., Masri, S.F., Caughey, T.K., Hunter, N.F.: Surveillance of mechanical systems on the basis of vibration signature analysis. ASME J. Appl. Mech. 67, 540–551 (2000)CrossRefMATHGoogle Scholar
  55. 55.
    Toussi, S., Yao, J.: Hysteretic identification of existing structures. J. Eng. Mech. Div. ASCE 109, 1189–1202 (1983)CrossRefGoogle Scholar
  56. 56.
    Van Overschee, P., De Moor, B.: Subspace Identification for Linear Systems: Theory-Implementation-Application. Kluwer, Dordrecht (1996)CrossRefMATHGoogle Scholar
  57. 57.
    Vinogradoc, O., Pivovarov, L.: Vibrations of a system with nonlinear hysteresis. J. Sound Vib. 111, 145–152 (1986)CrossRefGoogle Scholar
  58. 58.
    Wen, Y.: Method for random vibration of hysteretic systems. J. Eng. Mech. Div. ASCE 102, 249–263 (1976)Google Scholar
  59. 59.
    Wen, Y.: Equivalent linearization for hysteretic systems under random excitation. ASME J. Appl. Mech. 47(1), 150–154 (1980)CrossRefMATHGoogle Scholar
  60. 60.
    Wu, M., Smyth, A.: Real-time parameter estimation for degrading and pinching hysteretic models. Int. J. Non-linear Mech. 43, 822–833 (2008)CrossRefGoogle Scholar
  61. 61.
    Wu, M., Smyth, A.W.: Application of the unscented Kalman filter for real-time nonlinear structural system identification. Struct. Control Health Monit. 14, 971–990 (2007)CrossRefGoogle Scholar
  62. 62.
    Wu, T., Kareem, A.: Modeling hysteretic nonlinear behavior of bridge aerodynamics via cellular automata nested neural network. J. Wind Eng. Ind. Aerodyn. 99, 378–388 (2011)Google Scholar
  63. 63.
    Yang, J., Lin, S., Huang, H., Zhou, L.: An adaptive extended Kalman filter for structural damage identification. Struct. Control Health Monit. 13(4), 849–867 (2006)CrossRefGoogle Scholar
  64. 64.
    Yar, M., Hammond, J.: Modeling and response of bilinear hysteretic systems. J. Eng. Mech. Div. ASCE 113, 1000–1013 (1987)CrossRefGoogle Scholar
  65. 65.
    Yar, M., Hammond, J.: Parameter estimation for hysteretic systems. J. Sound Vib. 117, 161–172 (1987)CrossRefGoogle Scholar
  66. 66.
    Zhang, H., Foliente, G., Yang, Y., Ma, F.: Parameter identification of inelastic structures under dynamic loads. Earthq. Eng. Struct. Dyn. 31(5), 1113–1130 (2002)CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2016

Authors and Affiliations

  1. 1.Viterbi School of EngineeringUniversity of Southern CaliforniaLos AngelesUSA

Personalised recommendations