Full Text:   <1086>

Summary:  <560>

CLC number: TP277

On-line Access: 2015-08-04

Received: 2014-07-20

Revision Accepted: 2015-05-03

Crosschecked: 2015-07-08

Cited: 0

Clicked: 9589

Citations:  Bibtex RefMan EndNote GB/T7714

 ORCID:

Shi-jin Ren

http://orcid.org/0000-0002-8321-1879

-   Go to

Article info.
Open peer comments

Frontiers of Information Technology & Electronic Engineering  2015 Vol.16 No.8 P.617-633

http://doi.org/10.1631/FITEE.1400263


A novel multimode process monitoring method integrating LDRSKM with Bayesian inference


Author(s):  Shi-jin Ren, Yin Liang, Xiang-jun Zhao, Mao-yun Yang

Affiliation(s):  1National Laboratory of Industrial Control Technology, Zhejiang University, Hangzhou 310027, China; more

Corresponding email(s):   sjren_phd@163.com

Key Words:  Multimode process monitoring, Local discriminant regularized soft k-means clustering, Kernel support vector data description, Bayesian inference, Tennessee Eastman process


Share this article to: More |Next Article >>>

Shi-jin Ren, Yin Liang, Xiang-jun Zhao, Mao-yun Yang. A novel multimode process monitoring method integrating LDRSKM with Bayesian inference[J]. Frontiers of Information Technology & Electronic Engineering, 2015, 16(8): 617-633.

@article{title="A novel multimode process monitoring method integrating LDRSKM with Bayesian inference",
author="Shi-jin Ren, Yin Liang, Xiang-jun Zhao, Mao-yun Yang",
journal="Frontiers of Information Technology & Electronic Engineering",
volume="16",
number="8",
pages="617-633",
year="2015",
publisher="Zhejiang University Press & Springer",
doi="10.1631/FITEE.1400263"
}

%0 Journal Article
%T A novel multimode process monitoring method integrating LDRSKM with Bayesian inference
%A Shi-jin Ren
%A Yin Liang
%A Xiang-jun Zhao
%A Mao-yun Yang
%J Frontiers of Information Technology & Electronic Engineering
%V 16
%N 8
%P 617-633
%@ 2095-9184
%D 2015
%I Zhejiang University Press & Springer
%DOI 10.1631/FITEE.1400263

TY - JOUR
T1 - A novel multimode process monitoring method integrating LDRSKM with Bayesian inference
A1 - Shi-jin Ren
A1 - Yin Liang
A1 - Xiang-jun Zhao
A1 - Mao-yun Yang
J0 - Frontiers of Information Technology & Electronic Engineering
VL - 16
IS - 8
SP - 617
EP - 633
%@ 2095-9184
Y1 - 2015
PB - Zhejiang University Press & Springer
ER -
DOI - 10.1631/FITEE.1400263


Abstract: 
A local discriminant regularized soft k-means (LDRSKM) method with bayesian inference is proposed for multimode process monitoring. LDRSKM extends the regularized soft k-means algorithm by exploiting the local and non-local geometric information of the data and generalized linear discriminant analysis to provide a better and more meaningful data partition. LDRSKM can perform clustering and subspace selection simultaneously, enhancing the separability of data residing in different clusters. With the data partition obtained, kernel support vector data description (KSVDD) is used to establish the monitoring statistics and control limits. Two bayesian inference based global fault detection indicators are then developed using the local monitoring results associated with principal and residual subspaces. Based on clustering analysis, bayesian inference and manifold learning methods, the within and cross-mode correlations, and local geometric information can be exploited to enhance monitoring performances for nonlinear and non-Gaussian processes. The effectiveness and efficiency of the proposed method are evaluated using the Tennessee Eastman benchmark process.

The manuscript considers a very complete combination of machine learning algorithms for multimode process monitoring, including the modified clustering (LDRSKM), dimension reduction (generalized LDA method), construction of statistics (SVDD and Bayesian method). The aspects mentioned here are such a large collection that any individual analysis for the ultimate monitoring performance is too difficult to make. This is a credible and strong presentation and deserves publication.

一种融合贝叶斯推理与LDRSKM的多模态过程监测算法

目的:针对复杂多模态工业过程故障监测存在的问题,充分利用过程数据的局部和非局部几何信息,同时实现数据聚类和低维子空间选择,提高非线性、非高斯多模态过程监测性能。
创新点:提出融合局部鉴别正则化软k-均值与贝叶斯推理的多模态过程监测算法。该方法充分利用过程数据的局部和非局部几何信息,较好地发挥了无监督学习和有监督学习的优点,提高了模态数据的分离性和解释性,监测性能良好。
方法:该方法分为二个阶段:第一阶段,首先,考虑过程数据的局部和非局部几何信息,提出一种局部保持的正则化软k-均值聚类算法(LPRSKM)。然后,建立有监督学习与无监督学习的统一框架,提出融合LPRSKM与广义线性鉴别分析算法(GELDA)的局部鉴别正则化软k-均值算法(LDRSKM)(图1)。第二阶段,使用核支持向量数据描述(KSVDD)对各局部子空间建立监测统计量及控制限。然后,基于贝叶斯推理方法建立多模态过程全局监测统计量。最后,在TE仿真平台对所提方法进行仿真分析。
结论:针对非线性、非高斯的多模态过程监测,提出一种新的数据划分和最优低维子空间选择的迭代算法来提高不同模态数据的分离效果。在此基础上使用KSVDD和贝叶斯推理方法,较好地解决了多个非高斯和非线性的过程模态的监测准确性和可靠性问题。

关键词:多模态过程监测;局部鉴别正则化软k-均值;核支持向量数据描述;贝叶斯推理

Darkslateblue:Affiliate; Royal Blue:Author; Turquoise:Article

Reference

[1]Cai, L.F., Tian, X.M., Zhang, N., 2014. A kernel time structure independent component analysis method for nonlinear process monitoring. Chin. J. Chem. Eng., 22(11-12):1243-1253.

[2]Chiang, L.H., Russell, E.L., Braatz, R.D., 2000. Fault diagnosis in chemical processes using Fisher discriminant analysis, discriminant partial least squares, and principal component analysis. Chemometr. Intell. Lab. Syst., 50(2):243-252.

[3]Deng, X.G., Tian, X.M., 2013. Sparse kernel locality preserving projection and its application in nonlinear process fault detection. Chin. J. Chem. Eng., 21(2):163-170.

[4]Deng, X.G., Tian, X.M., Chen, S., 2013. Modified kernel principal component analysis based on local structure analysis and its application to nonlinear process fault diagnosis. Chemometr. Intell. Lab. Syst., 127:195-209.

[5]Dong, W.W., Yao, Y., Gao, F.R., 2012. Phase analysis and identification method for multiphase batch processes with partitioning multi-way principal component analysis (MPCA) model. Chin. J. Chem. Eng., 20(6):1121-1127.

[6]Downs, J.J., Vogel, E.F., 1993. A plant-wide industrial process control problem. Comput. Chem. Eng., 17(3):245-255.

[7]Feital, T., Kruger, U., Dutra, J., et al., 2013. Modeling and performance monitoring of multivariate multimodal processes. AIChE J., 59(5):1557-1569.

[8]Ge, Z.Q., Song, Z.H., 2010. Maximum-likelihood mixture factor analysis model and its application for process monitoring. Chemometr. Intell. Lab. Syst., 102(1):53-61.

[9]Ge, Z.Q., Song, Z.H., 2012. A distribution-free method for process monitoring. Exp. Syst. Appl., 38(8):9812-9829.

[10]Ge, Z.Q., Zhang, M.G., Song, Z.H., 2010. Nonlinear process monitoring based on linear subspace and Bayesian inference. J. Process Contr., 20(5):676-688.

[11]Ge, Z.Q., Song, Z.H., Gao, F.R., 2013. Review of recent research on data-based process monitoring. Ind. Eng. Chem. Res., 52(10):3543-3562.

[12]Ghosh, K., Ramteke, M., Srinivasan, R., 2014. Optimal variable selection for effective statistical process monitoring. Comput. Chem. Eng., 60:260-276.

[13]He, X.F., Cai, D., Shao, Y.L., et al., 2011. Laplacian regularized Gaussian mixture model for data clustering. IEEE Trans. Knowl. Data Eng., 23(9):1406-1418.

[14]Howland, P., Wang, J., Park, H., 2006. Solving the small sample size problem in face recognition using generalized discriminant analysis. Patt. Recog., 39(2):277-287.

[15]Jing, L.P., Ng, M.K., Huang, J.Z., 2007. An entropy weighting k-means algorithm for subspace clustering of high-dimensional sparse data. IEEE Trans. Knowl. Data Eng., 19(8):1026-1041.

[16]Kano, M., Nagao, K., Hasebe, S., et al., 2002. Comparison of multivariate statistical process monitoring methods with applications to the Eastman challenge problem. Comput. Chem. Eng., 26(2):161-174.

[17]Kano, M., Fujioka, T., Tonomura, O., et al., 2007. Data-based and model-based blockage diagnosis for stacked microchemical processes. Chem. Eng. Sci., 62(4):1073-1080.

[18]Lee, D., Lee, J., 2007. Domain described support vector classifier for multi-classification problems. Patt. Recog., 40(1):41-51.

[19]Lee, J., Kang, B., Kang, S., 2011. Integrating independent component analysis and local outlier factor for plant-wide process monitoring. J. Process Contr., 21(7):1011-1021.

[20]Liu, J.L., Cai, D., He, X.F., 2010. Gaussian mixture model with local consistency. Proc. 24th AAAI Conf. on Artificial Intelligence, p.512-517.

[21]Miao, A.M., Ge, Z.Q., Song, Z.H., et al., 2015. Nonlocal structure constrained neighborhood preserving embedding model and its application for fault detection. Chemometr. Intell. Lab. Syst., 142:184-196.

[22]Miyamoto, S., Mukaidono, M., 1997. Fuzzy C-means as a regularization and maximum entropy approach. Proc. IFSA, p.86-92.

[23]Molina, G.D., Zumoffen, D.A.R., Basualdo, M.S., 2011. Plant-wide control strategy applied to the Tennessee Eastman process at two operating points. Comput. Chem. Eng., 35(10):2081-2097.

[24]Ng, Y.S., Srinivasan, R., 2009. An adjoined multi-model approach for monitoring batch and transient operations. Comput. Chem. Eng., 33(4):887-902.

[25]Perez, C.F.A., 2011. Fault Diagnosis with Reconstruction-Based Contributions for Statistical Process Monitoring. PhD Thesis, University of Southern California, USA.

[26]Serradilla, J., Shi, J.Q., Morris, A.J., 2011. Fault detection based on Gaussian process latent variable models. Chemometr. Intell. Lab. Syst., 109(1):9-21.

[27]Shen, J.F., Bu, J.J., Ju, B., et al., 2012. Refining Gaussian mixture model based on enhanced manifold learning. Neurocomputing, 87:19-25.

[28]Song, B., Ma, Y.X., Shi, H.B., 2014. Multimode process monitoring using improved dynamic neighborhood preserving embedding. Chemometr. Intell. Lab. Syst., 135:17-30.

[29]Tan, S.C., Lim, C.P., Rao, M.V.C., 2007. A hybrid neural network model for rule generation and its application to process fault detection and diagnosis. Eng. Appl. Artif. Intell., 20(2):203-213.

[30]Teppola, P., Mujunen, S.P., Minkkinen, P., 1999. Adaptive fuzzy C-means clustering in process monitoring. Chemometr. Intell. Lab. Syst., 45(1-2):23-38.

[31]Tong, C.D., Palazoglu, A., Yan, X.F., 2013. An adaptive multimode process monitoring strategy based on mode clustering and mode unfolding. J. Process Contr., 23(10):1497-1507.

[32]Venkatasubramanian, V., Rengaswamy, R., Yin, K., et al., 2003. A review of process fault detection and diagnosis: Part I: quantitative model-based methods. Comput. Chem. Eng., 27(3):293-311.

[33]Xie, L., Liu, X.Q., Zhang, J.M., et al., 2009. Non-Gaussian process monitoring based on NGPP-SVDD. Acta Autom. Sin., 35(1):107-112 (in Chinese).

[34]Xie, X., Shi, H.B., 2012. Multimode process monitoring based on fuzzy C-means in locality preserving projection subspace. Chin. J. Chem. Eng., 20(6):1174-1179.

[35]Xu, X., Xie, L., Wang, S., 2011. Multi-mode process monitoring method based on PCA mixture model. CIESC J., 62(3):743-752 (in Chinese).

[36]Yang, Y.H., Li, X., Liu, X.Z., et al., 2015. Wavelet kernel entropy component analysis with application to industrial process monitoring. Neurocomputing, 147:395-402.

[37]Yin, X.S., Chen, S.C., Hu, E.L., 2013. Regularized soft K-means for discriminant analysis. Neurocomputing, 103:29-42.

[38]Yu, J., 2012. A nonlinear kernel Gaussian mixture model based inferential monitoring approach for fault detection and diagnosis of chemical processes. Chem. Eng. Sci., 68(1):506-519.

[39]Zang, X., Vista Iv, F.P., Chong, K.T., 2014. Fast global kernel fuzzy c-means clustering algorithm for consonant/vowel segmentation of speech signal. J. Zhejiang Univ.-Sci. C (Comput. & Electron.), 15(7):551-563.

[40]Zhang, M., Ge, Z.Q., Song, Z.H., et al., 2011. Global-local structure analysis model and its application for fault detection and identification. Ind. Eng. Chem. Res., 50(11):6837-6848.

[41]Zhang, S.J., Wang, Z.L., Qian, F., 2010. FS-SVDD based on LTSA and its application to chemical process monitoring. CIESC J., 61(8):1894-1900 (in Chinese).

[42]Zhang, Y.W., 2009. Enhanced statistical analysis of nonlinear processes using KPCA, KICA and SVM. Chem. Eng. Sci., 64(5):801-811.

[43]Zhang, Y.W., Li, S., 2014. Modeling and monitoring of nonlinear multi-mode processes. Contr. Eng. Pract., 22:194-204.

[44]Zhang, Y.W., An, J.Y., Li, Z.M., et al., 2013. Modeling and monitoring for handling nonlinear dynamic processes. Inform. Sci., 235:97-105.

[45]Zhu, Z.B., Wang, P.L., Song, Z.H., 2010. PCA-SVDD based fault detection and self-learning identification. J. Zhejiang Univ. (Eng. Sci.), 44(4):652-658 (in Chinese).

[46]Zhu, Z.B., Song, Z.H., Palazoglu, A., 2012. Process pattern construction and multi-mode monitoring. J. Process Contr., 22(1):247-262.

Open peer comments: Debate/Discuss/Question/Opinion

<1>

Please provide your name, email address and a comment





Journal of Zhejiang University-SCIENCE, 38 Zheda Road, Hangzhou 310027, China
Tel: +86-571-87952783; E-mail: cjzhang@zju.edu.cn
Copyright © 2000 - Journal of Zhejiang University-SCIENCE