多变量统计过程监控:进展及其在化学工业的应用

摘要/Abstract

摘要： Multivariate statistical process monitoring and control (MSPM&C) methods for chemical
process monitoring with statistical projection techniques such as principal component
analysis (PCA) and partial least squares (PLS) are surveyed in this paper. The four-step
procedure of performing MSPM&C for chemical process, modeling of processes, detecting
abnormal events or faults, identifying the variable(s) responsible for the faults and
diagnosing the source cause for the abnormal behavior, is analyzed. Several main research
directions of MSPM&C reported in the literature are discussed, such as multi-way principal
component analysis (MPCA) for batch process, statistical monitoring and control for
nonlinear process, dynamic PCA and dynamic PLS, and on-line quality control by inferential
models. Industrial applications of MSPM&C to several typical chemical processes, such as
chemical reactor, distillation column, polymerization process, petroleum refinery units,
are summarized. Finally, some concluding remarks and future considerations are made.

关键词: multivariate statistical process monitoring and control (MSPM&C);fault detection and isolation (FDI);principal component analysis (PCA);partial least squares (PLS);quality control;inferential mode

Abstract: Multivariate statistical process monitoring and control (MSPM&C) methods for chemical
process monitoring with statistical projection techniques such as principal component
analysis (PCA) and partial least squares (PLS) are surveyed in this paper. The four-step
procedure of performing MSPM&C for chemical process, modeling of processes, detecting
abnormal events or faults, identifying the variable(s) responsible for the faults and
diagnosing the source cause for the abnormal behavior, is analyzed. Several main research
directions of MSPM&C reported in the literature are discussed, such as multi-way principal
component analysis (MPCA) for batch process, statistical monitoring and control for
nonlinear process, dynamic PCA and dynamic PLS, and on-line quality control by inferential
models. Industrial applications of MSPM&C to several typical chemical processes, such as
chemical reactor, distillation column, polymerization process, petroleum refinery units,
are summarized. Finally, some concluding remarks and future considerations are made.

Key words: multivariate statistical process monitoring and control (MSPM&, C), fault detection and isolation (FDI), principal component analysis (PCA), partial least squares (PLS), quality control

梁军, 钱积新. 多变量统计过程监控:进展及其在化学工业的应用[J]. CIESC Journal.

LIANG Jun, QIAN Jixin. Multivariate Statistical Process Monitoring and Control: Recent Developments and
Applications to Chemical Industry[J]. .

参考文献

[1] Martin, E.B., Morris, A.J., Zhang, J., "Process performance monitoring using
multivariate statistical process control", IEE Proc. Control Theory and Appl., 143, 132-
144(1996).
[2] Wise, B.M., Gallagher, N.B., "The process chemometrics approach to process monitoring
and fault detection", J.Process Control, 6, 329-348 (1996).
[3] MacGregor, J.F., Kourti, T., Nomikos, P., "Analysis, monitoring and fault diagnosis of
industrial processes using multivariate statistical projection methods", In: Proceedings of
13th IFAC World Congress Elsevier Science Ltd., San Francisco, 785-796 (1996).
[4] MacGregor, J.F., Kourti, T., "Statistical process control of multivariate process",
Control Eng. Practice, 3, 403-414 (1995).
[5] Frank, P.M., "Fault diagnosis in dynamic systems using analytical and knowledge-based
redundancy-a survey and some new results", Automatica, 26 (3), 459-474 (1990).
[6] Yoon, S., MacGregor, J.F., "Statistical and causal modelbased approaches to fault
detection and isolation", AIChE J., 46, 1813-1824 (2000).
[7] Zhang, J., Martin, E.B., Morris, A.J., "Process monitoring using nonlinear statistical
techniques", Chem. Eng. J., 67,181-189 (1997).
[8] Santen, A., Koot, G.L.M., Zullo, L.C., "Statistical data analysis of a chemical plant",
Computers Chem. Eng., 21(suppl.), 1123-1129 (1997).
[9] Baffi, G., Martin, E.B., Morari, A.J., "Nonlinear PLS revisited: The quadratic PLS
algorithm", Computers Chem. Eng., 23,395-411 (1999).
[10] Dong, D., McAvoy, T.J., "Nonlinear principal component analysis-based on principal
curves and neural network",Computers Chem. Eng., 20, 65-78 (1996).
[11] Kaspar, M.H., Ray, W.H., "Dynamic PLS modeling for process control", Chem. Eng. Sci.,
48 (20), 3447-3461 (1993).
[12] Kosanovich, K.A., Dahl, K.S., Piovoso, M.J., "Improved process understanding using
multi-way principal component analysis", Indus. Eng. Chem. Res., 35, 138-146(1996).
[13] Nomikos, P., "Detection and diagnosis of abnormal batch operations based on multi-way
PCA", ISA Trans., 35,259-266 (1996).
[14] MacGregor, J.F., Jaeckle, C., Kiparissides, C., Koutoudi, M., "Process monitoring and
diagnosis by multi-block PLS methods", AIChE J., 40, 826-838 (1994).
[15] Negiz, A., Cinar, A., "Monitoring of multivariable dynamic process and sensor", J.
Process Control, 8, 375-380 (1998).
[16] Qin, S.J., McAvoy, T.J., "Nonlinear PLS modeling using neural networks", Computers
Chem. Eng., 16, 379-391(1992).
[17] Wold, S., "Exponentially weighted moving PCA and PLS",Chemometrics Intel. Lab.
Systems, 23, 149 (1994).
[18] Piovoso, M.J., Kosanovich, K.A., "Applications of multivariate statistical methods to
process monitoring and controller design", International J. Control, 59, 743-765(1994).
[19] Doymaz, F.J., Romagnoli, A., Polazoglu, A., "A strategy for detection and isolation of
sensor failures and process upsets", Chemometrics Intel Lab. Systems, 55, 109-123(2001).
[20] Boque, R., Smilde, A.K., "Monitoring and diagnosing batch processes with multi-way
covariates regression models",AIChE J., 45, 1504-1520 (1999).
[21] Chen, J., Liu, J., "Derivation of function space analysis based PCA control charts for
batch process monitoring",Chem. Eng. Sci., 56, 3289-3304 (2001).
[22] Gallagher, N.B., Wise, B.M., Steuart, C.W., "Application of multi-way principal
components analysis to nuclear waste storage tank monitoring", Computers Chem. Eng., 20
(suppt.), 739-744 (1996).
[23] Henrion, R., Andersson, C.A., "A new criterion for simplestructure transformations of
core arrays in N-way principal components analysis", Chemometrics Intel Lab. Systems, 47,
189-204 (1999).
[24] Thornhill, N.F., Hutchison, R.J., Haughie, B.T., "A probability chart for statistical
process analysis with application",J. Process Control, 6, 161-167 (1996).
[25] Nomikos, P., MacGregor, J.F., "Monitoring batch processes using multi-way PCA", AIChE
J., 40, 1361-1375 (1994).
[26] Wold, S., Kettaneh-Wold, N., Skagerberg, B., "Nonlinear PLS modeling", Chemometrics
Intel. Lab. Systems, 7,53-65 (1989).
[27] Nomikos, P., MacGregor, J.F., "Multivariate SPC charts for monitoring batch process",
Technometrics, 37, 41-59(1995).
[28] Sanchez, E., Kowalski, B.R., "Tensorial resolution: A direct trilinear decomposition",
J. Chemometrics, 4, 29-45(1990).
[29] Smilde, A.K., Doornbos, D.A., "Three way methods for the calibration of
chromatographic systems: Comparing PARAFAC and three-way PLS", J. Chemometrics, 5, 345-360
(1991).
[30] Smilde, A.K., Wang, Y., Kowalski, B.R., "Theory of medium-rank second-order
calibration with restrictedTucker models", J. Chemometrics, 8, 21-36 (1994).
[31] Rannar, S., MacGregor, J.F., Wold, S., "Adaptive batch monitoring using hierarchical
PCA", Chemometrics Intel. Lab. Systems, 41, 73-81 (1998).
[32] Chen, J., Liu, J., "Post analysis on different operating time process using
orthonormal function approximation and multi-way PCA", J. Process Control, 10,411-418
(2000).
[33] Louwerse, D.J., Smilde, A.K., "Multivariate statistical process control of batch
processes based on three-way models",Chem. Eng. Sci., 55, 1225-1235 (2000).
[34] Dahl, K.S., Piovoso, M.J., Kosanovich, K.A., "Translating third-order data analysis
methods to chemical batch process", Chemometrics Intel. Lab. Systems, 46, 161-180(1999).
[35] Wold, S., "Nonlinear PLS modeling, Ⅱ. Spline inner function", Chemometrics Intel.
Lab. Systems, 14, 71-84(1992).
[36] Fourie, S.H., deVaol, P., "Advances process monitoring using an on-line nonlinear
multiscale PCA methodology",Computers Chem. Eng., 24, 755-760 (2000).
[37] Hastie, T., Stuetzle, W., "Principal curves", J. Amer. Stat.Assoc., 84, 502-516
(1989).
[38] McLain, R.B., Henson, M.A., "Principal component analysis for nonlinear model
reference adaptive control", Computers Chem. Eng., 24, 99-110 (2000).
[39] McAvoy, T.J., Donat, J. S., Bhat, N., "Interpreting biosensor data via
backpropagation", In: Proceedings of International Joint Conference on Neural Network, IEEE
Press, Washington DC, Vol. I , 227-233 (1989).
[40] Frank, I.E., "A nonlinear PLS model", Chemometrics and Intelligent Laboratory Systems,
8, 109-119 (1990).
[41] Cybenko, G., "Approximation by superposition of a sigmoidal function", SIAM J. Sci.
Stat. Computing, 5, 175-191 (1989).
[42] Holcomb, T.R., Morari, M., "PLS/neural networks", Computers Chem. Eng., 16, 393-411
(1992).
[43] Malthouse, E.C., Tamhane, A.C., Mah, R.S.H., "Nonlinear partial least squares",
Computers Chem. Eng., 21, 875-890 (1997).
[44] Shao, R., Jia, F., Martin, E.B., Morris, A.J., "Wavelet and nonlinear PCA for process
monitoring", Control Eng. Practice, 7, 865-879 (1999).
[45] Eeczadi-Amoli, J., McDonald, R.P., "A second generation nonlinear factor analysis",
Psychiometrika, 48 (3), 315-327 (1983).
[46] Gertler, J., McAvoy, T.J., "Principal components analysis and parity relations a
strong duality", In: Preprints of IFAC Safe-process Symposium, Hull, UK, 2, 837-842 (1997).
[47] Huang, Y., McAvoy, T.J., Gertler, J., "Fault isolation in nonlinear systems with
structured partial PCA and clustering Analysis", The Canadian J. Chem. Eng., 78, 569-577
(2000).
[48] Reddy, N.R., Riley, P.M., Mavrovouniotis, M.L., "Analysis of plant measurements
through input-training neural network", Computers Chem. Eng., 20 (Suppl.), 889-894(1996).
[49] Kramer, M.A., "Nonlinear principal component analysis using autoassociative neural
network", AIChE J., 37, 233-243(1991).
[50] Doymaz, F., Chen, J., Romagnoli, J.A., Polazoglu, A., "A robust strategy of real time
process monitoring", J. Process Control, 11, 343-359 (2001).
[51] Gomm, J.B., Weerasinghe, M., Williams, D., "Diagnosis of process faults with neural
networks and principal component analysis", Proc. IME, 214, Pt.E, 131-143 (2000).
[52] Gnanadesikian, R., Methods for Statistical Data Analysis of Multivariate Observations,
John Wiley & Sons, New York(1977).
[53] Ku, W., Storer, R.H., Georgakis, C., "Disturbance detection and isolation by dynamic
principal component analysis", Chemometrics Intel. Lab. Systems, 30, 179-196(1995).
[54] Lakshaninarayanan, S., Shah, S.L., Nandakumar, K., "Modeling and control of
multivariable process: Dynamic PLS approach", AIChE J., 43, 2307-2322 (1997).
[55] Lin, W., Qian, Y., Li, X., "Nonlinear dynamic principal component analysis for on-line
process monitoring and diagnosis", Computers Chem. Eng., 24, 423-429 (2000).
[56] Negiz, A., Cinar, A., "Statistical monitory of multivariable dynamic process with
state-space model", AIChE J., 43,2002-2020 (1997).
[57] Hartnett, M.K., Lightbody, G., Irwin, G.W., "Dynamic inferential estimation using
principal component regression (PCR)", Chemometrics Intel. Lab. Systems, 40, 215-224(1998).
[58] Kano, M., Miyazaki, K., Hasebe, S., Hashimoto, I., "Inferential control system of
distillation compositions using dynamic PLS regression", J. Process Control, 10, 157-166
(2000).
[59] Russell, E.L., Chiang, L.H., Braatz, R.D., "Fault detection in industrial processes
using canonical variate analysis and dynamic principal component analysis", Chemometrics
Intel. Lab. Systems, 51, 81-93 (2000).
[60] Rickel N.L., "The use of biased least square estimators for parameters in discrete-
time pulse response models", Indus. Eng. Chem. Res., 27, 343-350 (1988).
[61] Qin, S.J., McAvoy, T.J., "A data-based process modeling approach and its
applications", In: Proceedings of3rd IFAC Symposium on Dynamics and Control of Chemical
Reactors, Distillation Columns and Batch Processing, Pergamon, Qxford, 341-350 (1992).
[62] Hartnett, M.K., Lightbody, G., Irwin, G.W., "Identification of state models using
principle components analysis",Chemometrics Intel. Lab. Systems, 46, 181-196 (1999).
[63] Wise, B.M., Ricker, N.L., Veltkamp, D.F., Kowalski, B.R., "A theoretical basis for the
use of principal component models for monitoring multivariate processes", Process Control
and Quality, 1, 41-51 (1990).
[64] Kaspar, M.H., Ray, W.H., "Chemometric methods for process monitoring and high-
performance controller design", AIChE J., 38, 1593-1608 (1992).
[65] Qin, S.J., McAvoy, T.J., "Nonlinear FIR modeling via a neural net PLS approach",
Computers Chem. Eng., 20,147-159 (1996).
[66] Russell, S.A., Kesavan, P., Lee, J.H., Ogunnaike, B.A., "Recursive data-based
prediction and control of batch product quality", AIChE J., 44, 2442-2458 (1998).
[67] Kresta, J., Marlin, T., MacGregor, J., "Choosing inferential variables for
multicomponent distillation using PLS",In: AIChE Annual Meeting, Chicago, (1990).
[68] MacGregor, J.F., Skagerberg, B., Kiparissides, C., "Multivariate SPC and property
inference applied to low density polyethylene reactor", In: Proceedings of IFAC Symp
ADCHEM’91, Toulouse, France, 663-670 (1991).
[69] Zhang, W., Yu, J., "MIMO soft sensors for estimating product quality with on-line
correction", Trans. IChemE, 78A,612-620 (2000).
[70] Qin, S.J., Yue, H., Dunia, R., "Self-validating inferential sensors with application
to air emission monitoring", Indus. Eng. Chem. Res., 36, 1675-1685 (1997).
[71] Wadi, I., "Neural network model predicts naphtha cut point", Oil & Gas J., 11, 67-70
(1996).
[72] Brambille, A., Trivella, F., "Estimate product quality with ANNs", Hydrocarbon
Processing, 75 (9), 61-66 (1996).
[73] Nijhuis, A., Tong, S., Vandeginste, B.G.M., "The application of multivariate quality
control in gas chromatography",Chemometrics Intel. Lab. Systems, 47, 107-125 (1999).
[74] Chen, G., McAvoy, T.J., "Process control utilizing data based multivariate statistical
models", The Canadian J. Chem. Eng., 74, 1010-1024 (1996).
[75] Chen, G., McAvoy, T.J., Piovoso, M.J., "A multivariate statistical controller for on-
line quality improvement", J. Process Control, 8, 139-149 (1998).
[76] Moore, C., Hackney, J., Canter, D., "Selecting sensor location and type for
multivariate processes", In: Shell Process Control Workshop, Burrerworths, Toronto (1986).
[77] MacGregor, J.F., Wong, A.K.L., "Multivariate model identification and stochastic
control of a chemical reactor",Technometrics, 22, 453-364 (1980).
[78] Mejdell, T., Skogestad, S., "Composition control of distillation columns using
multiple temperature measurements",In: Proceedings of AIChE Annual Meeting, Chicago, 917-
922 (1990).
[79] Kresta, J., Marlin, T.E., MacGregor, J.F., "Multivariate statistical monitoring of
process operating performance",The Canadian J. Chem. Eng., 69 35-47 (1990).
[80] Mejdell, T., Skogestad, S., "Estimation of distillation compositions from multiple
temperature measurements using PLS", Indus. Eng. Chem. Res., 30, 2543-2555 (1991).
[81] Mejdell, T., Skogestad, S., "Composition estimator in a pilot-plant distillation
column using multiple temperatures", Indus. Eng. Chem. Res., 30, 2555-2564 (1991).
[82] Skagerberg, B., Sjoestroem, M., Wold, S., "Multivariate data analysis applied to low
density polyethylene reactors",Chemometrics Intel. Lab. Systems, 14, 341-356 (1992).
[83] Kresta, J., Marlin, T.E., MacGregor, J.F., "Development of inferential process models
using PLS", Computers Chem.Eng., 18, 597-611 (1994).
[84] Yabuki, Y., MacGregor, J.F., "Product quality control in semibatch reactors using
midcourse corrections policies",In: proceedings of IFAC Symposium on Advanced Control of
Chemical Process, Elsevier Science Ltd., Banff, 189-194(1997).
[85] Dong, D., McAvoy, T.J., "Batch tracking via nonlinear PCA", AIChE J., 26, 2199-2208,
(1996).
[86] Simoglou, A., Martin, E.B., Morris, A.J., "Multivariate statistical process control of
an industrial fluidized-bed reactor", Control Eng. Practice, 8, 893-909 (2000).
[87] Ruiz, D., Nougues, J., M., Calderon, Z., Espuna, A., Puigianer, L., "Neural network
based framework for fault diagnosis in batch chemical plants", Computers Chem.Eng., 24,
777-784 (2000).
[88] Zhang, J., Martin, E.B., Morris, A.J., "Fault detection and diagnosis using
multivariate statistical techniques", Trans. IChemE, 74A, 89-96 (1996).
[89] Hiden, H.G., Wills, M.J., Tham, M.T., Montague, G.A., "Nonlinear principal components
analysis using genetic programming", Computers Chem. Eng., 23, 413-425 (1999).
[90] Skagerberg, B., Lehtinen, J., Wiklunel, J., "Multivariate statistical monitoring of
polypropylene production", Pla.st.Rubber Compos. Proce. Appl., 18, 299-305 (1992).
[91] Martin, E.B., Morris, A. J., "Non-parametric confidence bounds for process performance
monitoring charts", J. Process Control, 6, 349-358 (1996).
[92] Kourti, T., MacGregor, J.F., "Multivariate SPC methods for process and product
monitoring", J. Qual. Tech., 28,409-428 (1996).
[93] Kourti, T., Nomikos, P., MacGregor, J.F., "Analysis, monitoring and fault diagnosis of
batch processing multiblock and multi-way PLS", J. Process Control, 5, 277-284(1995).
[94] Chen, F.Z., Wang, X.Z., "Discovery of operational spaces from process data for
production of multiple grades of products", Indus. Eng. Chem. Res., 39, 2378-2383 (2000).
[95] Wold, S., Kettaneh, N., Tjessem, K., "Hierarchical multiblock PLS and PC models for
easier model interpretation and as an alternative to variable selection", J. Chemometrics,
10, 463-482 (1996).
[96] Vedam, H., Venkatasubramanian, V., "PCA-SDG based process monitoring and fault
diagnosis", Control Eng.Practice, 7, 903-917 (1999).
[97] Teppola, P., Mujunen, S.P., Minkkinen, P., "Adaptive fuzzy C-means clustering in
process monitoring", Chemometrics Intel. Lab. Systems, 45, 23-38 (1999).
[98] Rotem, Y., Wachs, A., Lewin, D.R., "Ethylence compressor monitoring using model-based
PCA", AIChE J., 46,1825-1836 (2000).
[99] Stork, A.L., Kowalski, B.R., "Distinguishing between process upsets and sensor
malfunctions using sensor redundancy", Chemometrics Intel Lab. Systems, 46, 117-131(1999).