Process monitoring in principal component subspace: part 1 - fault reconstruction study

The principal component analysis (PCA) is a kind of data-driven modeling method that has wide applications in the field of industrial process monitoring and product quality control. However, it was shown that some faults can only be detected in the principal component subspace (PCS) and the T/sup 2/ statistic in PCS is more robust than SPE statistic while the latter is in the residual subspace (RS). A reconstruction approach for these faults in the PCS is proposed to estimate the fault magnitude and then judge its type. The reconstructability conditions both for complete and partial ones are derived mathematically and the obtained results are illustrated and verified by simulation studies on a double-effective evaporator.

[1]  Jin Cao,et al.  PCA-based fault diagnosis in the presence of control and dynamics , 2004 .

[2]  Zhihuan Song,et al.  Fault detection behavior and performance analysis of principal component analysis based process monitoring methods , 2002 .

[3]  S. Joe Qin,et al.  Statistical process monitoring: basics and beyond , 2003 .

[4]  Richard D. Braatz,et al.  Fault Detection and Diagnosis in Industrial Systems , 2001 .

[5]  B. Bakshi.  Multiscale PCA with application to multivariate statistical process monitoring , 1998 .

[6]  S. Joe Qin,et al.  Subspace approach to multidimensional fault identification and reconstruction , 1998 .

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

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

[9]  S. J. Qin,et al.  Extracting fault subspaces for fault identification of a polyester film process , 2001, Proceedings of the 2001 American Control Conference. (Cat. No.01CH37148).

[10]  Haiqing Wang,et al.  Process monitoring in principal component subspace: part 2. Fault identification and isolation study , 2004, 2004 IEEE International Conference on Systems, Man and Cybernetics (IEEE Cat. No.04CH37583).

[11]  Chonghun Han,et al.  Fault Detection and Operation Mode Identification Based on Pattern Classification with Variable Selection , 2004 .

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