Skip to main content

Advertisement

Springer Nature Link
Log in

Design of an auto-associative neural network by using design of experiments approach

  • KES 2008
  • Published:
Neural Computing and Applications Aims and scope Submit manuscript

Abstract

Advanced monitoring systems enable integration of data-driven algorithms for various tasks, for e.g., control, decision support, fault detection and isolation (FDI), etc. Due to improvement of monitoring systems, statistical or other computational methods can be implemented to real industrial systems. Algorithms which rely on process history data sets are promising for real-time operation especially for online process monitoring tasks, e.g., FDI. However, a reliable FDI system should be robust to uncertainties and small process deviations, thus, false alarms can be avoided. To achieve this, a good model for comparison between process and model is needed and for easier FDI implementation, the model has to be derived directly from process history data. In such cases, model-based FDI approaches are not very practical. In this paper a nonlinear statistical multivariate method (nonlinear principal component analysis) was used for modeling, and realized with auto-associative artificial neural network (AANN). A Taguchi design of experiments (DoE) technique was used and compared with a classic approach, where according to the analysis best AANN model structure was chosen for nonlinear model. Parameters that are important for neural network’s performance have been included into a joint orthogonal array to consider interactions between noise and control process variables. Results are compared to AANN design recommendations by other authors, where obtained nonlinear model was designed for reliable fault detection of very small faults under closed-loop conditions. By using Taguchi DoE robust design on AANN, an improved and reliable FDI scheme was achieved even in case of small faults introduced to the system. The accuracy and performance of AANN and FDI scheme were tested by experiments carried out on a real laboratory hydraulic system, to validate the proposed design for industrial cases.

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

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

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

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11

Similar content being viewed by others

Explore related subjects

Discover the latest articles and news from researchers in related subjects, suggested using machine learning.

Abbreviations

FDI:

Fault detection and isolation

PCA:

Principal component analysis

NLPCA:

Nonlinear PCA

AANN:

Auto-associative artificial neural network

DOE:

Design of experiments

FPE:

Final prediction error

AIC:

Information theoretic criterion

OA:

Orthogonal array

ANOVA:

Analysis of variance

S/N:

Signal to noise ratio

OPC:

OLE for process control

PLC:

Programmable logic controller

LM:

Levenberg–Marquardt backpropagation

GD:

Gradient descent backpropagation

GDX:

Gradient descent backpropagation with adaptation

Tansig:

Hyperbolic tangent sigmoid transfer function

Logsig:

Logarithmic sigmoid transfer function

Purelin:

Linear transfer function

References

  1. Iserman R (1984) Fault diagnosis of machines via parameter estimation and knowledge processing—tutorial paper. Automatica 20:387–404

    Article  Google Scholar 

  2. Basseville M, Nikiforov IV (1993) Detection of abrupt changes: theory and applications. Prentice Hall, Englewood Cliffs

    Google Scholar 

  3. Patton R, Frank P (1989) Fault diagnosis in dynamic systems. Prentice Hall, New York

    Google Scholar 

  4. Chiang LH, Russell EL, Braatz RD (2001) Fault detection and diagnosis in industrial control. Springer, New York

    Google Scholar 

  5. Korbitz JK, Kościelny JM, Kowalczuk Z, Cholewa W (2004) Fault diagnosis. Springer, Berlin

    Google Scholar 

  6. Portillo E, Marcos M, Cabanes I, Zubizarreta A (2009) Recurrent ANN for monitoring degraded behaviours in a range of workpiece thicknesses. doi:10.1016/j.engappai.2009.03.009

  7. Mok HT, Chan CW (2008) Online fault detection and isolation of nonlinear systems based on neurofuzzy networks. Eng Appl Artif Intel 21:171–181

    Article  Google Scholar 

  8. Venkatesan D, Kannan K, Saravanan R (2009) A genetic algorithm-based artificial neural network model for the optimization of machining processes. Neural Comput Appl 18:135–140

    Article  Google Scholar 

  9. Jolliffe IT (2004) Principal component analysis, 2nd edn. Springer, New York

    Google Scholar 

  10. Klančar G, Škrjanc I (2002) A principal component analysis in fault detection and isolation: hydraulic and fermentation process example. Electrotech Rev 69:311–316

    Google Scholar 

  11. Chen J, Liao C (2002) Dynamic process fault monitoring based on neural network and PCA. J Process Control 12:277–289

    Article  Google Scholar 

  12. Kramer MA (1991) Nonlinear principal component analysis using autoassociative neural networks. AIChE J 37:233–243

    Article  Google Scholar 

  13. Frank PM (1992) Robust model-based fault detection in dynamic systems. In: IFAC symposium on on-line fault detection and supervision in the chemical process industries. Newark, Delaware

  14. Dunia R, Qin SJ (1998) Joint diagnosis of process and sensor faults using principal component analysis. Control Eng Pract 6:457–469

    Article  Google Scholar 

  15. Gertler J (1998) Fault detection and diagnosis in engineering systems. Marcel Dekker, New York

    Google Scholar 

  16. Venkatasubramanian V, Raghunathan R, Yin K, Kavuri NS (2003) A review of process fault detection and diagnosis. Comput Chem Eng 27:293–311 (Part I)

    Article  Google Scholar 

  17. Venkatasubramanian V, Raghunathan R, Yin K, Kavuri NS (2003) A review of process fault detection and diagnosis. Comput Chem Eng 27:313–326 (Part II)

    Article  Google Scholar 

  18. Venkatasubramanian V, Raghunathan R, Yin K, Kavuri NS (2003) A review of process fault detection and diagnosis. Comput Chem Eng 27:327–346 (Part III)

    Article  Google Scholar 

  19. Uraikul V, Chan CW, Tontiwachwuthikul P (2007) Artificial intelligence for monitoring and supervisory control of process systems. Eng Appl Artif Intel 20:115–131

    Article  Google Scholar 

  20. MacGregor JF, Kourti T (1995) Statistical process control of multivariate processes. Control Eng Pract 3:403–414

    Article  Google Scholar 

  21. Zhang J, Martin EB, Morris AJ (1997) Process monitoring using non-linear statistical techniques. Chem Eng J 67:181–189

    Article  Google Scholar 

  22. Patton RJ, Uppal FJ, Lopez-Toribio CJ (2000) Soft computing approaches to fault diagnosis for dynamic systems: a survey. In: IFAC symposium SAFEPROCESS 2000, pp 298–311

  23. Jiang J, Wang J, Chu X, Yu R (1996) Neural network learning to non-linear principal component analysis. Anal Chim Acta 336:209–222

    Article  Google Scholar 

  24. Milde S, Kobe A (1997) An exact learning algorithm for autoassociative neural networks with binary couplings. J Phys A Math Gen 30:2349–2352

    Article  MATH  Google Scholar 

  25. Hidden HG, Willis MJ, Tham MT, Turner P, Montague GA (1997) Non-linear principal components analysis using genetic programming. Genetic algorithms in engineering systems: innovations and applications, Glasgow, UK, pp 302–307

  26. Jia F, Martin EB, Morris AJ (1998) Non-linear principal components analysis for process fault detection. Comput Chem Eng 20:851–854

    Article  Google Scholar 

  27. Hines JW, Uhrig RE, Wrest DJ (1998) Use of autoassociative neural networks for signal validation. J Intell Robot Syst 21:143–154

    Article  Google Scholar 

  28. Yang T, Wang S (2000) Fuzzy auto-associative neural networks for principal component extraction of noisy data. IEEE T Neural Netw 11:808–810

    Article  Google Scholar 

  29. Hsieh WW (2007) Nonlinear principal component analysis of noisy data. Neural Netw 20:434–443

    Article  MATH  MathSciNet  Google Scholar 

  30. Malthouse EC (1998) Limitations of nonlinear PCA as performed with generic neural networks. IEEE T Neural Netw 9:165–173

    Article  Google Scholar 

  31. Yegnanarayana B, Kishore SP (2002) AANN: an alternative to GMM for pattern recognition. Neural Netw 15:459–469

    Article  Google Scholar 

  32. Yeh T, Huang M, Huang C (2003) Estimate of process compositions and plantwide control from multiple secondary measurements using artificial neural networks. Comput Chem Eng 27:55–72

    Article  Google Scholar 

  33. Scholz M, Kaplan F, Guy CG, Kopka J, Selbig J (2005) Non-linear PCA: a missing data approach. Bioinformatics 21:3887–3895

    Article  Google Scholar 

  34. Zabiri H, Ramasamy M (2009) NLPCA as a diagnostic tool for control valve stiction. J Process Contr. doi:10.1016/j.jprocont.2009.04.010

  35. Taguchi G, Chowdhury S, Wu Y (2004) Taguchi’s quality engineering handbook. Wiley and Sons, New Jersey

    Book  Google Scholar 

  36. Taguchi G, Jugulum R (2002) The Mahalanobis–Taguchi strategy. Wiley and Sons, New York

    Book  Google Scholar 

  37. Di Mascio R, Barton GW (2001) The economic assessment of process control quality using a Taguchi-based method. J Process Control 11:81–88

    Article  Google Scholar 

  38. Vlachogiannis JG, Roy RK (2005) Robust PID controllers by Taguchi’s method. TQM Mag 17:456–466

    Article  Google Scholar 

  39. Kim Y, Yum B (2004) Robust design of multilayer feedforward neural networks: an experimental approach. Eng Appl Artif Intel 17:249–263

    Article  Google Scholar 

  40. Sukthomya W, Tannock J (2005) The optimisation of neural network parameters using Taguchi’s design of experiments approach: an application in manufacturing process modeling. Neural Comput Appl 14:337–344

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Faculty of Electrical Engineering and Computer Science, University of Maribor, Smetanova 17, 2000, Maribor, Slovenia

    Božidar Bratina, Nenad Muškinja & Boris Tovornik

Authors
  1. Božidar Bratina
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Nenad Muškinja
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Boris Tovornik
    View author publications

    You can also search for this author inPubMed Google Scholar

Corresponding author

Correspondence to Božidar Bratina.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Bratina, B., Muškinja, N. & Tovornik, B. Design of an auto-associative neural network by using design of experiments approach. Neural Comput & Applic 19, 207–218 (2010). https://doi.org/10.1007/s00521-009-0287-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00521-009-0287-6

Keywords