Multivariate
Statistical Process Control (MSPC)
Barry Lennox
There has been enormous
interest in using MSPC techniques, such as Principal Component Analysis (PCA)
and Partial Least Squares (PLS), to detect and diagnose abnormal conditions in
process systems. Work at Manchester
in this field began in the mid-1990s when it was supported by Invensys. In 2002
researchers from the University
of Manchester and
engineers from Invensys formed the spin-out company, Perceptive Engineering, which now has
control and monitoring applications
throughout the world.
The research at the University
of Manchester has focused
on the application of MSPC techniques to large-scale continuous and batch
industrial processes. A selection of the case-studies that have been
investigated are described in the papers below:
Batch applications:
Marjanovic, O, Lennox, B., Sandoz, D and Smith, K. (2006), ‘Real-time monitoring of an industrial batch process’,
Computers and Chemical Engineering,
30 (10-12), 1476-1481
Zhang, H. and Lennox, B., (2003), ‘Integrated
condition monitoring and control of fed-batch fermentation processes’, Journal of Process Control,
14 (1), 41-50
Lennox, B., Kipling, K., Glassey, J., and Montague, G., (2002), ‘Automated production support for the bioprocess industry’,
Biotechnology Progress, 18 (2), 269-275
Lennox,
B., Hiden, H.G., Montague, G.A., Kornfeld, G. and Goulding, P.R., (2001), ‘Process monitoring of an industrial fed-batch fermentation’,
Biotechnology and Bioengineering, 74 (2), 125-135
Refining:
Al-Ghazzawi, A and Lennox,
B., (2008), ‘Condition monitoring of a complex refining
process using multivariate statistical process control methods’, Control Engineering Practice, 16 (3),
294-307
Al-Ghazzawi, A and Lennox, B., (2009), ‘Model
predictive control monitoring using multivariate statistics’, Journal of Process Control, 19 (2),
314-327
Monitoring non-linear and time-varying processes:
Al-Ghazzawi, A and Lennox,
B., (2008), ‘Condition monitoring of a complex refining
process using multivariate statistical process control methods’, Control Engineering Practice, 16 (3),
294-307
Wang, X., Kruger, U. and Lennox,
B., (2003), ‘Recursive partial least squares algorithms
for monitoring complex industrial processes’, Control Engineering
Practice, 11 (6), 613-632
Monitoring the performance of model
predictive control system:
Al-Ghazzawi, A and Lennox, B., (2009), ‘Model
predictive control monitoring using multivariate statistics’, Journal of Process Control, 19 (2),
314-327
Applications in the steel industry:
Sandberg, E., Lennox, B. and Undvall, P., (2007), ‘Scrap
management by statistical evaluation of EAF process data’, Control Engineering Practice, 15(9),
1063-1075
Lennox, B., Sandberg, E. and Marjanovic, O., (2003), ‘Recent experiences in the application of PLS in the
steel industry’, PLS’03, Lisbon
Back to Home Page