Abstract
This paper considers the robust recursive stochastic gradient algorithm for identification of multivariable Hammerstein model with a static nonlinear block in polynomial form and a linear block described by output-error model. The algorithm is designed for unknown parameters in vector form. It is assumed that there is a priori information about a distribution class to which a real disturbance belongs. Such class of distributions describes the presence of outliers in observations. The main contributions of the paper are: (i) design of robust stochastic approximation algorithm for MIMO Hammerstein models using robust statistics (Huber’s theory); (ii) design of general form of nonlinear block; (iii) a strong consistency of estimated parameter whereby proof is based on martingale theory, generalized strictly positive real condition and persistent excitation condition. The properties of algorithm are illustrated by simulations.
Similar content being viewed by others
References
Kushner, K.J., Yin, G.G.: Stochastic Approximation Algorithms and Applications. Springer, Berlin (2003)
Chen, H.F.: Stochastic Approximation and its Applications. Kluwer, New York (2002)
Benveniste, A., Metivier, M., Priouret, P.: Adaptive Algorithms and Stochastic Approximations. Springer, Berlin (1990)
Chen, H.F., Zhao, W.: Recursive Identification and Parameter Estimation. CRC Press, Boca Raton (2014)
Ding, F., Chen, T.: Identification of Hammerstein nonlinear ARMAX systems. Automatica 41, 1479–1489 (2005)
Wang, F., Ding, F.: Gradient—based iterative identification methods for multivariate pseudo-linear mobbing average systems using the data filtering. Nonlinear Dyn. 84, 2003–2015 (2016)
Ma, L., Liu, X.: A nonlinear recursive instrumental variable identification method of Hammerstein ARMAX system. Nonlinear Dyn. 79, 1601–1613 (2015)
Voros, J.: Recursive identification of discrete-time nonlinear cascade systems with time-varying output hysteresis. Nonlinear Dyn. (2016). doi:10.1007/s11071-0163124-3
Abu-Arqub, O., Abo-Hammour, Z.: Numerical solution of systems of second-order boundary value problems using continuous genetic algorithm. Inf. Sci. 279, 396–415 (2014)
Zhang, Y.Q., Wang, X.Y.: A symmetric image encryption algorithm based on mixed linear-nonlinear coupled map lattice. Inf. Sci. 273, 329–351 (2014)
Zhang, Y.Q., Wang, X.Y.: A new image encryption algorithm based on non-adjacent coupled map lattice. Appl. Soft Comput. 26, 10–20 (2015)
Zhang, Y.Q., Wang, X.Y., Liu, J., Chi, Z.L.: An image encryption scheme based on the MLNCML system using DNA sequences. Opt. Lasers Eng. 82, 95–103 (2016)
Giri, E., Bai, E. (eds.): Block-Oriented Nonlinear System Identification. Springer, Berlin (2010)
Mzyk, G.: Combined Parametric-Nonparametric Identification of Block-Oriented Systems. Springer, Berlin (2014)
Sliwinski, P.: Nonlinear Systems Identification by Haar Wavelets. Springer, Berlin (2013)
Narendra, K., Galmann, P.: An iterative method for the identification of nonlinear systems using a Hammerstein model. IEEE Trans. Autom. Control 11, 546–550 (1966)
Yan, J., Bernstein, S.: Minimum modeling retrospective cost adaptive control of uncertain Hammerstein system using auxiliary nonlinearities. Int. J. Control 87, 483–505 (2014)
Fabri, S.G., Wittenmark, B., Bugeja, M.K.: Dual adaptive extremum control of Hammerstein systems. Int. J. Control 88, 1271–1286 (2015)
Barnett, V., Lewis, T.: Outliers in Statistical Data. Wiley, New York (1994)
Pearson, R.K.: Mining Imperfect Data. SIAM, Philadelphia (2005)
Pearson, R.K.: Exploring Data in Engineering, The Sciences and Medicine. Oxford University Press, Oxford (2011)
Huber, P.: Robust estimation of a location parameter. Ann. Math. Stat. 35, 73–101 (1964)
Huber, P., Ronchetti, E.: Robust Statistics. Wiley, London (2009)
Tsypkin, Y.Z.: On the Foundation of Information Identification Theory. Nauka, Moscow (1984). (in Russian)
Filipovic, V.Z.: Consistency of the robust recursive Hammerstein model identification algorithm. J. Frankl. Inst. 352, 1932–1945 (2015)
Filipovic, V.Z.: Recursive identification of multivariable ARX models in the presence of a priori information: robustness and regularization. Signal Process. 116, 68–77 (2015)
Filipovic, V.Z.: Stochastic multivariable self-tuning tracker for non-Gaussian systems. Int. J. Appl. Math. Comput. Sci. 15, 351–357 (2005)
Filipovic V.Z.: A global convergent outlier robust adaptive predictor for MIMO Hammerstein models. Int. J. Robust. Nonlinear Control. (2016). doi:10.1002/rnc:3705
Greblicki, W.: Stochastic approximation in nonparametric identification of Hammerstein systems. IEEE Trans. Autom. Control 47, 1800–1810 (2002)
Hersh, M., Zarrop, M.: Strong consistency of a class of recursive stochastic gradient algorithms. Int. J. Control 43, 1115–1123 (1986)
Stout, W.: Almost Sure Convergence. Academic Press, New York (1974)
Desoer, C., Vidyasagar, M.: Feedback Systems: Input–Output Properties. Academic Press, New York (1975)
Kottenstette, N., McCourt, M., Xia, M., Gupta, A., Antsaklis, P.: On relationships among positivity, positive realness, and dissipativity. Automatica 50, 1003–1016 (2014)
Ljung, L., Soderstrom, T.: Theory and Practice of Recursive Identification. MIT Press, Massachusetts (1983)
Papoulis, A.: Probability, Random Variables, and Stochastic Processes. McGraw-Hill, New York (1965)
Caines, P.: Linear Stochastic Systems. Wiley, New York (1988)
Tukey, J.: A survey for sampling from contaminated distributions. In: Olkin, I. (ed.) Contributions for Probability and Statistics, pp. 448–485. Stanford University Press, Stanford (1960)
Bromwich, T.: An Introduction to the Theory of Infinite Series. MacMilan, London (1947)
Ding, F., Chen, T.: Performance analysis of multi-inovation gradient type identification methods. Automatica 43, 1–14 (2007)
Acknowledgements
This work was supported by the Republic of Serbia, Ministry of Education and Science, through project No. 33026-TR. The author thanks to the referees for their constructive comments that have helped me to improve the article significantly.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Filipovic, V.Z. Outlier robust stochastic approximation algorithm for identification of MIMO Hammerstein models. Nonlinear Dyn 90, 1427–1441 (2017). https://doi.org/10.1007/s11071-017-3736-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11071-017-3736-2