The role of parameter constraints in EE and OE methods for optimal identification of continuous LTI models

Witold Byrski; Jedrzej Byrski

International Journal of Applied Mathematics and Computer Science (2012)

  • Volume: 22, Issue: 2, page 379-388
  • ISSN: 1641-876X

Abstract

top
The paper presents two methods used for the identification of Continuous-time Linear Time Invariant (CLTI) systems. In both methods the idea of using modulating functions and a convolution filter is exploited. It enables the proper transformation of a differential equation to an algebraic equation with the same parameters. Possible different normalizations of the model are strictly connected with different parameter constraints which have to be assumed for the nontrivial solution of the optimal identification problem. Different parameter constraints result in different quality of identification. A thorough discussion on the role of parameter constraints in the optimality of system identification is included. For time continuous systems, the Equation Error Method (EEM) is compared with the continuous version of the Output Error Method (OEM), which appears as a special sub-case of the EEM.

How to cite

top

Witold Byrski, and Jedrzej Byrski. "The role of parameter constraints in EE and OE methods for optimal identification of continuous LTI models." International Journal of Applied Mathematics and Computer Science 22.2 (2012): 379-388. <http://eudml.org/doc/208115>.

@article{WitoldByrski2012,
abstract = {The paper presents two methods used for the identification of Continuous-time Linear Time Invariant (CLTI) systems. In both methods the idea of using modulating functions and a convolution filter is exploited. It enables the proper transformation of a differential equation to an algebraic equation with the same parameters. Possible different normalizations of the model are strictly connected with different parameter constraints which have to be assumed for the nontrivial solution of the optimal identification problem. Different parameter constraints result in different quality of identification. A thorough discussion on the role of parameter constraints in the optimality of system identification is included. For time continuous systems, the Equation Error Method (EEM) is compared with the continuous version of the Output Error Method (OEM), which appears as a special sub-case of the EEM.},
author = {Witold Byrski, Jedrzej Byrski},
journal = {International Journal of Applied Mathematics and Computer Science},
keywords = {continuous systems; parameter constraints in identification; modulating functions; transfer function normalization; least squares method},
language = {eng},
number = {2},
pages = {379-388},
title = {The role of parameter constraints in EE and OE methods for optimal identification of continuous LTI models},
url = {http://eudml.org/doc/208115},
volume = {22},
year = {2012},
}

TY - JOUR
AU - Witold Byrski
AU - Jedrzej Byrski
TI - The role of parameter constraints in EE and OE methods for optimal identification of continuous LTI models
JO - International Journal of Applied Mathematics and Computer Science
PY - 2012
VL - 22
IS - 2
SP - 379
EP - 388
AB - The paper presents two methods used for the identification of Continuous-time Linear Time Invariant (CLTI) systems. In both methods the idea of using modulating functions and a convolution filter is exploited. It enables the proper transformation of a differential equation to an algebraic equation with the same parameters. Possible different normalizations of the model are strictly connected with different parameter constraints which have to be assumed for the nontrivial solution of the optimal identification problem. Different parameter constraints result in different quality of identification. A thorough discussion on the role of parameter constraints in the optimality of system identification is included. For time continuous systems, the Equation Error Method (EEM) is compared with the continuous version of the Output Error Method (OEM), which appears as a special sub-case of the EEM.
LA - eng
KW - continuous systems; parameter constraints in identification; modulating functions; transfer function normalization; least squares method
UR - http://eudml.org/doc/208115
ER -

References

top
  1. Byrski, W. and Fuksa, S. (1995). Optimal identification of continuous systems in 2 space by the use of compact support filter, International Journal of Modelling & Simulation, IASTED 15(4): 125-131. 
  2. Byrski, W. and Fuksa, S. (1996). Linear adaptive controller for continuous system with convolution filter, Proceedings of the 13th IFAC Triennial World Congress, San Francisco, CA, USA, pp. 379-384. 
  3. Byrski, W. and Fuksa, S. (1999). Time variable gram matrix eigen-problem and its application to optimal identification of continuous systems, Proceedings of the European Control Conference ECC99, Karlsruhe, Germany, F0256. 
  4. Byrski, W. and Fuksa, S. (2000). Optimal identification of continuous systems and a new fast algorithm for on line mode, Proceedings of the International Conference on System Identification, SYSID2000, Santa Barbara, CA, USA, PM 2-5. Zbl0547.93013
  5. Byrski, W. and Fuksa, S. (2001). Stability analysis of CLTI state feedback system with simultaneous state and parameter identification, Proceedings of the IASTED International Conference on Applied Simulation and Modelling, ASM01, 2001, Marbella, Spain, pp. 7-12. 
  6. Byrski, W., Fuksa, S. and Byrski, J. (1999). A fast algorithm for the eigenproblem in on-line continuous model identification, Proceedings of the 18 IASTED International Conference on Modelling, Identification and Control, MIC99, Innsbruck, Austria, pp. 22-24. Zbl0547.93013
  7. Byrski, W., Fuksa, S. and Nowak, M. (2003). The quality of identification for different normalizations of continuous transfer functions, Proceedings of the 22 IASTED International Conference on Modelling, Identification and Control, MIC03, 2003, Innsbruck, Austria, pp. 96-101. 
  8. Co, T. and Ydstie, B. (1990). System identification using modulating functions and fast Fourier transforms, Computers & Chemical Engineering 14(10): 1051-1066. 
  9. Eykhoff, P. (1974). System Identification, Parameter and State Estimation, J. Wiley, London. Zbl0667.93102
  10. Garnier, H. and Wang, L. (Eds.) (2008). Identification of Continuous-time Models from Sampled Data, Advances in Industrial Control, Springer-Verlag, London. 
  11. Gillberg, J. and Ljung, L. (2009). Frequency domain identification of continuous time ARMA models from sampled data, Automatica 4(6): 1371-1378. Zbl1166.93027
  12. Johansson, R. (2010). Continuous-time model identification and state estimation using non-uniformly sampled data, Proceedings of the 19th International Symposium on Mathematical Theory of Networks and Systems, MTNS 2010, Budapest, Hungary, pp. 347-354. 
  13. Ljung, L. and Wills, A. (2010). Issues in sampling and estimating continuous-time models with stochastic disturbances, Automatica 46(5): 925-931. Zbl1191.93137
  14. Maletinsky, V. (1979). Identification of continuous dynamical systems with spline-type modulating functions method, Proceedings of the IFAC Symposium on Identification and SPE, Darmstadt, Germany, Vol. 1, p. 275. 
  15. Preisig, H.A. and Rippin, D.W.T. (1993). Theory and application of the modulating function method,Computers & Chemical Engineering 17(1): 1-16. 
  16. Schwartz, L. (1966). Théorie des distributions, Hermann, Paris. Zbl0149.09501
  17. Shinbrot, M. (1957). On the analysis of linear and nonlinear systems, Transactions of the American Society of Mechanical Engineers, Journal of Basic Engineering 79: 547-552. 
  18. Sinha, N.K. and Kuszta, B. (1983). Modeling and Identification of Dynamic Systems, Van Nostrand RC, New York, NY. Zbl0365.93011
  19. Soderstrom, T. and Stoica, P. (1994). System Identification, Prentice Hall, London. Zbl0695.93108
  20. Unbehauen, H. and Rao, G. P. (1987). Identification of Continuous Systems, North-Holland, Amsterdam. Zbl0647.93001
  21. Yeredor, A. (2006). On the role of constraints in system identification, 4th International Workshop on Total Least Squares and Errors-invariables Modelling, Leuven, Belgium, (see also http://diag.mchtr.pw.edu.pl/damadics). 
  22. Young, P. (1981). Parameter estimation for continuous-time models-A survey, Automatica 17(1): 23-39. Zbl0451.93052

NotesEmbed ?

top

You must be logged in to post comments.