Abstract

In the process monitoring procedure, Data-driven (statistical) methods usually rely on the process measurements. In most industrial process this measurements has a multi-scale substance in time and frequency. Therefore the statistical methods which are proper for one scale may not be able to detect events at several scales. A Multi-Scale Partial Least Squares (MSPLS) algorithm consists of Wavelet Transforms for extracting multi-scale nature of measurements and Partial Least Squares (PLS) as a popular technique of statistical monitoring methods. In this paper the MSPLS algorithm is applied for monitoring of the Tennessee Eastman Process as a benchmark. To show the advantages of MSPLS, its process monitoring performance is compared with the standard PLS and is proved that MSPLS can be a more efficient technique than standard PLS for fault detection in industrial processes.

References

• L.H Chaing,., E.L Russel, and R.D Braatz,.. Fault Detection and Diagnosis in Industrial Systems, Springer, London, 2001.
• H.W Lee, M.W Lee, and J.M. Park., Multi-scale extension of PLS for advanced on-line process monitoring, Chemometrics and Intelligent Laboratory Systems, 98:201-212, 2009.
• H.B. Aradye, B.R. Bakshi, R.A Strauss,.and J. F. Davis, Multi-scale SPC using wavelet, AIChEJ, 44(7):1596-1610, 2003.
• B. R. Bakshi, Multiscale PCA with application to multivariate statistical process monitoring. AIChEJ, 44: 1596–1610, 1998.
• D. J. H. Wilson and G. W. Irwin. PLS modelling and fault detection on the Tennessee Eastman benchmark, International Journal of Systems Science, 31(11): 1449 – 1457, 2000.
• H. Abdi, Partial Least Squares Regression. In Lewis-Beck, M., A. Bryman, & T. Futing, editors, Encyclopedia of Social Sciences Research Methods. Thousand Oaks: Sage, 2003.
• S. Yoon and J.F. MacGregor, principal component analysis of multi scale data for process monitoring. AIChEJ, 50(11):2892-2903, 2004
• S.G. Mallat, A theory for multiresolution signal decomposition: the wavelet representation. IEEE Trans. Pattn Anal. Mach. Intell., 11: 674-693, 1989.
• G. P Nason and T. Sapatinas, Wavelet packet transfer function modelling of nonstationary time series. Statist and Computing, 12: 45–56, 2002.
• J.J., Downs and E.F Vogel,. A plant-wide industrial process control problem. Computer and chemical engineering 17: 245-255, 1993.
• P.R. Lyman, Plant-wide control structure for Teenesse Eastman problem. Computers and chemical engineering,19:321-331, 1995.
• I., Hashimoto, M., Kano, and K. Nagao, A new method for process monitoring using principal component analysis, AIChE, Annual meeting, paper 224a, 1999.
• W.E Larimore and D.E. Seborg, Process monitoring and identification of dynamic systems using statistical techniques. Los Angeles, CA, USA, 1997.
• A. Maulud, D. Wang, and J.A. Romagnoli, Process Control 16: 671–683, 2006.
• A. Simoglou, E.B. Martin, and A.J. Morris, Comput. Chem. Eng. 26 909–920, 2002.
• X. Wang, U.K. Kruger and B. Lennox, Control Eng. Pract. 11 613–632, 2003.
• Y Zhang and C. Ma, Fault diagnosis of nonlinear processes using multi scale KPCA and multi scale KPLS, Chemical engineering science, 66:64-72, 2011
• Y.Zhang, S.J.Qin,. improved nonlinear fault detection technology and statistical analysis. AIChEJ, 45(12): 3270-3220, 2008.