Mixture Bayesian Regularization Method of PPCA for Multimode Process Monitoring

被引:172
作者
Ge, Zhiqiang [1 ]
Song, Zhihuan [1 ]
机构
[1] Zhejiang Univ, Inst Ind Proc Control, State Key Lab Ind Control Technol, Hangzhou 310027, Zhejiang, Peoples R China
基金
中国国家自然科学基金;
关键词
multimode process monitoring; Bayesian regularization; principal component analysis; model localization; PRINCIPAL COMPONENT ANALYSIS; PCA;
D O I
10.1002/aic.12200
中图分类号
TQ [化学工业];
学科分类号
0817 ;
摘要
This article intends to address two drawbacks of the traditional principal component analysis (PCA)-based monitoring method: (1) nonprobabilistic; (2) single operation mode assumption. On the basis of the monitoring framework of probabilistic PCA (PPCA), a Bayesian regularization method is introduced for performance improvement, through which the effective dimensionality of the latent variable can be determined automatically. For monitoring processes with multiple operation modes, the Bayesian regularization method is extended to its mixture form, thus a mixture Bayesian regularization method of PPCA has been developed. To enhance the monitoring performance, a novel probabilistic strategy has been proposed for result combination in different operation modes. In addition, a new mode localization approach has also been developed, which can provide additional information and improve process comprehension for the operation engineer. A numerical example and a real industrial application case study have been used to evaluate the efficiency of the proposed method. (C) 2010 American Institute of Chemical Engineers AIChE J, 56: 2838-2849, 2010
引用
收藏
页码:2838 / 2849
页数:12
相关论文
共 16 条
[1]   Monitoring a complex refining process using multivariate statistics [J].
AlGhazzawi, Ashraf ;
Lennox, Barry .
CONTROL ENGINEERING PRACTICE, 2008, 16 (03) :294-307
[2]  
Bishop C., 2006, PATTERN RECOGN, V1st, P559
[3]  
Bishop CM, 1999, ADV NEUR IN, V11, P382
[4]   Mixture principal component analysis models for process monitoring [J].
Chen, JH ;
Liu, JL .
INDUSTRIAL & ENGINEERING CHEMISTRY RESEARCH, 1999, 38 (04) :1478-1488
[5]   Using mixture principal component analysis networks to extract fuzzy rules from data [J].
Chen, JH ;
Liu, JL .
INDUSTRIAL & ENGINEERING CHEMISTRY RESEARCH, 2000, 39 (07) :2355-2367
[6]   Fault detection based on a maximum-likelihood principal component analysis (PCA) mixture [J].
Choi, SW ;
Martin, EB ;
Morris, AJ ;
Lee, IB .
INDUSTRIAL & ENGINEERING CHEMISTRY RESEARCH, 2005, 44 (07) :2316-2327
[7]   Process monitoring using a Gaussian mixture model via principal component analysis and discriminant analysis [J].
Choi, SW ;
Park, JH ;
Lee, IB .
COMPUTERS & CHEMICAL ENGINEERING, 2004, 28 (08) :1377-1387
[8]   Process monitoring based on probabilistic PCA [J].
Kim, DS ;
Lee, IB .
CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS, 2003, 67 (02) :109-123
[9]   Diagnosis of process faults in chemical systems using a local partial least squares approach [J].
Kruger, Uwe ;
Dimitriadis, Grigorios .
AICHE JOURNAL, 2008, 54 (10) :2581-2596
[10]   Bayesian principal component analysis [J].
Nounou, MN ;
Bakshi, BR ;
Goel, PK ;
Shen, XT .
JOURNAL OF CHEMOMETRICS, 2002, 16 (11) :576-595