A novel dynamic PCA algorithm for dynamic data modeling and process monitoring

Abstract Principal component analysis (PCA) has been widely applied for data modeling and process monitoring. However, it is not appropriate to directly apply PCA to data from a dynamic process, since PCA focuses on variance maximization only and pays no attention to whether the components contain dynamics or not. In this paper, a novel dynamic PCA (DiPCA) algorithm is proposed to extract explicitly a set of dynamic latent variables with which to capture the most dynamic variations in the data. After the dynamic variations are extracted, the residuals are essentially uncorrelated in time and static PCA can be applied. The new models generate a subspace of principal time series that are most predictable from their past data. Geometric properties are explored to give insight into the new dynamic model structure. For the purpose of process monitoring, fault detection indices based on DiPCA are developed based on the proposed model. Case studies on simulation data, data from an industrial boiler process, and the Tennessee Eastman process are presented to illustrate the effectiveness of the proposed dynamic models and fault detection methods.

[1]  Christos Georgakis,et al.  Disturbance detection and isolation by dynamic principal component analysis , 1995 .

[2]  S. Qin,et al.  Selection of the Number of Principal Components: The Variance of the Reconstruction Error Criterion with a Comparison to Other Methods† , 1999 .

[3]  Ping Zhang,et al.  A comparison study of basic data-driven fault diagnosis and process monitoring methods on the benchmark Tennessee Eastman process , 2012 .

[4]  Donghua Zhou,et al.  A New Method of Dynamic Latent-Variable Modeling for Process Monitoring , 2014, IEEE Transactions on Industrial Electronics.

[5]  Fuli Wang,et al.  Two‐dimensional dynamic PCA for batch process monitoring , 2005 .

[6]  Ali Cinar,et al.  Statistical quality control of multivariable continuous processes , 1994, Proceedings of 1994 American Control Conference - ACC '94.

[7]  S. Joe Qin,et al.  Reconstruction-Based Fault Identification Using a Combined Index , 2001 .

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

[9]  Junghui Chen,et al.  On-line batch process monitoring using dynamic PCA and dynamic PLS models , 2002 .

[10]  Catherine Porte,et al.  Automation and optimization of glycine synthesis , 1996 .

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

[12]  I. Helland.  ON THE STRUCTURE OF PARTIAL LEAST SQUARES REGRESSION , 1988 .

[13]  Donghua Zhou,et al.  Dynamic latent variable modeling for statistical process monitoring , 2011 .

[14]  Petre Stoica,et al.  Decentralized Control , 2018, The Control Systems Handbook.

[15]  Xiuxi Li,et al.  Nonlinear dynamic principal component analysis for on-line process monitoring and diagnosis , 2000 .

[16]  Donghua Zhou,et al.  Total projection to latent structures for process monitoring , 2009 .

[17]  Thomas E. Marlin,et al.  Multivariate statistical monitoring of process operating performance , 1991 .

[18]  John F. MacGregor,et al.  Process monitoring and diagnosis by multiblock PLS methods , 1994 .

[19]  S. Joe Qin,et al.  Dynamic-Inner Partial Least Squares for Dynamic Data Modeling , 2015 .

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