Nonlinear Statistical Process Monitoring based on Competitive Principal Component Analysis

Traditional process monitoring techniques assume the normal operating conditions (NOC) to be distributed normally. However, for processes with more than one operating regime, building a single subspace model to monitor the whole process operation performance may not be efficient and will lead to high rate of missing alarm. To handle this situation, a monitoring strategy using multiple subspace models is presented in this paper. From the experimental results using a simulation model of a continuous flow aerated bioreactor for wastewater treatment in pulp and paper industry it has been shown that the proposed approach is very promising.

[1]  John F. MacGregor.  STATISTICAL PROCESS CONTROL OF MULTIVARIATE PROCESSES , 1994 .

[2]  J. E. Jackson,et al.  Control Procedures for Residuals Associated With Principal Component Analysis , 1979 .

[3]  Sam T. Roweis,et al.  EM Algorithms for PCA and SPCA , 1997, NIPS.

[4]  T. Hastie,et al.  Principal Curves , 2007 .

[5]  Andrew R. Webb.  An approach to non-linear principal components analysis using radially symmetric kernel functions , 1996, Stat. Comput..

[6]  Silvio Simani,et al.  Model-based fault diagnosis in dynamic systems using identification techniques , 2003 .

[7]  Janos Gertler,et al.  A new structural framework for parity equation-based failure detection and isolation , 1990, Autom..

[8]  Thomas F. Edgar,et al.  Use of principal component analysis for sensor fault identification , 1996 .

[9]  Nanda Kambhatla,et al.  Dimension Reduction by Local Principal Component Analysis , 1997, Neural Computation.

[10]  James C. Bezdek,et al.  Pattern Recognition with Fuzzy Objective Function Algorithms , 1981, Advanced Applications in Pattern Recognition.

[11]  Thomas J. McAvoy,et al.  Nonlinear PLS Modeling Using Neural Networks , 1992 .

[12]  Rolf Isermann,et al.  Supervision, fault-detection and fault-diagnosis methods — An introduction , 1997 .

[13]  Paul M. Frank,et al.  Fault diagnosis in dynamic systems using analytical and knowledge-based redundancy: A survey and some new results , 1990, Autom..

[14]  Jin Cao,et al.  Partial PCA-based optimal structured residual design for fault isolation , 2004, Proceedings of the 2004 American Control Conference.

[15]  Weihua Li,et al.  Isolation enhanced principal component analysis , 1999 .

[16]  Janos Gertler,et al.  Fault detection and diagnosis in engineering systems , 1998 .

[17]  Mark A. Kramer,et al.  Autoassociative neural networks , 1992 .

[18]  G. Box.  Some Theorems on Quadratic Forms Applied in the Study of Analysis of Variance Problems, I. Effect of Inequality of Variance in the One-Way Classification , 1954 .

[19]  George W. Irwin,et al.  RBF principal manifolds for process monitoring , 1999, IEEE Trans. Neural Networks.

[20]  Geoffrey E. Hinton,et al.  Modeling the manifolds of images of handwritten digits , 1997, IEEE Trans. Neural Networks.

[21]  Michael E. Tipping,et al.  Probabilistic Principal Component Analysis , 1999 .

[22]  Christopher M. Bishop,et al.  Mixtures of Probabilistic Principal Component Analyzers , 1999, Neural Computation.

[23]  T. J. McAvov,et al.  BASE CONTROL FOR THE TENNESSEE EASTMAN PROBLEM , 2001 .

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