A novel Q-based online model updating strategy and its application in statistical process control for rubber mixing

doi:10.1016/j.cjche.2014.11.029

Abstract

Abstract: To overcome the large time-delay in measuring the hardness of mixed rubber, rheological parameters were used to predict the hardness. A novel Q-based model updating strategy was proposed as a universal platform to track time-varying properties. Using a few selected support samples to update the model, the strategy could dramatically save the storage cost and overcome the adverse influence of lowsignal-to-noise ratio samples.Moreover, it could be applied to any statistical processmonitoring systemwithout drastic changes to them,which is practical for industrial practices. As examples, the Q-based strategy was integrated with three popular algorithms (partial least squares (PLS), recursive PLS (RPLS), and kernel PLS (KPLS)) to form novel regression ones, QPLS, QRPLS and QKPLS, respectively. The applications for predicting mixed rubber hardness on a large-scale tire plant in east China prove the theoretical considerations.

Key words: Online model updating, Rubber mixing, Q statistic, Hardness, Rheological parameters, Statistical process control

摘要： To overcome the large time-delay in measuring the hardness of mixed rubber, rheological parameters were used to predict the hardness. A novel Q-based model updating strategy was proposed as a universal platform to track time-varying properties. Using a few selected support samples to update the model, the strategy could dramatically save the storage cost and overcome the adverse influence of lowsignal-to-noise ratio samples.Moreover, it could be applied to any statistical processmonitoring systemwithout drastic changes to them,which is practical for industrial practices. As examples, the Q-based strategy was integrated with three popular algorithms (partial least squares (PLS), recursive PLS (RPLS), and kernel PLS (KPLS)) to form novel regression ones, QPLS, QRPLS and QKPLS, respectively. The applications for predicting mixed rubber hardness on a large-scale tire plant in east China prove the theoretical considerations.

关键词: Online model updating, Rubber mixing, Q statistic, Hardness, Rheological parameters, Statistical process control

Chunying Zhang, Sun Chen, Fang Wu, Kai Song. A novel Q-based online model updating strategy and its application in statistical process control for rubber mixing[J]. Chin.J.Chem.Eng., 2015, 23(5): 796-803.

Chunying Zhang, Sun Chen, Fang Wu, Kai Song. A novel Q-based online model updating strategy and its application in statistical process control for rubber mixing[J]. Chinese Journal of Chemical Engineering, 2015, 23(5): 796-803.

References

[1] M. Kano, Y. Nakagawa, Data-based process monitoring, process control, and quality improvement: recent developments and applications in steel industry, Comput. Chem. Eng. 32 (1-2) (2008) 12-24.

[2] P. Kadlec, B. Gabrys, S. Strandt, Data-driven soft sensors in the process industry, Comput. Chem. Eng. 33 (4) (2009) 795-814.

[3] P. Nomikos, J.F. MacGregor, Multi-way partial least squares in monitoring batch processes, Chemom. Intell. Lab. Syst. 30 (1) (1995) 97-108.

[4] B.R. Bakshi, Multiscale PCA with application to multivariate statistical process monitoring, AICHE J. 44 (7) (1998) 1596-1610.

[5] X. Zhang,W.W. Yan, X. Zhao, H.H. Shao, Nonlinear real-time process monitoring and fault diagnosis based on principal component analysis and kernel fisher discriminant analysis, Chem. Eng. Technol. 30 (9) (2007) 1203-1211.

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

[7] L. Wang, H. Shi, Improved kernel PLS-based fault detection approach for nonlinear chemical processes, Chinese J. Chem. Eng. 22 (6) (2014) 657-663.

[8] K. Helland, H.E. Berntsen, O.S. Borgen, H. Martens, Recursive algorithm for partial least squares regression, Chemometr. Intell. Lab. Syst. 14 (1-3) (1992) 129-137.

[9] X. Wang, U. Kruger, B. Lennox, Recursive partial least squares algorithms for monitoring complex industrial processes, Control. Eng. Pract. 11 (6) (2003) 613-632.

[10] S.J. Mu, Y.Z. Zeng, R.L. Liu, P.Wu, H.Y. Su, J. Chu, Online dual updating with recursive PLS model and its application in predicting crystal size of purified terephthalic acid (PTA) process, J. Process Control 16 (6) (2006) 557-566.

[11] K. Song, H.Wang, P. Li, Discounted-measurement RPLS algorithm and its application to quality control of rubber mixing process, CIESC J. 55 (6) (2004) 942-946 (in Chinese).

[12] S.W. Choi, E.B. Martin, A.J. Morris, I.B. Lee, Adaptive multivariate statistical process control for monitoring time-varying processes, Ind. Eng. Chem. Res. 45 (9) (2006) 3108-3118.

[13] M.J. Atalla, D.J. Inman, On model updating using neural networks, Mech. Syst. Signal Process. 12 (1) (1998) 135-161.

[14] H.S. Tang, S.T. Xue, R. Chen, T. Sato, Onlineweighted LS-SVMfor hysteretic structural system identification, Eng. Struct. 28 (12) (2006) 1728-1735.

[15] J.A.Westerhuis, S.P. Gurden, A.K. Smilde, Generalized contribution plots inmultivariate statistical processmonitoring, Chemometr. Intell. Lab. Syst. 51 (1) (2000) 95-114.

[16] J.F. MacGregor, T. Kourti, Statistical process control of multivariate processes, Control. Eng. Pract. 3 (3) (1995) 403-414.

[17] D.G. Alvarez, M.J. Fuente, P. Vega, G. Sainz, Fault detection and diagnosis using multivariate statistical techniques in a wastewater treatment plant, in: S. Engell (Ed.), 7th IFAC International Symposium on Advanced Control of Control of Chemical Processes, Curran Associates, Inc., USA, 2009, pp. 952-957.

[18] Q. Jiang, X. Yan, Statistical monitoring of chemical processes based on sensitive kernel principal components, Chinese J. Chem. Eng. 21 (6) (2013) 633-643.

[19] Y. Zhang, C. Ma, Fault diagnosis of nonlinear processes using multiscale KPCA and multiscale KPLS, Chem. Eng. Sci. 66 (1) (2011) 64-72.

[20] R. Jia, Z. Mao, F.Wang, KPLS model based product quality control for batch processes, CIESC J. 64 (4) (2013) 1332-1339 (in Chinese).

[21] S. Bersimis, S. Psarakis, J. Panaretos, Multivariate statistical process control charts: an overview, Qual. Reliab. Eng. Int. 23 (5) (2007) 517-543.

[22] L.Y. Jiang, B.J. Xu, J.H. Xi, J.G. Cui, L. Fu, Improved confidence limits of T² statistic for monitoring batch processes, 24th Chinese Control and Decision Conference (CCDC), IEEE, USA, 2012, pp. 2928-2932.

[23] T. Chen, J. Morris, E. Martin, Probability density estimation via an infinite Gaussian mixture model: application to statistical process monitoring, J. R. Stat. Soc.: Ser. C: Appl. Stat. 55 (2006) 699-715.

[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.E. Jackson, G.S. Mudholkar, Control procedures for residuals associated with principal component analysis, Technometrics 21 (3) (1979) 341-349.

[26] Y.H. Shi, R.C. Eberhart, Fuzzy adaptive particle swarm optimization, Proceedings of the 2001 Congress on Evolutionary Computation, IEEE, USA, 2001, pp. 101-106.

[27] O. Chapelle, V. Vapnik, O. Bousquet, S.Mukherjee, Choosingmultiple parameters for support vector machines, Mach. Learn. 46 (1-3) (2002) 131-159.

[28] J. Haddock, J. Mittenthal, Simulation optimization using simulated annealing, Comput. Ind. Eng. 22 (4) (1992) 387-395.

[29] R. Rubinstein, The cross-entropy method for combinatorial and continuous optimization, Methodol. Comput. Appl. Probab. 1 (2) (1999) 127-190.

[30] R. Battiti, M. Brunato, F. Mascia, Reactive Search and Intelligent Optimization, Springer, Boston, 2008.

[31] S. Boettcher, A.G. Percus, Extremal Optimization: An Evolutionary Local-search Algorithm, Kluwer Academic, Berlin, 2003. 61-77.

[32] A.J. Burnham, J.F. MacGregor, R. Viveros, Latent variable multivariate regression modeling, Chemometr. Intell. Lab. Syst. 48 (2) (1999) 167-180.

[33] S. Joe Qin, Recursive PLS algorithms for adaptive data modeling, Comput. Chem. Eng. 22 (4-5) (1998) 503-514.

[34] F. Lindgren, P. Geladi, S.Wold, The kernel algorithmfor PLS, J. Chemom. 7 (1) (1993) 45-59.

[35] P.K. Freakley, Rubber Processing and Production Organization, Plenum Press, New York, 1985.

[36] Z. Zhang, K. Song, T.P. Tong, F. Wu, A novel nonlinear adaptive Mooney-viscosity model based on DRPLS-GP algorithm for rubber mixing process, Chemometr. Intell. Lab. Syst. 112 (2012) 17-23.

[37] P. Geladi, Notes on the history and nature of partial least squares (PLS) modelling, J. Chemom. 2 (4) (1988) 231-246.

[38] S.Wold, K. Esbensen, P. Geladi, Principal component analysis, Chemometr. Intell. Lab. Syst. 2 (1-3) (1987) 37-52.

[39] C. Cortes, V. Vapnik, Support-vector networks, Mach. Learn. 20 (3) (1995) 273-297.