Recent experiences in the industrial exploitation of principal component based fault detection methods

Summary form only given. There has been a plethora of research and many industrial studies involving the application of multivariate statistical techniques for the detection and isolation of faults on process plants. Despite all this work and reported success of the studies, there remain very few online applications of the technology. To address this issue and to determine the capabilities and limitations of the multivariate statistical techniques, the Control Technology Centre Ltd. has undertaken a comprehensive study aimed specifically at the development of online multivariate solutions for process plants. This paper provides details of several of these investigations. These investigations highlight a number of factors which should be considered when conducting multivariate analysis. For example, there appears to be a need for consideration of cause and effect relationships, which is often ignored in other application studies. In addition the integration of the multivariate statistical routines into a hierarchical system is demonstrated to provide the potential for robust and self-diagnosing control systems.

[1]  E. F. Vogel,et al.  A plant-wide industrial process control problem , 1993 .

[2]  S. Qin.  Recursive PLS algorithms for adaptive data modeling , 1998 .

[3]  Paul M. Frank,et al.  Analytical and Qualitative Model-based Fault Diagnosis - A Survey and Some New Results , 1996, Eur. J. Control.

[4]  Uwe Kruger,et al.  Recursive partial least squares algorithms for monitoring complex industrial processes , 2003 .

[5]  Julian Morris,et al.  Non-Linear Model Based Predictive Control Through Dynamic Non-Linear Partial Least Squares , 2002 .

[6]  José R. Dorronsoro,et al.  Autoassociative neural networks and noise filtering , 2003, IEEE Trans. Signal Process..

[7]  A. J. Morris,et al.  Impurity Detection in Batch Pharmaceutical Production , 2001 .

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

[9]  Richard J. Beckman,et al.  Intelligent Instrumentation. , 1986, Journal of research of the National Bureau of Standards.

[10]  Venkat Venkatasubramanian,et al.  PCA-SDG based process monitoring and fault diagnosis , 1999 .

[11]  John F. MacGregor,et al.  Multivariate SPC charts for monitoring batch processes , 1995 .

[12]  A. J. Morris,et al.  An overview of multivariate statistical process control in continuous and batch process performance monitoring , 1996 .

[13]  B. Kowalski,et al.  Partial least-squares regression: a tutorial , 1986 .

[14]  G. Barney,et al.  Intelligent instrumentation , 1988 .

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

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

[17]  J. Macgregor,et al.  Identification of finite impulse response models : Methods and robustness issues , 1996 .

[18]  J. Macgregor,et al.  Monitoring batch processes using multiway principal component analysis , 1994 .

[19]  Sirish L. Shah,et al.  Monitoring Batch Processes Using Multivariate Statistical Tools: Extensions and Practical Issues , 1996 .

[20]  T. McAvoy,et al.  Batch tracking via nonlinear principal component analysis , 1996 .

[21]  J. A. Rossiter,et al.  Non-linear model based predictive control , 1999 .

[22]  John F. MacGregor,et al.  Multi-way partial least squares in monitoring batch processes , 1995 .

[23]  Christos Georgakis,et al.  Dynamic simulator for a Model IV fluid catalytic cracking unit , 1993 .

[24]  B Lennox,et al.  Process monitoring of an industrial fed-batch fermentation. , 2001, Biotechnology and bioengineering.

[25]  P. A. Taylor,et al.  Missing data methods in PCA and PLS: Score calculations with incomplete observations , 1996 .

[26]  Barry Lennox,et al.  Fault detection in continuous processes using multivariate statistical methods , 2000, Int. J. Syst. Sci..

[27]  P. A. Taylor,et al.  Synchronization of batch trajectories using dynamic time warping , 1998 .

[28]  Barry M. Wise,et al.  The process chemometrics approach to process monitoring and fault detection , 1995 .

[29]  Michael J. Piovoso,et al.  A multivariate statistical controller for on-line quality improvement , 1998 .

[30]  Barry Lennox,et al.  Industrial application of neural networks — an investigation , 2001 .

[31]  S. Wold.  Cross-Validatory Estimation of the Number of Components in Factor and Principal Components Models , 1978 .

[32]  Chen Gang,et al.  An improved base control for the Tennessee Eastman problem , 1995, Proceedings of 1995 American Control Conference - ACC'95.

[33]  Thomas F. Edgar,et al.  Identification of faulty sensors using principal component analysis , 1996 .