Parameter selection algorithm with self adaptive growing neural network classifier for diagnosis issues

  • M. Barakat
  • D. Lefebvre
  • M. Khalil
  • F. Druaux
  • O. Mustapha
Original Article


Neural networks have been widely used in the field of intelligent information processing such as classification, clustering, prediction, and recognition. In this paper, a non-parametric supervised classifier based on neural networks is proposed for diagnosis issues. A parameter selection with self adaptive growing neural network (SAGNN) is developed for automatic fault detection and diagnosis in industrial environments. The growing and adaptive skill of SAGNN allows it to change its size and structure according to the training data. An advanced parameter selection criterion is embedded in SAGNN algorithm based on the computed performance rate of training samples. This approach (1) improves classification results in comparison to recent works, (2) achieves more optimization at both stages preprocessing and classification stage, (3) facilitates data visualization and data understanding, (4) reduces the measurement and storage requirements and (5) reduces training and time consumption. In growing stage, neurons are added to hidden subspaces of SAGNN while its competitive learning is an adaptive process in which neurons become more sensitive to different input patterns. The proposed classifier is applied to classify experimental machinery faults of rotary elements and to detect and diagnose disturbances in chemical plant. Classification results are analyzed, explained and compared with various non-parametric supervised neural networks that have been widely investigated for fault diagnosis.


Adaptive neural network Fault diagnosis Parameter selection 


  1. 1.
    Isermann R (2005) Fault diagnosis systems: an introduction from fault detection to fault tolerance. Springer, BerlinGoogle Scholar
  2. 2.
    Korbicz J, Koscielny JM, Kowalczuk Z, Cholewa W (2004) Fault diagnosis models, artificial intelligence applications. Springer, BerlinMATHGoogle Scholar
  3. 3.
    Patton RJ, Chen J (1999) Robust model-based fault diagnosis for dynamic systems. Kluwer Academic Publishers, LondonMATHGoogle Scholar
  4. 4.
    Haykin S (1999) Neural networks, a comprehensive foundation, 2nd edn. Macmillan, New YorkMATHGoogle Scholar
  5. 5.
    Vikas C, Anil K, Ahlawat, R.S. Bhatia, (2011) Growing neural networks using soft competitive learning. Int J Comput Appl 21(3)Google Scholar
  6. 6.
    Xinjian Q, Guojian C, Zheng W, (2010) An overview of some classical growing neural networks and new developments. In: 2nd international conference on education technology and computer (ICETC)Google Scholar
  7. 7.
    Marsland S, Sphairo J, Nehmzow U (2002) A self-organizing network that grows when required. Neural Netw 15:1041–1058CrossRefGoogle Scholar
  8. 8.
    Daxin T, Yueou R, Qiuju L (2008) Dynamic growing self-organizing neural network for clustering. Springer, Berlin, pp 589–595Google Scholar
  9. 9.
    Kamimura R (2003) Competitive learning by information maximization: eliminating dead neurons in competitive learning. In: Kaynak O, Alpaydın E, Oja E, Xu L (eds) ICANN 2003 and ICONIP 2003, vol 2714. LNCS, Springer, Heidelberg, pp 99–106Google Scholar
  10. 10.
    Andrew L (2000) Analyses on the generalized lotto-type competitive learning. In: Leung K-S, Chan L, Meng H (eds) IDEAL 2000, vol 1983. LNCS, Springer, Heidelberg, pp. 9–16Google Scholar
  11. 11.
    Dellomo MR (1999) Helicopter gearbox fault detection: a neural network based approach. J Vib Acoust 121(3):265–272CrossRefGoogle Scholar
  12. 12.
    Su H, Chong KT (2007) Induction machine condition monitoring using neural network modeling. IEEE Trans Ind Electron 54(1):241–249CrossRefGoogle Scholar
  13. 13.
    Wasserman PD (1999) Advanced methods in neural computing. Van Nostrand Reinhold, New YorkGoogle Scholar
  14. 14.
    Specht DF (1999) Probabilistic neural networks. Neural Netw 3:109–118CrossRefGoogle Scholar
  15. 15.
    Guang-Bin H, Dian HW, Yuan L (2011) Extreme learning machines: a survey. Int J Mach Learn Cybern 2(2):107–122CrossRefGoogle Scholar
  16. 16.
    Xi-zhao W, Aixia C, Huimin F (2011) Upper integral network with extreme learning mechanism. Neurocomputing 74(16):2520–2525CrossRefGoogle Scholar
  17. 17.
    Fukunaga K (1972) Introduction to statistical pattern recognition. Purdue University/Academic Press, Lafayette/LondonGoogle Scholar
  18. 18.
    Dong LT, Robert M (2010) Genetic algorithm-neural network (GANN): a study of neural network activation functions and depth of genetic algorithm search applied to feature selection. Int J Mach Learn Cybern 1(1–4):75–87Google Scholar
  19. 19.
    Xi-Zhao W, Li Chun-Guo, Daniel SY, Shi JS, Hui MF (2008) A definition of partial derivative of random functions and its application to RBFNN sensitivity analysis. Neurocomputing 71(7–9):1515–1526Google Scholar
  20. 20.
    Samanta B, Al-Baulshi KR (2003) Artificial neural network based fault diagnostics of rolling element bearings using time domain features. Mech Syst Signal Process 17(2):317–328CrossRefGoogle Scholar
  21. 21.
    Fernández S, Jesús M, Murakami S (2004) A comparison of two learning mechanisms for the automatic design of fuzzy diagnosis systems for rotating machinery. Appl Soft Comput 4:413–422CrossRefGoogle Scholar
  22. 22.
    Nandi AK (2000) Advanced digital vibration signal processing for condition monitoring. In: Proceedings of COMADEM2000, Houston, TX, pp 129–143Google Scholar
  23. 23.
    Jack LB, Nandi AK (2000) Genetic algorithms for feature extraction in machine condition monitoring with vibration signals. IEEE Proc Vis Image Signal Process 147:205–212CrossRefGoogle Scholar
  24. 24.
    Belotti V, Crenna F, Michelini RC, Rossi GB (2006) Wheel-flat diagnostic tool via wavelet transform. Mech Syst Signal Process 20:1953–1966CrossRefGoogle Scholar
  25. 25.
    Binu P, Chacko V, Vimal Krishnan R, Raju G, Babu Anto P (2011) Handwritten character recognition using wavelet energy and extreme learning machine. Int J Mach Learn Cybern. doi: 10.1007/s13042-011-0049-5
  26. 26.
    Wang XZ, Dong CR (2009) Improving generalization of fuzzy if-then rules by maximizing fuzzy entropy. IEEE Trans Fuzzy Syst 17(3):556–567CrossRefGoogle Scholar
  27. 27.
    Wang XZ, Zhai JH, Lu SX (2008) Induction of multiple fuzzy decision trees based on rough set technique. Inf Sci 178(16):3188–3202MathSciNetCrossRefMATHGoogle Scholar
  28. 28.
    Mendel T, Rauber W, Varejao FM, Batista RJ (2009), Rolling element bearing fault diagnosis in rotating machines of oil extraction rigs. 17th European signal processing conference, ScotlandGoogle Scholar
  29. 29.
    Barakat M, Lefebvre D, Khalil M, Mustapha O, Druaux F (2010), Input–output classification mapping for the fault detection, identification and accommodation. In: Proc. IEEE-SMC, Istanbul (to appear)Google Scholar
  30. 30.
    Hsin-Chang Y, Chung-Hong L (2003) A text mining approach on automatic generation of web directories and hierarchies. In: Proceedings of the IEEE/WIC international conference on web intelligence (WI’03)Google Scholar
  31. 31.
    Yan Y, Pederson JO (1997) Comparative study of feature selection in text categorization. In: Proceedings on fourteenth international conference on machine learning (ICML’97), pp 412–420Google Scholar
  32. 32.
    Hecht-Nielsen R (1990) Neurocomputing. Addison-Wesley, ReadingGoogle Scholar
  33. 33.
    Cybenko G (1989) Approximations by superposition of a sigmoidal function. Math Control Signal Syst 2:303–314MathSciNetCrossRefMATHGoogle Scholar
  34. 34.
    Park J, Sandberg JW (1991) Universal approximation using radial basis functions network. Neural Comput 3:246–257Google Scholar
  35. 35.
    Poggio T, Girosi F (1990) Networks for approximation and learning. Proc IEEE 78:1481–1497CrossRefGoogle Scholar
  36. 36.
    Jack LB, Nandi AK (2002) Fault detection using support vector machines and artificial neural networks, augmented by genetic algorithms. Mech Syst Signal Process 16:373–390CrossRefGoogle Scholar
  37. 37.
    Yang W (2006) Establishment of the mathematical model for diagnosing the engine valve faults by genetic programming. J Sound Vib 293:213–226CrossRefGoogle Scholar
  38. 38.
    Merry R, Steinbuch M (2005) Wavelet theory and applications, literature study. Eindhoven University of Technology, Department of Mechanical Engineering, Control systems technology groupGoogle Scholar
  39. 39.
    Arthur A (2003) Signal processing applications of wavelets. University of California, Information and Computer Sciences, USAGoogle Scholar
  40. 40.
    Addison PS (2002) The illustrated wavelet transform handbook. IOP Publishing Ltd, ISBNCrossRefMATHGoogle Scholar
  41. 41.
    Schneiders MGE (2001) Wavelets in control engineering. Master’s thesis, Eindhoven University of TechnologyGoogle Scholar
  42. 42.
    Barakat M, Druaux F, Lefebvre D, Khalil M, Mustapha O (2011) Self adaptive growing neural network classifier for faults detection and diagnosis. Neurocomput J Elsevier 74:3865–3876CrossRefGoogle Scholar
  43. 43.
    Chen G, McAvoy TJ (1997) Predictive on-line monitoring of continuous processes, J. Process Control 8:409–420CrossRefGoogle Scholar
  44. 44.
    McAvoy TJ (1998) A methodology for screening level control structures in plant-wide control systems. Comput Chem Eng 22:1543–1552CrossRefGoogle Scholar
  45. 45.
    Downs JJ, Vogel EF (1993) A plant-wide industrial control problem. Comput Chem Eng 17:245–255CrossRefGoogle Scholar
  46. 46.
    Multiscale Systems Research Laboratory.
  47. 47.
    Leclercq E, Druaux F, Lefebvre D, Zerkaoui S (2005) Autonomous learning algorithm for fully connected recurrent networks. Neurocomputing 63:25–44CrossRefGoogle Scholar
  48. 48.
    Zerkaoui S, Druaux F, Leclercq E, and Lefebvre D (2007) Multivariable adaptive control for non-linear systems: application to the Tennessee Eastman Challenge Process, GreeceGoogle Scholar
  49. 49.
    Barakat M, Druaux F, Lefebvre D, Khalil M, and Mustapha O (2011) Intelligent condition monitoring of rotary machine by mean of self adaptive growing neural network, IEEE-MED, GreeceGoogle Scholar
  50. 50.
    Barakat M, Lefebvre D, Khalil M, Mustapha O, Druaux F (2010) FDI based on wavelet decomposition combined with parameters selection and RBF networks: application to the diagnosis of TECP reactor. In: Proc. IEEE-MED, Invited session ‘Diagnosis and Environmental issues’, Marrakech, pp 1347–1352Google Scholar
  51. 51.
    Barakat M, Lefebvre D, Khalil M, Mustapha O, Druaux F (2010) BSP-BDT Classification technique: application to rolling elements bearing. In: Conference on control and fault tolerant systems, “sys-Tol”, Nice, pp 654–659Google Scholar
  52. 52.
    Barakat M, Lefebvre D, Khalil M, Mustapha O, Druaux F (2009) FDD using wavelet decomposition combined with parameters elimination method and radial basis function network, LAAS16, BeirutGoogle Scholar
  53. 53.
    Khalil M, Khalil K (2007) Faults identification in industrial machines using non parametric classification methods. Int J Comput Cognit 5:21–26Google Scholar

Copyright information

© Springer-Verlag 2012

Authors and Affiliations

  • M. Barakat
    • 1
    • 2
  • D. Lefebvre
    • 1
  • M. Khalil
    • 2
  • F. Druaux
    • 1
  • O. Mustapha
    • 3
  1. 1.GREAH, Le Havre UniversityLe HavreFrance
  2. 2.Azm Center for Research in Biotechnology, Doctoral School for Sciences and TechnologyLebanese UniversityTripoliLebanon
  3. 3.Islamic University of LebanonKhaldehLebanon

Personalised recommendations