Kalman filter based fault detection for two-dimensional systems

Abstract Fault detection and isolation for two-dimensional (2-D) systems represent a great challenge in both theoretical development and applications. Use of Kalman filters in fault detection has been well developed for one-dimensional (1-D) system. In this paper, a fault detection algorithm is developed for 2-D systems described by the Fornasini & Marchesini (F–M) model. Based on the state estimate from a recursive 2-D Kalman filter, a residual is generated. From the model of residual over a 2-D evaluation window, the residual is explicitly related to faults within the evaluation window. Using generalized likelihood ratio (GLR), residual evaluation is carried out and compared with a calculated threshold value for fault detection. Simulation results indicate that faults can be detected effectively using the proposed method.

[1]  Steven X. Ding,et al.  Threshold computation for fault detection in linear discrete-time Markov jump systems , 2014 .

[2]  M. Ohki,et al.  Two-dimensional LMS adaptive filters , 1991 .

[3]  Sandro Zampieri.  2D residual generation and dead-beat observers , 1991 .

[4]  M. Kosaka,et al.  Recursive filtering algorithm for a two-dimensional system , 1979 .

[5]  Lorenzo Ntogramatzidis,et al.  Detectability subspaces and observer synthesis for two-dimensional systems , 2012, Multidimens. Syst. Signal Process..

[6]  G. E. Antoniou,et al.  Minimal state-space realization of factorable 2-D transfer functions , 1988 .

[7]  R. Roesser.  A discrete state-space model for linear image processing , 1975 .

[8]  Michèle Basseville,et al.  Detecting changes in signals and systems - A survey , 1988, Autom..

[9]  Ahmed Alenany,et al.  Subspace algorithms for identifying separable-in-denominator two-dimensional systems with deterministic inputs , 2011 .

[10]  Zidong Wang,et al.  A fault detection scheme for linear discrete-time systems with an integrated online performance evaluation , 2014, Int. J. Control.

[11]  Frank L. Lewis,et al.  A note on Kalman filtering , 1999 .

[12]  A. Willsky,et al.  A generalized likelihood ratio approach to the detection and estimation of jumps in linear systems , 1976 .

[13]  Steven X. Ding,et al.  Fault detection scheme for discrete-time Markov jump linear systems with mode-independent residual , 2012, 2012 IEEE International Conference on Control Applications.

[14]  Jinsong Zhao,et al.  A hybrid fault diagnosis strategy for chemical process startups , 2012 .

[15]  Huijun Gao,et al.  I filtering for 2D Markovian jump systems , 2008, Autom..

[16]  G. Marchesini,et al.  State-space realization theory of two-dimensional filters , 1976 .

[17]  M. Sebek,et al.  Polynomial solution of 2D Kalman-Bucy filtering problem , 1992 .

[18]  Eric Rogers,et al.  A Geometric Approach to 3D Fault Identification , 2013 .

[19]  Yun Zou,et al.  H ∞ control for 2-D singular delayed systems , 2011, Int. J. Syst. Sci..

[20]  Józef Korbicz,et al.  Analytical and knowledge-based redundancy for fault diagnosis in process plants , 1993 .

[21]  Ligang Wu,et al.  Mixed H2/H∞ approach to fault detection of discrete linear repetitive processes , 2011, J. Frankl. Inst..

[22]  Steven X. Ding,et al.  Model-based Fault Diagnosis Techniques: Design Schemes, Algorithms, and Tools , 2008 .

[23]  Guillaume Mercère,et al.  Subspace algorithms for identifying separable-in-denominator 2D systems with deterministic–stochastic inputs , 2016, Int. J. Control.

[24]  Shengyuan Xu,et al.  A Generalized Kalman Filter for 2D Discrete Systems , 2004, The 2004 47th Midwest Symposium on Circuits and Systems, 2004. MWSCAS '04..

[25]  Wei Xing Zheng,et al.  Generalized H2 fault detection for two-dimensional Markovian jump systems , 2012, Autom..

[26]  Mauro Bisiacco,et al.  Unknown input observers for 2D state-space models , 2004 .

[27]  Erik Frisk,et al.  Data-driven and adaptive statistical residual evaluation for fault detection with an automotive application , 2014 .

[28]  T. Kaczorek.  Perfect Observers for Singular 2D Linear Systems , 2001 .

[29]  S. Attasi.  Modelling and Recursive Estimation for Double Indexed Sequences , 1976 .

[30]  Yun Zou,et al.  Non-Fragile Robust Guaranteed Cost Control of 2-D Discrete Uncertain Systems Described by the General Models , 2011, Circuits Syst. Signal Process..

[31]  Helen Shang,et al.  Observer based fault detection for two dimensional systems described by Roesser models , 2015, Multidimens. Syst. Signal Process..

[32]  Alf J. Isaksson,et al.  Analysis of identified 2-D noncausal models , 1993, IEEE Trans. Inf. Theory.

[33]  Huijun Gao,et al.  Robust finite frequency H∞ filtering for uncertain 2-D systems: The FM model case , 2013, Autom..

[34]  Ettore Fornasini,et al.  Doubly-indexed dynamical systems: State-space models and structural properties , 1978, Mathematical systems theory.

[35]  Yun Zou,et al.  Robust H∞ filtering for uncertain two-dimensional discrete systems with state-varying delays , 2010 .

[36]  Maria Elena Valcher,et al.  Observer-based Fault Detection and Isolation for 2D State-space Models , 2006, Multidimens. Syst. Signal Process..

[37]  Cishen Zhang,et al.  H2 and mixed H2/Hinfinity control of two-dimensional systems in Roesser model , 2006, Autom..

[38]  A. Kawakami,et al.  A realization method of two-dimensional rational transfer functions , 1990 .

[39]  Huijun Gao,et al.  Robust finite frequency Hinfinity filtering for uncertain 2-D Roesser systems , 2012, Autom..

[40]  Hans Zwart,et al.  A Review on Realization Theory for Infinite-Dimensional Systems , 2017 .

[41]  Ping-Ya Zhao,et al.  An unbiased and computationally efficient LS estimation method for identifying parameters of 2D noncausal SAR models , 1993, IEEE Trans. Signal Process..

[42]  Raghunathan Rengaswamy,et al.  A review of process fault detection and diagnosis: Part I: Quantitative model-based methods , 2003, Comput. Chem. Eng..

[43]  S. Treitel,et al.  Stability and synthesis of two-dimensional recursive filters , 1972 .

[44]  T. Çiftçibasi,et al.  Sufficient or necessary conditions for modal controllability and observability of Roesser's 2-D system model , 1983 .

[45]  JOHN w. WOODS,et al.  Kalman filtering in two dimensions , 1977, IEEE Trans. Inf. Theory.

[46]  Chuei-Tin Chang,et al.  Simplification techniques for EKF computations in fault diagnosis—suboptimal gains , 1998 .

[47]  Zhiping Lin,et al.  A direct-construction approach to multidimensional realization and LFR uncertainty modeling , 2008, Multidimens. Syst. Signal Process..