Investigation of Dynamic Multivariate Chemical Process Monitoring

• SYSTEM ENGINEERING • Next Articles

Investigation of Dynamic Multivariate Chemical Process Monitoring

XIELei; ZHANG Jianming; WANG Shuqing

National Key Laboratory of Industrial Control Technology, Institute of Advanced Process Control, Zhejiang Uni-versity, Hangzhou 310027, China

Received:1900-01-01 Revised:1900-01-01 Online:2006-10-28 Published:2006-10-28
Contact:
谢磊

动态多变量过程监控研究

谢磊; 张建明; 王树青

National Key Laboratory of Industrial Control Technology, Institute of Advanced Process Control, Zhejiang Uni-versity, Hangzhou 310027, China

通讯作者: XIE Lei

Abstract

Abstract: Chemical process variables are always driven by random noise and disturbances. The closed-loop control yields process measurements that are auto and cross correlated. The influence of auto and cross correlations on statistical process control (SPC) is investigated in detail by Monte Carlo experiments. It is revealed that in the sense of average performance, the false alarms rates (FAR) of principal component analysis (PCA), dynamic PCA are not affected by the time-series structures of process variables. Nevertheless, non-independent identical distribution will cause the actual FAR to deviate from its theoretic value apparently and result in unexpected consecutive false alarms for normal operating process. Dynamic PCA and ARMA-PCA are demonstrated to be inefficient to remove the influences of auto and cross correlations. Subspace identification-based PCA (SI-PCA) is proposed to improve the monitoring of dynamic processes. Through state space modeling, SI-PCA can remove the auto and cross correlations efficiently and avoid consecutive false alarms. Synthetic Monte Carlo experiments and the application in Tennessee Eastman challenge process illustrate the advantages of the proposed approach.

Key words: multivariate statistical processes control, subspace identification, false alarms rate, dynamic processes

摘要： Chemical process variables are always driven by random noise and disturbances. The closed-loop control yields process measurements that are auto and cross correlated. The influence of auto and cross correlations on statistical process control (SPC) is investigated in detail by Monte Carlo experiments. It is revealed that in the sense of average performance, the false alarms rates (FAR) of principal component analysis (PCA), dynamic PCA are not affected by the time-series structures of process variables. Nevertheless, non-independent identical distribution will cause the actual FAR to deviate from its theoretic value apparently and result in unexpected consecutive false alarms for normal operating process. Dynamic PCA and ARMA-PCA are demonstrated to be inefficient to remove the influences of auto and cross correlations. Subspace identification-based PCA (SI-PCA) is proposed to improve the monitoring of dynamic processes. Through state space modeling, SI-PCA can remove the auto and cross correlations efficiently and avoid consecutive false alarms. Synthetic Monte Carlo experiments and the application in Tennessee Eastman challenge process illustrate the advantages of the proposed approach.

关键词: multivariate statistical processes control;subspace identification;false alarms rate;dynamic processes

XIELei, ZHANG Jianming, WANG Shuqing. Investigation of Dynamic Multivariate Chemical Process Monitoring[J]. , DOI: .

谢磊, 张建明, 王树青. 动态多变量过程监控研究[J]. , DOI: .

References

1    Nomikos, P., MacGregor, J.F., “Multivariate SPC charts for monitoring batch processes”, Technometrics, 37,   41—59(1995).
2    Wise, B., Gallagher, N.B., “The process chemometrics approach to process monitoring and fault detection”, J. Process Contr., 6, 329—348(1996).
3    Cinar, A., Ündey, C., “Statistical process and controller performance monitoring: A tutorial on current methods and future direction”, In: Proceedings of American Con-trol Conference, Ⅳ, 2625—2639(1999).
4    Xie, L., Zhang, J., Wang, S., “A robust statistical batch process monitoring framework and its application”, Chin. J. Chem. Eng., 12, 682—687(2004).
5    Zhao, L.J., Cai, T.Y., Wang, G., “Double moving win-dow MPCA for online adaptive batch monitoring”, Chin. J. Chem. Eng., 13, 649—655(2005).
6    Zhou, S., Xie, L., Wang, S., “On-line fault diagnosis in industrial processes using variable moving window and Hidden Markov Model”, Chin. J. Chem. Eng., 13, 388—395(2005)
7    Jackson, J.E., A User’s Guide to Principal Components, Wiley Inter-science, New York (1991).
8    Ku, W., Storer, R.H., Georgakis, C., “Disturbance rejec-tion and isolation by dynamic principal component analysis”, Chemometr. Intell. Lab., 30, 179—196(1995).
9    Negiz, A., Cinar, A., “Statistical monitoring of multi-variate dynamic processes with state space models”, AIChE J., 43, 2002—2020(1997).
10    Callao, M.P., Rius, A., “Time series: a complementary technique to control charts for monitoring analytical sys-tems”, Chemometr. Intell. Lab., 66, 79—87(2003).
11    Simoglou, A., Martin, E.B., Morris, A.J., “Statistical performance monitoring of dynamic multivariate proc-esses using state space modeling”, Comput. Chem. Eng., 26, 909—920(2002).
12    Chen, J., Liu, K., “On-line batch process monitoring us-ing dynamic PCA and dynamic PLS models”, Chem. Eng. Sci., 57, 63—75(2002).
13    Kruger, U., Zhou, Y., Irwin G.W., “Improved principal component monitoring of large scale processes”, J. Proc-ess Contr., 14, 879—888(2004).
14    Åström, K.J., Introduction to Stochastic Control Theory, Academic Press, New York (1970).
15    Ljung, L., System Identification-Theory for the User, Prentice Hall, Englewood Clifs (1999).
16    Ljung, L., System Identification Toolbox-for Use with MATLAB, Mathwork, Natick (2002).
17    George, B., Gwilym, M.J., Gregory, R., Time series analysis: forecasting & control, 3rd edition, Prentice Hall, Englewood Clifs (1994).
18    Van, Overschee, P., De, Moor, B., Subspace Identifica-tion for Linear Systems: Theory Implementation Appli-cations, Kluwer Academic Publishers, Boston-London- Dordrecht (1996).
19    Downs, J.J., Vogel, E.F., “A plant-wide industrial proc-ess control problem”, Comput. Chem. Eng., 17, 245—255(1993).
20    Chiang, L.H., Russell, E.L., Braatz, R.D., Fault Detection and Diagnosis in Industrial Systems, Springer-Verlag, Heidelberg (2000).
21    Kano, M., Nagao, K., Hasebe, S., Hashimoto, I., “Com-parison of statistical process monitoring method: appli-cation to the Eastman challenge problem”, Comput. Chem. Eng., 24, 175—181(2000).
22    Ricker, N.L., “Decentralized control of the Tennessee Eastman challenge process”, J. Process Contr., 6, 205—221(1996).
23    Russell, E.L., Chiang, L.H., Braatz, R.D., “Fault detec-tion in industrial processes using canonical variate analysis and dynamic principal component analysis”, Chemometr. Intell. Lab., 51, 81—93(2000).