A Synaptic Indicator Based Approach For Hidden Parameters Extraction In Industrial Environment

  • Kurosh Madani
  • Ion Berechet
Conference paper
Part of the Advances in Soft Computing book series (AINSC, volume 19)


In the case of a large number of applications, especially complex industrial ones, the knowledge on system’s (process, plant, etc.) parameters during the operation of the system is of major importance. However, in real cases, there are always parameters, which are not accessible. In the present work, we focus our interest around the extraction possibility of information relative to inaccessible parameters, which is a difficult problem in a general context. We will discuss some realistic and especially, realizable conditions for which a solution could be approached. In proposed approach, we use the neural network’s learning and a synaptic weight based indicator to detect changes related to system’s inaccessible parameters. Experimental results relative to a real industrial process have been reported validating our approach.


Artificial Neural Network Output Neuron Synaptic Weight Internal Parameter Virtual Sensor 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    M. Mayoubi, M. Schafer, S. Sinsel 1995 ), “Dynamic Neural Units for Non-linear Dynamic Systems Identification”, LNCS Vol. 930, Springer Verlag, 1995, pp. 10451051.Google Scholar
  2. [2]
    Faller W., Schreck S. (1995), “Real-Time Prediction Of Unsteady Aerodynamics: Application for Aircraft Control and Maneuverability Enhancement”, IEEE Transactions on Neural Networks, Vol. 6, Ni 6, Nov. 1995.Google Scholar
  3. [3]
    Y. Maidon, B. W. Jervis, N. Dutton, S. Lesage (1996), “Multifault Diagnosis of Analogue Circuits Using Multilayer Perceptrons,” IEEE European Test Workshop 96, Montpellier, June 12–14, 1996.Google Scholar
  4. [4]
    I. Grabec (1997), “Continuation of Chaotic Fields by RBFNN”, LNCS Vol. 1240–Springer Verlag 1997, pp. 597–606.Google Scholar
  5. [5]
    A. Sachenko, V. Kochan, V. Turchenko, V. Golovko, J. Savitsky, A. Dunets, T. Laopoulos (2000), “Sensor errors prediction using neural networks”, Proceedings IJCNN’2000, Jul 24-Jul 27 2000, Como, Italy, pp. 441–446.Google Scholar
  6. [6]
    A. Sachenko (1999), “IMCS development on the basis of distributed sensor networks”, IEEE AFRICON Conference, vl, 1999 IEEE, Piscataway, NJ, USA, p 345–350.Google Scholar
  7. [7]
    Y. Zhang, Y. MA (1997). MA (1997), “CGHA for Principal Component extraction in the complex domain”, IEEE Trans. On Neural Networks, Vol. 8, No. 5, 1997, pp. 1031–1036.CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Kurosh Madani
    • 1
  • Ion Berechet
    • 2
  1. 1.Intelligence in Instrumentation and Systems Lab. (I2S) - SENART Institute of TechnologyUniversity PARIS XIILieusaintFrance
  2. 2.CREATA Holding SANeuchâtelSwitzerland

Personalised recommendations