Consistency of the robust recursive Hammerstein model identification algorithm

In this paper, it is proposed a robust recursive algorithm for identification of a Hammerstein model with a static nonlinear block in polynomial form and a linear block described by ARMAX model. It is assumed that there is a priori information about a distribution class to which a disturbance belongs. Such assumption introduces a nonlinear transformation of the prediction error in the recursive algorithm. The obtained algorithm is robust in relation to the uncertainty of the disturbance distribution. By using the stochastic Lyapunov function and the martingale theory a strong consistency of estimated parameters is proved under generalized strict real positivity conditions, based on the theory of passive operators and the weakest possible excitation. The practical behavior of the robust algorithm is illustrated by simulations.

[1]  O. Nelles.  Nonlinear System Identification , 2001 .

[2]  J. Chu,et al.  Gradient-based and least-squares-based iterative algorithms for Hammerstein systems using the hierarchical identification principle , 2013 .

[3]  Feng Ding,et al.  An efficient hierarchical identification method for general dual-rate sampled-data systems , 2014, Autom..

[4]  Debasish Roy,et al.  Iterated gain-based stochastic filters for dynamic system identification , 2014, J. Frankl. Inst..

[5]  Robert Haber.  Nonlinear System Identification : Input-output Modeling Approach , 1999 .

[6]  R. Luus,et al.  A noniterative method for identification using Hammerstein model , 1971 .

[7]  Han-Fu Chen,et al.  Identification and Stochastic Adaptive Control , 1991 .

[8]  Feng Ding,et al.  Identification methods for Hammerstein nonlinear systems , 2011, Digit. Signal Process..

[9]  David T. Westwick,et al.  Identification of Hammerstein models with cubic spline nonlinearities , 2004, IEEE Transactions on Biomedical Engineering.

[10]  E. Bai,et al.  Block Oriented Nonlinear System Identification , 2010 .

[11]  Feng Ding,et al.  Identification of Hammerstein nonlinear ARMAX systems , 2005, Autom..

[12]  Feng Ding,et al.  Data filtering based recursive least squares algorithm for Hammerstein systems using the key-term separation principle , 2013, Inf. Sci..

[13]  C. Desoer,et al.  Feedback Systems: Input-Output Properties , 1975 .

[14]  Branko D. Kovacevic,et al.  On robust AML identification algorithms , 1994, Autom..

[15]  Vic Barnett,et al.  Outliers in Statistical Data , 1980 .

[16]  Han-Fu Chen,et al.  Convergence rate of least-squares identification and adaptive control for stochastic systems† , 1986 .

[17]  José Luis Figueroa,et al.  Robust model predictive control of a Wiener-like system , 2013, J. Frankl. Inst..

[18]  P. J. Huber.  Robust Estimation of a Location Parameter , 1964 .

[19]  K. Narendra,et al.  An iterative method for the identification of nonlinear systems using a Hammerstein model , 1966 .

[20]  S. Foss,et al.  An Introduction to Heavy-Tailed and Subexponential Distributions , 2011 .

[21]  Lennart Ljung,et al.  Some Aspects on Nonlinear System Identification , 2006 .

[22]  W. Stout.  Almost sure convergence , 1974 .

[23]  F. Ding.  Hierarchical multi-innovation stochastic gradient algorithm for Hammerstein nonlinear system modeling , 2013 .

[24]  V. Filipovic.  Stochastic multivariable self-tuning tracker for non-gaussian systems , 2005 .

[25]  Feng Ding,et al.  Combined parameter and output estimation of dual-rate systems using an auxiliary model , 2004, Autom..

[26]  Feng Ding,et al.  Hierarchical Least Squares Estimation Algorithm for Hammerstein–Wiener Systems , 2012, IEEE Signal Processing Letters.

[27]  V. Filipovic,et al.  On robustified adaptive minimum-variance controller , 1996 .

[28]  Eric Rogers,et al.  Recursive identification of Hammerstein systems with application to electrically stimulated muscle , 2012 .