A Kernel Time Structure Independent Component Analysis Method for Nonlinear Process Monitoring

doi:10.1016/j.cjche.2014.09.021

Abstract

Abstract: Kernel independent component analysis (KICA) is a newly emerging nonlinear process monitoring method, which can extractmutually independent latent variables called independent components (ICs) fromprocess variables. However, when more than one IC have Gaussian distribution, it cannot extract the IC feature effectively and thus its monitoring performance will be degraded drastically. To solve such a problem, a kernel time structure independent component analysis (KTSICA) method is proposed for monitoring nonlinear process in this paper. The original process data are mapped into a feature space nonlinearly and then the whitened data are calculated in the feature space by the kernel trick. Subsequently, a time structure independent component analysis algorithm, which has no requirement for the distribution of ICs, is proposed to extract the IC feature. Finally, two monitoring statistics are built to detect process faults. When some fault is detected, a nonlinear fault identification method is developed to identify fault variables based on sensitivity analysis. The proposed monitoring method is applied in the Tennessee Eastman benchmark process. Applications demonstrate the superiority of KTSICA over KICA.

Key words: Process monitoring, Independent component analysis, Kernel trick, Time structure, Fault identification

摘要： Kernel independent component analysis (KICA) is a newly emerging nonlinear process monitoring method, which can extractmutually independent latent variables called independent components (ICs) fromprocess variables. However, when more than one IC have Gaussian distribution, it cannot extract the IC feature effectively and thus its monitoring performance will be degraded drastically. To solve such a problem, a kernel time structure independent component analysis (KTSICA) method is proposed for monitoring nonlinear process in this paper. The original process data are mapped into a feature space nonlinearly and then the whitened data are calculated in the feature space by the kernel trick. Subsequently, a time structure independent component analysis algorithm, which has no requirement for the distribution of ICs, is proposed to extract the IC feature. Finally, two monitoring statistics are built to detect process faults. When some fault is detected, a nonlinear fault identification method is developed to identify fault variables based on sensitivity analysis. The proposed monitoring method is applied in the Tennessee Eastman benchmark process. Applications demonstrate the superiority of KTSICA over KICA.

关键词: Process monitoring, Independent component analysis, Kernel trick, Time structure, Fault identification

Lianfang Cai, Xuemin Tian, Ni Zhang. A Kernel Time Structure Independent Component Analysis Method for Nonlinear Process Monitoring[J]. , 2014, 22(11/12): 1243-1253.

References

[1] M.M. Rashid, J. Yu, Hidden Markov model based adaptive independent component analysis approach for complex chemical process monitoring and fault detection, Ind. Eng. Chem. Res. 51 (15) (2012) 5506-5514.
[2] L. Ye, X. Shi, J. Liang, A multi-level approach for complex fault isolation based on structured residuals, Chin. J. Chem. Eng. 19 (3) (2011) 462-472.
[3] Z.Q. Ge, M.G. Zhang, Z.H. Song, Nonlinear process monitoring based on linear subspace and Bayesian inference, J. Process Control 20 (5) (2010) 676-688.
[4] L.F. Cai, X.M. Tian, A new fault detection method for non-Gaussian process based on robust independent component analysis, Process. Saf. Environ. Protect. (2014), http://dx.doi.org/10.1016/j.psep.2013.11.003 (in press).
[5] Z.F. Wang, J.Q. Yuan, Online supervision of Penicillin cultivations based on rolling MPCA, Chin. J. Chem. Eng. 15 (1) (2007) 92-96.
[6] J.C. Wang, Y.B. Zhang, H. Cao, W.Z. Zhu, Dimension reduction method of independent component analysis for process monitoring based on minimum mean square error, J. Process Control 22 (2) (2012) 477-487.
[7] M.M. Rashid, J. Yu, A newdissimilaritymethod integrating multidimensionalmutual information and independent component analysis for non-Gaussian dynamic process monitoring, Chemom. Intell. Lab. Syst. 115 (15) (2012) 44-58.
[8] J. Lee, B. Kang, S.H. Kang, Integrating independent component analysis and local outlier factor for plant-wide process monitoring, J. Process Control 21 (7) (2011) 1011-1021.
[9] Y.W. Zhang, Y. Zhang, Fault detection of non-Gaussian processes based on modified independent component analysis, Chem. Eng. Sci. 65 (16) (2010) 4630-4639.
[10] L.F. Cai, X.M. Tian, S. Chen, A process monitoring method based on noisy independent component analysis, Neurocomputing 127 (2014) 231-246.
[11] J.M. Lee, C.K. Yoo, I.B. Lee, Statistical process monitoring with independent component analysis, J. Process Control 14 (5) (2004) 467-485.
[12] J.M. Lee, S.J. Qin, I.B. Lee, Fault detection of non-linear process using kernel independent component analysis, Can. J. Chem. Eng. 85 (4) (2007) 526-536.
[13] X.M. Tian, X.L. Zhang, X.G. Deng, S. Chen, Multiway kernel independent component analysis based on feature samples for batch process monitoring, Neurocomputing 72 (7) (2009) 1584-1596.
[14] Y.W. Zhang, J.Y. An, H.L. Zhang, Monitoring of time-varying processes using kernel independent component analysis, Chem. Eng. Sci. 88 (2013) 23-32.
[15] L.F. Cai, X.M. Tian, N. Zhang, Non-Gaussian process fault detection method based on modified KICA, CIESC J. 63 (9) (2012) 2864-2868 (in Chinese).
[16] R. Xing, S.Y. Zhuang, L. Xie, Nonlinear process monitoring based on Improve kernel ICA, Proceedings of the International Conference on Computational Intelligence and Security, Guangzhou, China, 2, 2006, pp. 1742-1746.
[17] L. Tong, R.W. Liu, V.C. Soon, Y.F. Huang, Indeterminacy and identifiability of blind identification, IEEE Trans. Circ. Syst. 38 (5) (1991) 499-509.
[18] W. Liu, D.P. Mandic, A. Cichocki, Blind source extraction based on a linear predictor, IET Signal Process. 1 (1) (2007) 29-34.
[19] L. Petzold, S. Li, Y. Cao, R. Serban, Sensitivity analysis of differential-algebraic equations and partial differential equations, Comput. Chem. Eng. 30 (10-12) (2006) 1553-1559.
[20] A. Hyvärinen, Fast and robust fixed-point algorithms for independent component analysis, IEEE Trans. Neural Netw. 10 (3) (1999) 626-634.
[21] Y. Hu, H.H. Ma, H.B. Shi, Enhanced batch process monitoring using just-in-timelearning based kernel partial least squares, Chemom. Intell. Lab. Syst. 123 (2013) 15-27.
[22] X.G. Deng, X.M. Tian, Sparse kernel locality preserving projection and its application in nonlinear process fault detection, Chin. J. Chem. Eng. 21 (2) (2013) 163-170.
[23] X.G. Deng, X.M. Tian, A new fault isolation method based on unified contribution plots, Proceedings of the 30th Chinese Control Conference, Yantai, China, 2011, pp. 4280-4285.
[24] W. Lu, J.C. Rajapakse, Approach and applications of constrained ICA, IEEE Trans. Neural Netw. 16 (1) (2005) 203-212.
[25] X.M. Tian, L.F. Cai, S. Chen, Noise-resistant joint diagonalization independent component analysis based process fault detection, Neurocomputing (2014), http://dx. doi.org/10.1016/j.neucom.2014.08.009 (in press).
[26] X.F. Cheng, Y.W. Tao, S.B. Zhang, X.J. Zhang, J. Liu, Applications of independent sub-band functions and wavelet analysis in single-channel noisy signal BSS: model and crucial technique, Acta Electron. Sin. 37 (7) (2009) 1522-1528 (in Chinese).
[27] Q.C. Jiang, X.F. Yan, Statistic monitoring of chemical processes based on sensitive kernel principal components, Chin. J. Chem. Eng. 21 (6) (2013) 633-643.
[28] L.Wang, H.B. Shi, Improved kernel PLS-based fault detection approach for nonlinear chemical processes, Chin. J. Chem. Eng. 22 (6) (2014) 657-663.
[29] X.Y. Chen, X.F. Yan, Fault diagnosis in chemical process based on self-organizing map integrated with Fisher discriminant analysis, Chin. J. Chem. Eng. 21 (4) (2013) 382-387.
[30] L.H. Chiang, E.L. Russell, R.D. Braatz, Fault Detection and Diagnosis in Industrial Systems, Springer Verlag, London, 2001.
[31] S. Mahadevan, S.L. Shah, Fault detection and diagnosis in process data using one-class support vector machines, J. Process Control 19 (10) (2009) 1627-1639.
[32] Y.W. Zhang, C.Ma, Fault diagnosis of nonlinear processes using multiscale KPCA and multiscale KPLS, Chem. Eng. Sci. 66 (1) (2011) 64-72.
[33] X. Liu, L. Xie, U. Kruger, T. Littler, S. Wang, Statistical-based monitoring of multivariate non-Gaussian systems, AIChE J. 54 (9) (2008) 2379-2391.