Soft Sensor Modeling via Support Vector Regression with KICA-Based Feature Extraction

Abstract:

A novel two-level integrated soft sensor modeling method using kernel independent component analysis (KICA) and support vector regression (SVR) is proposed for chemical processes. In the first level, the KICA approach is adopted to extract information of input variables in the high dimensional feature space. Based on this strategy, the correlation of input variables can be eliminated and thus the complexity is reduced. Then, the model is established using SVR in the second level. The KICA-SVR soft sensor modeling method is applied to estimate product compositions in the Tennessee Eastman process. The obtained results show that it can exhibit better performance, compared to the traditional ICA, principal component analysis (PCA) and kernel PCA based information extraction methods, under different operating conditions.

• Periodical: Advanced Materials Research (Volume 186)
• Pages: 560-564
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.186.560
• Online since: January 2011
• Authors: Yi Liu, Hong Ying Deng, Zeng Liang Gao, Ping Li
• Keywords: Feature Extraction, Kernel Independent Component Analysis (KICA), Soft Sensor Modeling, Support Vector Regression (SVR)

References

[1] P. Kadlec, B. Gabrys and S. Strandt: Data-driven soft sensors in the process industry. Comput. Chem. Eng. Vol. 33 (2009), p.795.

DOI: 10.1016/j.compchemeng.2008.12.012

[2] C.M. Bishop: Pattern Recognition and Machine Learning (Springer-Verlag, New York, 2006).

[3] Y. Liu, N.P. Hu, H.Q. Wang and P. Li: Soft chemical analyzer development using adaptive least-squares support vector regression with selective pruning and variable moving window size. Ind. Eng. Chem. Res. Vol. 48 (2009), p.5731.

DOI: 10.1021/ie8012709

[4] E. Zamprogna, M. Barolo and D.E. Seborg: Optimal selection of soft sensor inputs for batch distillation columns using principal component analysis. J. Process Control Vol. 15 (2005), p.39.

DOI: 10.1016/j.jprocont.2004.04.006

[5] L.J. Cao, K.S. Chua, W.K. Chong, H.P., Lee and Q.M. Gu: A comparison of PCA, KPCA and ICA for dimensionality reduction in support vector machine. Neurocomput. Vol. 55 (2003), p.321.

DOI: 10.1016/s0925-2312(03)00433-8

[6] A.J. Chen, Z.H. Song and P. Li: Soft sensor modeling based on DICA-SVR. Lect. Notes Comput. Sci. Vol. 3644 (2005), p.868.

[7] Z. Li and X.M. Tian: Study of soft sensor modeling method based on KPCA-SVM. In Proceedings of the 7th World Congress on Intelligent Control and Automation, (2008), p.9162.

DOI: 10.1109/wcica.2006.1713311

[8] J.M. Lee, S.J. Qin and I.B. Lee: Fault detection of non-linear processes using kernel independent component analysis. The Canadian J. Chem. Eng. Vol. 85 (2007), p.526.

DOI: 10.1002/cjce.5450850414