Generalized Kernel Regression Estimate for the Identification of Hammerstein Systems

Grzegorz Mzyk

doi:10.2478/v10006-007-0018-z

Abstract

A modified version of the classical kernel nonparametric identification algorithm for nonlinearity recovering in a Hammerstein system under the existence of random noise is proposed. The assumptions imposed on the unknown characteristic are weak. The generalized kernel method proposed in the paper provides more accurate results in comparison with the classical kernel nonparametric estimate, regardless of the number of measurements. The convergence in probability of the proposed estimate to the unknown characteristic is proved and the question of the convergence rate is discussed. Illustrative simulation examples are included.

References

Bai E.W. (2003): Frequency domain identification of Hammerstein models. — IEEE Trans. Automat. Contr., Vol. 48, No. 4, pp. 530-542.10.1109/TAC.2003.809803
Search in Google Scholar
Bai E.W. and Li D. (2004): Convergence of the iterative Hammerstein system identification algorithm. — IEEE Trans. Automat. Contr., Vol. 49, No. 11, pp. 1929-1940.10.1109/TAC.2004.837592
Search in Google Scholar
Billings S.A. and Fakhouri S.Y. (1982): Identification of systems containing linear dynamic and static nonlinear elements. — Automat., Vol. 18, No. 1, pp. 15-26.10.1016/0005-1098(82)90022-X
Search in Google Scholar
Chang F.H.I. and Luus R. (1971): A non-iterative method for identification using Hammerstein model. — IEEE Trans. Automat. Contr., Vol. AC-16, No. 4, pp. 464-468.10.1109/TAC.1971.1099787
Search in Google Scholar
Chen H.F. (2005): Strong consistency of recursive identification for Hammerstein systems with discontinuous piecewise-linear memoryless block. — IEEE Trans. Automat. Contr., Vol. 50, No. 10, pp. 1612-1617.10.1109/TAC.2005.856658
Search in Google Scholar
Giannakis G.B. and Serpedin E. (2001): A bibliography on non-linear system identification. — Signal Process., Vol. 81, No. 3, pp. 533-580.
Search in Google Scholar
Giunta G., Jacovitti G. and Neri A. (1991): Bandpass nonlinear system identification by higher cross correlation. — IEEE Trans. Signal Process., Vol. 39, No. 9, pp. 2092-2095.10.1109/78.134443
Search in Google Scholar
Gomez J.C. and Basualdo M. (2000): Nonlinear model identification of batch distillation process. — Proc. Int. IFAC Symp. Advanced Control of Chemical Processes, ADCHEM, Pisa, Italy, pp. 953-959.
Search in Google Scholar
Greblicki W. (1989): Nonparametric orthogonal series identification of Hammerstein systems. — Int. J. Syst. Sci., Vol. 20, No. 12, pp. 2355-2367.10.1080/00207728908910318
Search in Google Scholar
Greblicki W. (2001): Recursive identification of Wiener systems. — Int. J. Appl. Math. Comp. Sci., Vol. 11, No. 4, pp. 977-991.
Search in Google Scholar
Greblicki W., Krzyżak A. and Pawlak M. (1984): Distribution free pointwise consistency of kernel regression estimate. — Ann. Stat., Vol. 12, No. 4, pp. 1570-1575.10.1214/aos/1176346815
Search in Google Scholar
Greblicki W. and Pawlak M. (1986): Identification of discrete Hammerstein systems using kernel regression estimates. — IEEE Trans. Automat. Contr., Vol. 31, No. 1, pp. 74-77.10.1109/TAC.1986.1104096
Search in Google Scholar
Greblicki W. and Pawlak M. (1989): Nonparametric identification of Hammerstein systems. — IEEE Trans. Inf. Theory, Vol. 35, No. 2, pp. 409-418.10.1109/18.32135
Search in Google Scholar
Greblicki W. and Pawlak M. (1994): Cascade non-linear system identification by a non-parametric method. — Int. J. Syst. Sci., Vol. 25, No. 1, pp. 129-153.10.1080/00207729408928949
Search in Google Scholar
Haber M. and Keviczky L. (1999): Nonlinear System Identification - Input-Output Modeling Approach. — Dordrecht: Kluwer.10.1007/978-94-011-4481-0
Search in Google Scholar
Haber R. and Zeirfuss P. (1988): Identification of an electrically heated heat exchanger by several nonlinear models using different structures and parameter estimation methods. — Tech. Rep., Inst. Machine and Process Automation, Technical University of Vienna, Austria.
Search in Google Scholar
Hannan E.J. and Deistler M. (1998): The Statistical Theory of Linear Systems. — New York: Wiley.
Search in Google Scholar
Hasiewicz Z. and Mzyk G. (2004a): Combined parametric-nonparametric identification of Hammerstein systems. — IEEE Trans. Automat. Contr., Vol. 49, No. 8, pp. 1370-1376.10.1109/TAC.2004.832662
Search in Google Scholar
Hasiewicz Z. and Mzyk G. (2004b): Nonparametric instrumental variables for Hammerstein system identification. — Int. J. Contr., (submitted).
Search in Google Scholar
Hasiewicz Z., Pawlak M. and Śliwiński P. (2005): Nonparametric identification of nonlinearities in block-oriented systems by orthogonal wavelets with compact support. — IEEE Trans. Circ. Syst. I: Fund. Theory Applic., Vol. 52, No. 2, pp. 427-442.10.1109/TCSI.2004.840288
Search in Google Scholar
Härdle W. (1990): Applied Nonparametric Regression. — Cambridge: Cambridge University Press.10.1017/CCOL0521382483
Search in Google Scholar
Janczak A. (1999): Parameter estimation based fault detection and isolation in Wiener and Hammerstein systems. — Int. J. Appl. Math. Comput. Sci., Vol. 9, No. 3, pp. 711-735.
Search in Google Scholar
Jang W. and Kim G. (1994): Identification of loudspeaker non-linearities using the NARMAX modelling technique. — J. Audio Eng. Soc., Vol. 42, No. 1/2, pp. 50-59.
Search in Google Scholar
Krzyżak A. (1990): On estimation of a class of nonlinear systems by the kernel regression estimate. — IEEE Trans. Inf. Theory, Vol. IT-36, No. 1, pp. 141-152.10.1109/18.50380
Search in Google Scholar
Krzyżak A., Sąsiadek J. and Kégl B. (2001): Identification of dynamic nonlinear systems using the Hermite series approach. — Int. J. Syst. Sci., Vol. 32, No. 10, pp. 1261-1285.10.1080/00207720110045056
Search in Google Scholar
Latawiec K.J. (2004): The Power of Inverse Systems in Linear and Nonlinear Modeling and Control. — Opole: Opole University of Technology Press.
Search in Google Scholar
Ljung L. (1987): System Identification: Theory for the User. — Englewood Cliffs, NJ: Prentice Hall.
Search in Google Scholar
Narendra K.S. and Gallman P.G. (1966): An iterative method for the identification of nonlinear systems using the Hammerstein model. — IEEE Trans. Automat. Contr., Vol. 11, No. 3, pp. 546-550.10.1109/TAC.1966.1098387
Search in Google Scholar
Pawlak M. and Hasiewicz Z. (1998): Nonlinear system identification by the Haar multiresolution analysis. — IEEE Trans. Circ. Syst., Vol. 45, No. 9, pp. 945-961.10.1109/81.721260
Search in Google Scholar
Söderström T. and Stoica P. (1982): Instrumental-variable methods for identification of Hammerstein systems. — Int. J. Contr., Vol. 35, No. 3, pp. 459-476.
Search in Google Scholar
Söderström T. and Stoica P. (1989): System Identification. — Englewood Cliffs, NJ: Prentice Hall.
Search in Google Scholar
Van den Hof P., Heuberger P. and Bokor J. (1995): System identification with generalized orthonormal basis functions. — Automatica, Vol. 31, No. 12, pp. 1821-1834.10.1016/0005-1098(95)00074-4
Search in Google Scholar
Vörös J. (1999): Iterative algorithm for identification of Hammerstein systems with two-segment nonlinearities. — IEEE Trans. Automat. Contr., Vol. 44, No. 11, pp. 2145-2149.10.1109/9.802933
Search in Google Scholar
Wand M.P. and Jones H.C. (1995): Kernel Smoothing. — London: Chapman and Hall.10.1007/978-1-4899-4493-1
Search in Google Scholar
Zhang Y.K. and Bai E.W. (1996): Simulation of spring discharge from a limestone aquifer in Iowa. — Hydrogeol. J., Vol. 4, No. 4, pp. 41-54.10.1007/s100400050087
Search in Google Scholar
Zhu Y. (2000): Identification of Hammerstein models for control using ASYM. — Int. J. Contr., Vol. 73, No. 18, pp. 1692-1702.10.1080/00207170050201771
Search in Google Scholar

Generalized Kernel Regression Estimate for the Identification of Hammerstein Systems

Abstract

Paradigm

My account