Batch-to-Batch Steady State Identification Based on Variable Correlation and Mahalanobis Distance

Online steady state identification (SSID) is an important task to ensure the quality consistence of final products in batch processes. Additionally, it is also critical for satisfactory control of many batch processes. The existing approach for batch process SSID is based on the multiway principal component analysis (MPCA) method, which requires history data from dozens of batches for process modeling. Consequently, this limits its online applications. In this work, principal component analysis (PCA) models are built for each batch from which the variable correlation information is extracted. The changes in variable correlation structures are then quantified with a PCA similarity factor. At the same time, the Mahalanobis distances between batch trajectories are also calculated to indicate the changes in variable trajectory magnitudes. These two types of information are then used for online SSID in batch processes. This method is more suitable for online applications and can solve the problems of uneven op...

[1]  R. H. Kent,et al.  The Mean Square Successive Difference , 1941 .

[2]  J. Neumann.  Distribution of the Ratio of the Mean Square Successive Difference to the Variance , 1941 .

[3]  W. Krzanowski.  Between-Groups Comparison of Principal Components , 1979 .

[4]  Christian Jutten,et al.  Blind separation of sources, part I: An adaptive algorithm based on neuromimetic architecture , 1991, Signal Process..

[5]  J. Edward Jackson,et al.  A User's Guide to Principal Components. , 1991 .

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

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

[8]  R. R. Rhinehart,et al.  An efficient method for on-line identification of steady state , 1995 .

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

[10]  R. Russell Rhinehart,et al.  Critical values for a steady-state identifier , 1997 .

[11]  R. R. Rhinehart,et al.  Automated steady-state identification in multivariable systems : Special report : Process control and information systems , 2000 .

[12]  A. Singhal,et al.  Pattern matching in historical batch data using PCA , 2002 .

[13]  S. Zhao,et al.  Monitoring of Processes with Multiple Operating Modes through Multiple Principle Component Analysis Models , 2004 .

[14]  Fuli Wang,et al.  PCA-Based Modeling and On-line Monitoring Strategy for Uneven-Length Batch Processes , 2004 .

[15]  Furong Gao,et al.  A Stage-Based Monitoring Method for Batch Processes with Limited Reference Data , 2004 .

[16]  Fuli Wang,et al.  Sub-PCA Modeling and On-line Monitoring Strategy for Batch Processes (R&D Note) , 2004 .

[17]  Deoki N. Saraf,et al.  Steady-State Identification, Gross Error Detection, and Data Reconciliation for Industrial Process Units , 2004 .

[18]  Bala Srinivasan,et al.  IMPROVEMENT OF PROCESS OPERATION IN THE PRODUCTION OF SPECIALTY CHEMICALS , 2006 .

[19]  L. Eriksson.  Multi- and megavariate data analysis , 2006 .

[20]  Zhi-huan Song,et al.  Process Monitoring Based on Independent Component Analysis - Principal Component Analysis ( ICA - PCA ) and Similarity Factors , 2007 .

[21]  Alberto Ferrer,et al.  Using Unfold‐PCA for batch‐to‐batch start‐up process understanding and steady‐state identification in a sequencing batch reactor , 2008 .

[22]  Yuan Yao,et al.  Phase and transition based batch process modeling and online monitoring , 2009 .