Nonlinear Dynamics

, Volume 62, Issue 3, pp 543–552 | Cite as

2–ℒ nonlinear system identification via recurrent neural networks

Original Paper


This paper proposes an ℒ2–ℒ identification scheme as a new robust identification method for nonlinear systems via recurrent neural networks. Based on linear matrix inequality (LMI) formulation, for the first time, the ℒ2–ℒ learning algorithm is presented to reduce the effect of disturbance to an ℒ2–ℒ induced norm constraint. New stability results, such as boundedness, input-to-state stability (ISS), and convergence, are established in some senses. It is shown that the design of the ℒ2–ℒ identification method can be achieved by solving LMIs, which can be easily facilitated by using some standard numerical packages. A numerical example is presented to demonstrate the validity of the proposed identification scheme.


2–ℒ identification Recurrent neural networks Input-to-state stability (ISS) Linear matrix inequality (LMI) Weight learning law 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Hunt, K.J., Sbarbaro, D., Zbikowski, R., Gawthrop, P.J.: Neural networks for control systems—a survey. Automatica 28, 1083–1112 (1992) MATHCrossRefMathSciNetGoogle Scholar
  2. 2.
    Ioannou, P.A., Sun, J.: Robust Adaptive Control. Prentice-Hall, Upper Saddle River (1996) MATHGoogle Scholar
  3. 3.
    Egardt, B.: Stability of Adaptive Controllers. Lecture Notes in Control and Information Sciences, vol. 20. Springer, Berlin (1979) MATHGoogle Scholar
  4. 4.
    Polycarpou, M.M., Ioannou, P.A.: Learning and convergence analysis of neural-type structured networks. IEEE Trans. Neural Netw. 3, 39–50 (1992) CrossRefGoogle Scholar
  5. 5.
    Jin, L., Gupta, M.M.: Stable dynamic backpropagation learning in recurrent neural networks. IEEE Trans. Neural Netw. 10, 1321–1334 (1999) CrossRefGoogle Scholar
  6. 6.
    Suykens, J.A.K., Vandewalle, J., De Moor, B.: nl q theory: checking and imposing stability of recurrent neural networks for nonlinear modelling. IEEE Trans. Signal Process. 45, 2682–2691 (1997) CrossRefGoogle Scholar
  7. 7.
    Kosmatopoulos, E.B., Polycarpou, M.M., Christodoulou, M.A., Ioannou, P.A.: High-order neural network structures for identification of dynamical systems. IEEE Trans. Neural Netw. 6, 431–442 (1995) CrossRefGoogle Scholar
  8. 8.
    Jagannathan, S., Lewis, F.L.: Identification of nonlinear dynamical systems using multilayered neural networks. Automatica 32, 1707–1712 (1996) MATHCrossRefMathSciNetGoogle Scholar
  9. 9.
    Song, Q.: Robust training algorithm of multilayered neural networks for identification of nonlinear dynamic systems. IEE Proc. Control Theory Appl. 145, 41–46 (1998) MATHCrossRefGoogle Scholar
  10. 10.
    Yu, W., Li, X.: Discrete-time neuro identification without robust modification. IEE Proc. Control Theory Appl. 150, 311–316 (2003) CrossRefGoogle Scholar
  11. 11.
    Rubio, J.J., Yu, W.: Stability analysis of nonlinear system identification via delayed neural networks. IEEE Trans. Circuits Syst. II 57, 161–165 (2007) CrossRefGoogle Scholar
  12. 12.
    Grigoriadis, K.M., Watson, J.T.: Reduced-order ℋ and ℒ2–ℒ filtering via linear matrix inequalities. IEEE Trans. Aerosp. Electron. Syst. 33, 1326–1338 (1997) CrossRefGoogle Scholar
  13. 13.
    Watson, J.T., Grigoriadis, K.M.: Optimal unbiased filtering via linear matrix inequalities. Syst. Control Lett. 35, 111–118 (1998) MATHCrossRefMathSciNetGoogle Scholar
  14. 14.
    Palhares, R.M., Peres, P.L.D.: Robust filtering with guaranteed energy-to-peak performance—an LMI approach. Automatica 36, 851–858 (2000) MATHCrossRefMathSciNetGoogle Scholar
  15. 15.
    Gao, H., Wang, C.: Robust ℒ2–ℒ filtering for uncertain systems with multiple time-varying state delays. IEEE Trans. Circuits Syst. I 50, 594–599 (2003) CrossRefMathSciNetGoogle Scholar
  16. 16.
    Gao, H., Wang, C.: Delay-dependent robust ℋ and ℒ2–ℒ filtering for a class of uncertain nonlinear time-delay systems. IEEE Trans. Autom. Control 48, 1661–1666 (2003) CrossRefMathSciNetGoogle Scholar
  17. 17.
    Mahmoud, M.S.: Resilient ℒ2–ℒ filtering of polytopic systems with state delays. IET Control Theory Appl. 1(1), 141–154 (2007) CrossRefMathSciNetGoogle Scholar
  18. 18.
    Qiu, J., Feng, G., Yang, J.: New results on robust energy-to-peak filtering for discrete-time switched polytopic linear systems with time-varying delay. IET Control Theory Appl. 2(9), 795–806 (2008) CrossRefMathSciNetGoogle Scholar
  19. 19.
    Zhou, Y., Li, J.: Energy-to-peak filtering for singular systems: the discrete-time case. IET Control Theory Appl. 2(9), 773–781 (2008) CrossRefMathSciNetGoogle Scholar
  20. 20.
    Boyd, S., Ghaoui, L.E., Feron, E., Balakrishinan, V.: Linear Matrix Inequalities in Systems and Control Theory. SIAM, Philadelphia (1994) Google Scholar
  21. 21.
    Gahinet, P., Nemirovski, A., Laub, A.J., Chilali, M.: LMI Control Toolbox. The Mathworks, Natick (1995) Google Scholar
  22. 22.
    Hopfield, J.J.: Neurons with grade response have collective computational properties like those of two-state neurons. Proc. Natl. Acad. Sci. USA 81, 3088–3092 (1984) CrossRefGoogle Scholar
  23. 23.
    Sontag, E.D., Wang, Y.: On characterizations of the input-to-state stability property. Syst. Control Lett. 24, 351–359 (1995) MATHCrossRefMathSciNetGoogle Scholar
  24. 24.
    Zeng, Y., Singh, S.N.: Adaptive control of chaos in Lorenz system. Dyn. Control 7, 143–154 (1997) MATHCrossRefMathSciNetGoogle Scholar

Copyright information

© Springer Science+Business Media B.V. 2010

Authors and Affiliations

  1. 1.Division of Electronics and Control EngineeringWonkwang UniversityIksanKorea

Personalised recommendations