Skip to main content
SpringerLink
Log in

Adaptive multiblock kernel principal component analysis for monitoring complex industrial processes

  • Published:
Journal of Zhejiang University SCIENCE C Aims and scope Submit manuscript

Abstract

Multiblock kernel principal component analysis (MBKPCA) has been proposed to isolate the faults and avoid the high computation cost. However, MBKPCA is not available for dynamic processes. To solve this problem, recursive MBKPCA is proposed for monitoring large scale processes. In this paper, we present a new recursive MBKPCA (RMBKPCA) algorithm, where the adaptive technique is adopted for dynamic characteristics. The proposed algorithm reduces the high computation cost, and is suitable for online model updating in the feature space. The proposed algorithm was applied to an industrial process for adaptive monitoring and found to efficiently capture the time-varying and nonlinear relationship in the process variables.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  • Cheng, C.Y., Hsu, C.C., Chen, M.C., 2010. Adaptive kernel principal component analysis (KPCA) for monitoring small disturbances of nonlinear processes. Ind. Eng. Chem. Res., 49(5):2254–2262. [doi:10.1021/ie900521b]

    Article  Google Scholar 

  • Chiang, L.H., Russell, F.L., Braatz, R.D., 2001. Fault Detection and Diagnosis in Industrial Systems. Springer, London.

    MATH  Google Scholar 

  • Elshenawy, L.M., Yin, S., Naik, A.S., Ding, S.X., 2010. Efficient recursive principal component analysis algorithms for process monitoring. Ind. Eng. Chem. Res., 49(1):252–259. [doi:10.1021/ie900720w]

    Article  Google Scholar 

  • Gallagher, V.B., Wise, R.M., Butler, S.W., White, D.D., Barna, G.G., 1997. Development and Benchmarking of Multivariate Statistical Process Control Tools for a Semiconductor Etch Process: Improving Robustness Through Model Updating. Proc. ADCHEM, p.78–83.

  • Ge, Z., Yang, C., Song, Z., 2009. Improved kernel PCA-based monitoring approach for nonlinear processes. Chem. Eng. Sci., 64(9):2245–2255. [doi:10.1016/j.ces.2009.01.050]

    Article  Google Scholar 

  • Jeng, J.C., Li, C.C., Huang, H.P., 2007. Fault detection and isolation for dynamic processes using recursive principal component analysis (PCA) based on filtering of signals. Asia-Pacific J. Chem. Eng., 2(6):501–509. [doi:10.1002/apj.94]

    Article  Google Scholar 

  • Jia, F., Martin, E.B., Morris, A.J., 2000. Nonlinear principal components analysis with application to process fault detection. Int. J. Syst. Sci., 31(11):1473–1487. [doi:10.1080/00207720050197848]

    Article  MATH  Google Scholar 

  • Kruger, U., Zhang, J., Xie, L., 2007. Principal Manifolds for Data Visualization and Dimension Reduction. In: Gorban, A.N., Kégl, B., Wunsch, D.C., et al. (Eds.), Lecture Notes in Computational Science and Engineering, Vol. 58. Springer.

  • Liu, X., Kruger, U., Littler, T., Xie, L., Wang, S.Q., 2009. Moving window kernel PCA for adaptive monitoring of nonlinear processes. Chemometr. Intell. Lab. Syst., 96(2):132–143. [doi:10.1016/j.chemolab.2009.01.002]

    Article  Google Scholar 

  • Maestri, M.L., Cassanello, M.C., Hororwitz, G.I., 2009. Kernel PCA performance in processes with multiple operation modes. Chem. Prod. Process Model., 4(5), Article 7, p. 1–6. [doi:10.2202/1934-2659.1383]

    Google Scholar 

  • Qin, S.J., Valle, S., Piovoso, M.J., 2001. On unifying multi-block analysis with application to decentralized process monitoring. J. Chemometr., 15(9):715–742. [doi:10.1002/cem.667]

    Article  Google Scholar 

  • Voegtlin, T., 2005. Recursive principal components analysis. Neur. Networks, 18(8):1051–1063. [doi:10.1016/j.neunet.2005.07.005]

    Article  Google Scholar 

  • Wang, X., Kruger, U., Lennox, B., 2003. Recursive partial least squares algorithms for monitoring complex industrial processes. Control Eng. Pract., 11(6):613–632. [doi:10.1016/S0967-0661(02)00096-5]

    Article  Google Scholar 

  • Zhang, Y., Qin, S.J., 2010. Decentralized fault diagnosis of large-scale processes using multiblock kernel principal component analysis. Acta Autom. Sin., 36(4):593–597. [doi:10.3724/SP.J.1004.2010.00593]

    Article  MathSciNet  Google Scholar 

  • Zhou, D.H., Li, G., Qin, S.J., 2010. Total projection to latent structures for process monitoring. AIChE J., 56(1):168–178.

    MathSciNet  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. MOE Key Lab of Integrated Automation of Process Industry, Northeastern University, Shenyang, 110004, China

    Ying-wei Zhang & Yong-dong Teng

Authors
  1. Ying-wei Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Yong-dong Teng
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Ying-wei Zhang.

Additional information

Project supported by the National Basic Research Program (973) of China (No. 2009CB320600) and the National Natural Science Foundation of China (No. 60974057)

Rights and permissions

Reprints and permissions

About this article

Cite this article

Zhang, Yw., Teng, Yd. Adaptive multiblock kernel principal component analysis for monitoring complex industrial processes. J. Zhejiang Univ. - Sci. C 11, 948–955 (2010). https://doi.org/10.1631/jzus.C1000148

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1631/jzus.C1000148

Key words

CLC number

Search

Navigation