Advertisement

Supervised process monitoring and fault diagnosis based on machine learning methods

  • Hajer Lahdhiri
  • Maroua Said
  • Khaoula Ben Abdellafou
  • Okba TaoualiEmail author
  • Mohamed Faouzi Harkat
ORIGINAL ARTICLE
  • 8 Downloads

Abstract

Data-driven techniques have been receiving considerable attention in the industrial process monitoring field due to their major advantages of easy implementation and less requirement for the prior knowledge and process mechanism. Principal component analysis (PCA) method is known as a popular method for monitoring and fault detection in industrial systems but as it is basically a linear method. However, most practical systems are nonlinear. To make the extension to nonlinear systems, kernel PCA (KPCA) method has been proposed for process modeling and monitoring. We present in this paper an online reduced rank optimized KPCA (RR-KPCA) technique for fault detection in order to extend the advantages of the KPCA models to online processes. Following the fault detection, the identification of the variables correlated to the fault occurred is of great importance. For this purpose, it is proposed to extend the approaches of localization by partial PCA and by elimination in the linear case to the nonlinear case, by exploiting the solution of reduction of the dimension of the kernel matrix in the feature space. The partial RR-KPCA and the elimination sensor identification (ESI-RRKPCA) are generated based on the static RR-KPCA and the online RR-KPCA methods. The idea of these approaches is to generate partial RR-KPCA models with reduced sets of variables. In other words, their goal is to generate indices of fault detection sensitive to certain faults and insensitive to others. The proposed fault isolation methods are applied for monitoring an air quality monitoring network (AIRLOR) data.

Keywords

Reduced rank KPCA Nonlinear process monitoring Fault detection Tabu search algorithm Air quality monitoring Fault isolation 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Notes

References

  1. 1.
    Jaffel I, Taouali O, Elaissi I, Messaoud H (2014) A new online fault detection method based on PCA technique. IMA J Math Control Inf 31(4):487–499MathSciNetCrossRefzbMATHGoogle Scholar
  2. 2.
    Tharrault Y, Mourot G, Ragot J, Maquin D (2008) Fault detection and isolation with robust principal component analysis. Int J Appl Math Comput Sci 18(4):429–442CrossRefzbMATHGoogle Scholar
  3. 3.
    Mika S, Schölkopf B, Smola AJ, Müller KR, Scholz M, Rätsch G (1998, December) Kernel PCA and de-noising in feature spaces. In: NIPS, vol 11, pp 536–542Google Scholar
  4. 4.
    Schölkopf B, Smola A, Müller KR (1998) Nonlinear component analysis as a kernel eigenvalue problem. Neural Comput 10(5):1299–1319CrossRefGoogle Scholar
  5. 5.
    Aronszajn N (1950) Theory of reproducing kernels. Trans Am Math Soc 68(3):337–404MathSciNetCrossRefzbMATHGoogle Scholar
  6. 6.
    Fazai R, Taouali O, Harkat MF, Bouguila N (2016) A new fault detection method for nonlinear process monitoring. Int J Adv Manuf Technol 87(9–12):3425–3436CrossRefGoogle Scholar
  7. 7.
    Honeine P (2012) Online kernel principal component analysis: a reduced-order model. IEEE Trans Pattern Anal Mach Intell 34(9):1814–1826CrossRefGoogle Scholar
  8. 8.
    Kazor K, Holloway RW, Cath TY, Hering AS (2016) Comparison of linear and nonlinear dimension reduction techniques for automated process monitoring of a decentralized wastewater treatment facility. Stoch Env Res Risk A 30(5):1527–1544CrossRefGoogle Scholar
  9. 9.
    Lee JM, Yoo C, Lee IB (2004) Statistical process monitoring with independent component analysis. J Process Control 14(5):467–485CrossRefGoogle Scholar
  10. 10.
    Aizerman M, Braverman E, Rozonoer L (1964) Theoretical foundations of the potential function method in pattern recognition learning. Autom Remote Control 25:821–837zbMATHGoogle Scholar
  11. 11.
    Chouaib C, Mohamed-Faouzi H, Messaoud D (2015) New adaptive kernel principal component analysis for nonlinear dynamic process monitoring. Appl Math Inf Sci 9(4):1833–1845Google Scholar
  12. 12.
    Taouali O, Elaissi I, Messaoud H (2015) Dimensionality reduction of RKHS model parameters. ISA Trans 57:205–210CrossRefGoogle Scholar
  13. 13.
    Taouali O, Jaffel I, Lahdhiri H, Harkat MF, Messaoud H (2016) New fault detection method based on reduced kernel principal component analysis (RKPCA). Int J Adv Manuf Technol 85(5–8):1547–1552CrossRefGoogle Scholar
  14. 14.
    Taouali O, Elaissi I, Messaoud H (2012) Online identification of nonlinear system using reduced kernel principal component analysis. Neural Comput & Applic 21(1):161–169CrossRefGoogle Scholar
  15. 15.
    Ding C (2004) K -means clustering via principal component analysis, in the 21st Int Conf Mach Learn. Banff, CanadaGoogle Scholar
  16. 16.
    Dhillon IS (2004) Kernel k-means, spectral clustering and normalized cuts. Compute 78712:551–556Google Scholar
  17. 17.
    Jaffel I, Taouali O, Harkat MF, Messaoud H (2016) Moving window KPCA with reduced complexity for nonlinear dynamic process monitoring. ISA Trans 64:184–192CrossRefGoogle Scholar
  18. 18.
    Fezaia R, Mansourib M, Taoualia O, Harkatc MF, Bouguilaa N (2017) Online reduced kernel principal component analysis for process monitoring. J Process Control 61:1–11CrossRefGoogle Scholar
  19. 19.
    Gertler J, McAvoy T (1997) Principal component analysis and parity relations – a strong duality. IFAC Conference SAFEPROCESS, Hull, UK, pp. 837–842Google Scholar
  20. 20.
    Huang Y, Gertler J (1999) Fault isolation by partial PCA and partial NLPCA.. IFAC’99, 14th triennial world congress. P. R. China, Beijing, pp 545–550Google Scholar
  21. 21.
    Said M, Fazai R, Adellafou KB, Taouali O (2018) Decentralized fault detection and isolation using bond graph and PCA methods. Int J Adv Manuf Technol 99:1–13.  https://doi.org/10.1007/s00170-018-2526-4 CrossRefGoogle Scholar
  22. 22.
    Downs JJ, Vogel EF (1993) A plant-wide industrial process control problem. Comput Chem Eng 17:245–255CrossRefGoogle Scholar
  23. 23.
    Lyman PR, Georgakis C (1995) Plant-wide control of the Tennessee Eastmanproblem. Comput Chem Eng 19:321–331.  https://doi.org/10.1016/0098-1354(94)00057-U CrossRefGoogle Scholar
  24. 24.
    Zhao Y, Xiao L, Wen J, Lu Y, Wang S (2014) A robust pattern recognition-based fault detection and diagnosis (FDD) method for chillers. HVAC&R RESEARCH 20:798–809.  https://doi.org/10.1080/10789669.2014.938006 CrossRefGoogle Scholar
  25. 25.
    Zhao Y, Wang S, Xiao F (2013) Pattern recognition-based chillers fault detection method using support vector data description (SVDD). Appl Energy 112:1041–1048CrossRefGoogle Scholar
  26. 26.
    Chetouani Y (2008) A neural network approach for the real-time detection of faults. Stoch Env Res Risk A 22(3):339–349MathSciNetCrossRefzbMATHGoogle Scholar
  27. 27.
    Dong D, McAvoy TJ (1996) Nonlinear principal component analysis based on principal curves and neural networks. Comput Chem Eng 20(1):65–78CrossRefGoogle Scholar
  28. 28.
    Patan K, Parisini T (2005) Identication of neural dynamic models for fault detection and isolation: the case of a real sugar evaporation process. J Process Control 15:67–79CrossRefGoogle Scholar
  29. 29.
    Cristóvão RO, Pinto VMS, Gonçalves A, Martins RJE, Loureiro JM, Boaventura RAR (2016) Fish canning industry wastewater variability assessment using multivariate statistical methods. Process Saf Environ Prot 102:263–276CrossRefGoogle Scholar
  30. 30.
    Li G, Qin SJ, Zhou D (2010) Geometric properties of partial least squares for process monitoring. Automatica 46(1):204–210MathSciNetCrossRefzbMATHGoogle Scholar
  31. 31.
    Kano M, Tanaka S, Hasebe S, Hashimoto I, Ohno H (2003) Monitoring independent components for fault detection. AICHE J 49(4):969–976CrossRefGoogle Scholar
  32. 32.
    Harrou F, Kadri F, Khadraoui S, Sun Y (2016) Ozone measurements monitoring using data-based approach. Process Saf Environ Prot 100:220–231CrossRefGoogle Scholar
  33. 33.
    Madakyaru M, Harrou F, Sun Y (2017) Improved data-based fault detection strategy and application to distillation columns. Process Saf Environ Prot 107:22–34CrossRefGoogle Scholar
  34. 34.
    Cai L, Tian X (2014) A new fault detection method for non-Gaussian process based on robust independent component analysis. Process Saf Environ Prot 92(6):645–658CrossRefGoogle Scholar
  35. 35.
    Choi S, Morris J, Lee I (2008) Nonlinear multiscale modelling for fault detection and identification. Chem Eng Sci 63(8):2252–2266CrossRefGoogle Scholar
  36. 36.
    Cho JH, Lee JM, Choi SW, Lee D, Lee IB (2005) Fault identification for process monitoring using kernel principal component analysis. Chem Eng Sci 60(1):279–288CrossRefGoogle Scholar
  37. 37.
    Kallas M, Mourot G, Maquin D, Ragot J (2014) Diagnosis of nonlinear systems using kernel principal component analysis. In Journal of Physics: Conference Series (Vol. 570, No. 7, p. 072004). IOP PublishingGoogle Scholar
  38. 38.
    Sheriff MZ, Mansouric M, Nazmul Karima M, Nounouc H, Nounou M (2017) Fault detection using multiscale PCA-based moving window GLRT. J Process Control 54:47–64CrossRefGoogle Scholar
  39. 39.
    Zhang Y, Ma C (2011) Fault diagnosis of nonlinear processes using multiscale KPCA and multiscale KPLS. Chem Eng Sci 66(1):64–72CrossRefGoogle Scholar
  40. 40.
    Jicong Fan S, Qin J, Wang Y (2014) Online monitoring of nonlinear multivariate industrial processes using filtering KICA–PCA. Control Eng Pract 22(2014):205–216Google Scholar
  41. 41.
    Zhang Y (2009) Enhanced statistical analysis of nonlinear processes using KPCA, KICA and SVM. Chem Eng Sci 64(5):801–811CrossRefGoogle Scholar
  42. 42.
    Choi S, Lee I (2004) Nonlinear dynamic process monitoring based on dynamic kernel PCA. Chem Eng Sci 59(24):5897–5908CrossRefGoogle Scholar
  43. 43.
    Zhang N, Tian X, Cai L, Deng X (2015) Process fault detection based on dynamic kernel slow feature analysis. Comput Electr Eng 41:9–17CrossRefGoogle Scholar
  44. 44.
    Navi M, Meskin N, Davoodi M (2018) Sensor fault detection and isolation of an industrial gas turbine using partial adaptive KPCA. J Process Control 64:37–48CrossRefGoogle Scholar
  45. 45.
    Liu X, Kruger U, Littler T, Xie L, Wang S (2009) Moving window kernel PCA for adaptive monitoring of nonlinear processes. Chemom Intell Lab Syst 96(2):132–143CrossRefGoogle Scholar
  46. 46.
    Li H, Zhang D (2013) Stochastic representation and dimension reduction for non-Gaussian random fields: review and reflection. Stoch Env Res Risk A 27(7):1621–1635CrossRefGoogle Scholar
  47. 47.
    Vapnik V (1999) An overview of statistical learning theory. IEEE Trans Neural Netw 10(5):988–999CrossRefGoogle Scholar
  48. 48.
    Mercer J (1909) Functions of positive and negative type and their connection with the theory of integral equations. Philosophical transactions of the royal society of London. Series A, containing papers of a mathematical or physical character 209:415–446zbMATHGoogle Scholar
  49. 49.
    Jaffel I, Taouali O, Harkat MF, Messaoud H (2016) Kernel principal component analysis with reduced complexity for nonlinear dynamic process monitoring. Int J Adv Manuf Technol 88:3265(1–15).  https://doi.org/10.1007/s00170-016-8987-4
  50. 50.
    Zhang Y, Li S, Teng Y (2012) Dynamic processes monitoring using recursive kernel principal component analysis. Chem Eng Sci 72:78–86CrossRefGoogle Scholar
  51. 51.
    Lahdhiri H, Taouali O, Elaissi I, Jaffel I, Harakat MF, Messaoud H (2017) A new fault detection index based on Mahalanobis distance and kernel method. Int J Adv Manuf Technol 91: 2799:1–11.  https://doi.org/10.1007/s00170-016-9887-3
  52. 52.
    Nomikos P, MacGregor JF (1995) Multivariate SPC charts for monitoring batch processes. Technometrics 37(1):41–59CrossRefzbMATHGoogle Scholar
  53. 53.
    Choi SW, Lee C, Lee JM, Park JH, Lee IB (2005) Fault detection and identification of nonlinear processes based on kernel PCA. Chemom Intell Lab Syst 75(1):55–67CrossRefGoogle Scholar
  54. 54.
    Lee JM, Yoo C, Choi SW, Vanrolleghem PA, Lee IB (2004) Nonlinear process monitoring using kernel principal component analysis. Chem Eng Sci 59(1):223–234CrossRefGoogle Scholar
  55. 55.
    Lahdhiri H, Elaissi I, Taouali O, Harakat MF, Messaoud H (2017b) Nonlinear process monitoring based on new reduced rank-KPCA method. Stoch Env Res Risk A 32(6):1833–1848CrossRefGoogle Scholar
  56. 56.
    Lahdhiri H, Ben Abdellafou K, Taouali O, Mansouri M, Korbaa O (2018) New online kernel method with the Tabu search algorithm for process monitoring. Trans Inst Meas Control:1–12.  https://doi.org/10.1177/0142331218807271
  57. 57.
    Stork CL, Veltkamp DJ, Kowalski BR (1997) Identification of multiple sensor disturbances during process monitoring. Anal Chem 69(24):5031–5036CrossRefGoogle Scholar
  58. 58.
    Harkat MF, Mourot G, Ragot J (2006) An improved PCA scheme for sensor FDI: application to an air quality monitoring network. J Process Control 16(6):625–634CrossRefGoogle Scholar
  59. 59.
    Harkat MF, Tharrault Y, Mourot G, Ragot J (2010) Multiple sensor fault detection and isolation of an air quality monitoring network using RBF-NLPCA model. Int J Adapt Innov Syst 1(3–4):267–284CrossRefGoogle Scholar

Copyright information

© Springer-Verlag London Ltd., part of Springer Nature 2019

Authors and Affiliations

  • Hajer Lahdhiri
    • 1
  • Maroua Said
    • 2
  • Khaoula Ben Abdellafou
    • 3
    • 4
  • Okba Taouali
    • 1
    • 4
    Email author
  • Mohamed Faouzi Harkat
    • 5
  1. 1.National Engineering School of MonastirUniversity of MonastirMonastirTunisia
  2. 2.National Engineering School of Sousse, MARS Research Laboratory, LR17ES05University of SousseHammam SousseTunisia
  3. 3.ISITCom, MARS Research Laboratory, LR17ES05University of SousseHammam SousseTunisia
  4. 4.Faculty of Computers and Information TechnologyUniversity of TabukTabukSaudi Arabia
  5. 5.Department of ElectronicsUniversity Badji MokhtarAnnabaAlgeria

Personalised recommendations