Robust identification of piecewise/switching autoregressive exogenous process

A robust identification approach for a class of switching processes named PWARX (piecewise autoregressive exogenous) processes is developed in this article. It is proposed that the identification problem can be formulated and solved within the EM (expectation-maximization) algorithm framework. However, unlike the regular EM algorithm in which the objective function of the maximization step is built upon the assumption that the noise comes from a single distribution, contaminated Gaussian distribution is utilized in the process of constructing the objective function, which effectively makes the revised EM algorithm robust to the latent outliers. Issues associated with the EM algorithm in the PWARX system identification such as sensitivity to its starting point as well as inability to accurately classify “un-decidable” data points are examined and a solution strategy is proposed. Data sets with/without outliers are both considered and the performance is compared between the robust EM algorithm and regular EM algorithm in terms of their parameter estimation performance. Finally, a modified version of MRLP (multi-category robust linear programming) region partition method is proposed by assigning different weights to different data points. In this way, negative influence caused by outliers could be minimized in region partitioning of PWARX systems. Simulation as well as application on a pilot-scale switched process control system are used to verify the efficiency of the proposed identification algorithm. © 2009 American Institute of Chemical Engineers AIChE J, 2010

[1]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[2]  R. H. Fariss,et al.  An efficient computational technique for generalized application of maximum likelihood to improve correlation of experimental data , 1979 .

[3]  L. Biegler,et al.  Simultaneous strategies for data reconciliation and gross error detection of nonlinear systems , 1991 .

[4]  Dale E. Seborg,et al.  Nonlinear control strategies for continuous fermenters , 1992 .

[5]  Paul I. Barton,et al.  Modeling of combined discrete/continuous processes , 1994 .

[6]  O. Mangasarian,et al.  Multicategory discrimination via linear programming , 1994 .

[7]  L. Biegler,et al.  Data reconciliation and gross‐error detection for dynamic systems , 1996 .

[8]  Yaman Arkun,et al.  Model predictive control of plant transitions using a new identification technique for interpolating nonlinear models , 1998 .

[9]  David A. Landgrebe,et al.  Robust parameter estimation for mixture model , 2000, IEEE Trans. Geosci. Remote. Sens..

[10]  Alberto Bemporad,et al.  Observability and controllability of piecewise affine and hybrid systems , 2000, IEEE Trans. Autom. Control..

[11]  Julio R. Banga,et al.  Optimization of hybrid discrete/continuous dynamic systems , 2000 .

[12]  Bart De Schutter,et al.  Equivalence of hybrid dynamical models , 2001, Autom..

[13]  Christophe Biernacki,et al.  Choosing starting values for the EM algorithm for getting the highest likelihood in multivariate Gaussian mixture models , 2003, Comput. Stat. Data Anal..

[14]  Didier Maquin,et al.  Parameter estimation of switching piecewise linear system , 2003, 42nd IEEE International Conference on Decision and Control (IEEE Cat. No.03CH37475).

[15]  Manfred Morari,et al.  A clustering technique for the identification of piecewise affine systems , 2001, Autom..

[16]  Panagiotis D. Christofides,et al.  Coordinating feedback and switching for control of hybrid nonlinear processes , 2003 .

[17]  Ravindra D. Gudi,et al.  Identification of complex nonlinear processes based on fuzzy decomposition of the steady state space , 2003 .

[18]  A. Juloski,et al.  A Bayesian approach to identification of hybrid systems , 2004, CDC.

[19]  Paul I. Barton,et al.  Design of process operations using hybrid dynamic optimization , 2004, Comput. Chem. Eng..

[20]  José Ragot,et al.  DATA RECONCILIATION: A ROBUST APPROACH USING CONTAMINATED DISTRIBUTION. APPLICATION TO A PETROCHEMICAL PROCESS , 2004 .

[21]  Alberto Bemporad,et al.  Identification of piecewise affine systems via mixed-integer programming , 2004, Autom..

[22]  Ravindra D. Gudi,et al.  Multi-model decomposition of nonlinear dynamics using a fuzzy-CART approach , 2005 .

[23]  Panagiotis D. Christofides,et al.  Output feedback control of switched nonlinear systems using multiple Lyapunov functions , 2001, Proceedings of the 2005, American Control Conference, 2005..

[24]  Alberto Bemporad,et al.  A bounded-error approach to piecewise affine system identification , 2005, IEEE Transactions on Automatic Control.

[25]  Panagiotis D. Christofides,et al.  Predictive control of switched nonlinear systems with scheduled mode transitions , 2005, IEEE Transactions on Automatic Control.

[26]  Sheetal Kalyani,et al.  Robust Statistics Based Expectation-Maximization Algorithm for Channel Tracking in OFDM Systems , 2007, 2007 IEEE International Conference on Communications.

[27]  René Vidal,et al.  Identification of Hybrid Systems: A Tutorial , 2007, Eur. J. Control.