Local component based principal component analysis model for multimode process monitoring

doi:10.1016/j.cjche.2020.10.030

摘要/Abstract

摘要： For plant-wide processes with multiple operating conditions, the multimode feature imposes some chal lenges to conventional monitoring techniques. Hence, to solve this problem, this paper provides a novel local component based principal component analysis (LCPCA) approach for monitoring the status of a multimode process. In LCPCA, the process prior knowledge of mode division is not required and it purely based on the process data. Firstly, LCPCA divides the processes data into multiple local components using finite Gaussian mixture model mixture (FGMM). Then, calculating the posterior probability is applied to determine each sample belonging to which local component. After that, the local component information (such as mean and standard deviation) is used to standardize each sample of local component. Finally, the standardized samples of each local component are combined to train PCA monitoring model. Based on the PCA monitoring model, two monitoring statistics T² and SPE are used for monitoring multimode pro cesses. Through a numerical example and the Tennessee Eastman (TE) process, the monitoring result demonstrates that LCPCA outperformed conventional PCA and LNS-PCA in the fault detection rate.

关键词: Principal component analysis, Finite Gaussian mixture model, Process monitoring, Tennessee Eastman (TE) process

Abstract: For plant-wide processes with multiple operating conditions, the multimode feature imposes some chal lenges to conventional monitoring techniques. Hence, to solve this problem, this paper provides a novel local component based principal component analysis (LCPCA) approach for monitoring the status of a multimode process. In LCPCA, the process prior knowledge of mode division is not required and it purely based on the process data. Firstly, LCPCA divides the processes data into multiple local components using finite Gaussian mixture model mixture (FGMM). Then, calculating the posterior probability is applied to determine each sample belonging to which local component. After that, the local component information (such as mean and standard deviation) is used to standardize each sample of local component. Finally, the standardized samples of each local component are combined to train PCA monitoring model. Based on the PCA monitoring model, two monitoring statistics T² and SPE are used for monitoring multimode pro cesses. Through a numerical example and the Tennessee Eastman (TE) process, the monitoring result demonstrates that LCPCA outperformed conventional PCA and LNS-PCA in the fault detection rate.

Key words: Principal component analysis, Finite Gaussian mixture model, Process monitoring, Tennessee Eastman (TE) process

Yuan Li, Dongsheng Yang. Local component based principal component analysis model for multimode process monitoring[J]. 中国化学工程学报, 2021, 34(6): 116-124.

Yuan Li, Dongsheng Yang. Local component based principal component analysis model for multimode process monitoring[J]. Chinese Journal of Chemical Engineering, 2021, 34(6): 116-124.

参考文献

[1] X.M. Zhang, M. Kano, M. Tani, J. Mori, J. Ise, K. Harada, Prediction and causal analysis of defects in steel products:Handling nonnegative and highly overdispersed count data, Control Eng. Pract. 95(2020) 104258.
[2] Y. Zheng, H.J. Fang, H.O. Wang, Takagi-Sugeno fuzzy model-based fault detection for networked control systems with markov delays, IEEE Trans. Syst. Man and Cybernetics Part B Cybernetics 36(3) (2006) 924-929.
[3] X. Ma, Y.B. Si, Z.Y. Yuan, Y.H. Qin, Y.Q. Wang, Multi-step dynamic slow feature analysis for industrial process monitoring, IEEE Trans. Inst. Measur. 69(12) (2020) 9535-9548.
[4] Y.B. Si, Y.Q. Wang, D.H. Zhou, Key-performance-indicator-related process monitoring based on improved kernel partial least squares, IEEE Trans. Ind. Electron. 68(3) (2021) 2626-2636.
[5] Y. Tao, H.B. Shi, B. Song, S. Tan, A novel dynamic weight principal component analysis method and hierarchical monitoring strategy for process fault detection and diagnosis, IEEE Trans. Ind. Electron. 67(9) (2020) 7994-8004.
[6] K. Wang, J.H. Chen, Z.H. Song, Performance analysis of dynamic PCA for closedloop process monitoring and its improvement by output oversampling scheme, IEEE Trans. Control Syst. Technol. 27(1) (2019) 378-385.
[7] Y.W. Zhang, S. Li, Z.Y. Hu, Improved multi-scale kernel principal component analysis and its application for fault detection, Chem. Eng. Res. Des. 90(9) (2012) 1271-1280.
[8] J.M. Lee, C.K. Yoo, S.W. Choi, P.A. Vanrolleghem, I.B. Lee, Nonlinear process monitoring using kernel principal component analysis, Chem. Eng. Sci. 59(1) (2004) 223-234.
[9] P. Nomikos, J.F. MacGregor, Monitoring batch processes using multiway principal component analysis, AIChE J. 40(1994) 1361-1375.
[10] F. Pozo, Y. Vidal, O. Salgado, Wind turbine condition monitoring strategy through multiway PCA and multivariate inference, Energies 11(4) (2018) 749.
[11] Z.Q. Ge, Z.H. Song, F.R. Gao, Review of recent research on data-based process monitoring, Ind. Eng. Chem. Res. 52(10) (2013) 3543-3562.
[12] Y.Q. Chang, R.X. Ma, L.P. Zhao, F.L. Wang, S. Wang, Online operating performance evaluation for the plant-wide industrial process based on a three-level and multi-block method, Can. J. Chem. Eng. 97(S1) (2019) 1371-1385.
[13] A. Jaworski, K. Wikiel, H. Wikiel, Application of multiblock and hierarchical PCA and PLS models for analysis of AC voltammetric data, Electroanalysis 17(15-16) (2005) 1477-1485.
[14] Z.Q. Ge, Z.H. Song, Distributed PCA model for plant-wide process monitoring, Ind. Eng. Chem. Res. 52(5) (2013) 1947-1957.
[15] M. Rezamand, M. Kordestani, R. Carriveau, D.-S.-K. Ting, M. Saif, A new hybrid fault detection method for wind turbine blades using recursive PCA and wavelet-based PDF, IEEE Sens. J. 20(4) (2020) 2023-2033.
[16] C. Tong, A. Palazoglu, X.F. Yan, An adaptive multimode process monitoring strategy based on mode clustering and mode unfolding, J. Process Control 23(10) (2013) 1497-1507.
[17] X. Wang, U. Kruger, G.W. Irwin, G. McCullough, N. McDowell, Nonlinear PCA with the local approach for diesel engine fault detection and diagnosis, IEEE Trans. Control Syst. Technol. 16(1) (2007) 122-129.
[18] Z.C. Li, X.F. Yan, Ensemble learning model based on selected diverse principal component analysis models for process monitoring, J. Chemom. 32(4) (2018) e3010.
[19] J.F. MacGregor, C. Jaeckle, C. Kiparissides, M. Koutoudi, Process monitoring and diagnosis by multiblock PLS methods, AIChE J. 40(5) (1994) 826-838.
[20] S.J. Qin, S. Valle, M.J. Piovoso, On unifying multiblock analysis with application to decentralized process monitoring, J. Chemom. 15(9) (2001) 715-742.
[21] S.J. Zhao, J. Zhang, Y.M. Xu, Monitoring of processes with multiple operating modes through multiple principle component analysis models, Ind. Eng. Chem. Res. 43(22) (2004) 7025-7035.
[22] Q.C. Jiang, X.F. Yan, Monitoring multi-mode plant-wide processes by using mutual information-based multi-block PCA, joint probability, and Bayesian inference, Chemometrics and Intelligent Laboratory Syst. 136(2014) 121-137.
[23] Z.Q. Ge, Z.H. Song, Mixture Bayesian regularization method of PPCA for multimode process monitoring, AIChE J. 56(11) (2010) 2838-2849.
[24] J. Yu, S.J. Qin, Multimode process monitoring with Bayesian inference-based finite Gaussian mixture models, AIChE J. 54(7) (2008) 1811-1829.
[25] J. Yu, Fault detection using principal components-based gaussian mixture model for semiconductor manufacturing processes, IEEE Trans. Semicond. Manuf. 24(3) (2011) 432-444.
[26] J.L. Zhu, Z.Q. Ge, Z.H. Song, Distributed Gaussian mixture model for monitoring plant-wide processes with multiple operating modes, IFAC J. Syst. Control 6(2018) 1-15.
[27] Q.C. Jiang, X.F. Yan, Multimode process monitoring using variational bayesian inference and canonical correlation analysis, IEEE Trans. Autom. Sci. Eng. 16(4) (2019) 1814-1824.
[28] Q.P. He, J. Wang, Fault detection using the k-nearest neighbor rule for semiconductor manufacturing processes, IEEE Trans. Semicond. Manuf. 20(4) (2007) 345-354.
[29] Q.P. He, J. Wang, Principal component based k-nearest-neighbor rule for semiconductor process fault detection, American Control Conference, IEEE, Seattle, WA, USA, (2008) 1606-1611.
[30] H.H. Ma, Y. H, H.B. Shi, A novel local neighborhood standardization strategy and its application in fault detection of multimode processes, Chemometr. Intell. Lab. Syst. 118(2012) 287-300.
[31] X.G. Deng, X.M. Tian, Multimode process fault detection using local neighborhood similarity analysis, Chin. J. Chem. Eng. 22(11) (2014) 1260-1267.
[32] Y. Zheng, S.M. Mao, S.J. Liu, D.S.H. Wong, Y.W. Wang, Normalized relative RBC based minimum risk Bayesian decision approach for fault diagnosis of industrial process, IEEE Trans. Ind. Electron. 63(12) (2016) 7723-7732.
[33] Y. Zheng, B. Ai, D.S.H. Wong, S.S. Jang, Y.W. Wang, J. Zhang, An EWMA algorithm with a cycled resetting (CR) discount factor for drift and fault of high-mix Run-To-Run Control, IEEE Trans. Ind. Informatics 6(2) (2010) 229-242.
[34] Z.Q. Ge, Z.H. Song, Multimode process monitoring based on Bayesian method, J. Chemom. 23(12) (2010) 636-650.
[35] J.J. Downs, E.F. Vogel, A plant-wide industrial process control problem, Comput. Chem. Eng. 17(3) (1993) 245-255.