DYNAMICAL CLUSTERING TECHNIQUE TO ESTIMATE THE PROBABILITY OF THE FAILURE OCCURRENCE OF PROCESS SUBJECTED TO SLOW DEGRADATION

M. Traore, E. Duviella, S. Lecoeuche

2009

Abstract

In this paper, we propose a supervision method which aims at determining pertinent indicators to optimize predictive maintenance strategies. The supervision method, based on the AUto-adaptative and Dynamical Clustering technique (AUDyC), consists in classifying in real time measured data into classes representative of the operating modes of the process. This technique also allows the detection and the tracking of the slow evolutions of the process modes. Based on the AUDyC technique, a method is proposed to estimate the probabilities of the failure occurence of components in real time. This method is illustrated on the real case of a temperature controller.

References

  1. Anguita, J. and Hernando, J. (2004). Inter-phone and interword distances for confusability prediction in speech recognition. Congreso de la Sociedad Espaola para el Procesamiento del Lenguaje Natural, (33):33-40.
  2. Desinde, M., Flaus, J. M., and Ploix, S. (2006). Tool and methodology for online risk assessement of process. In Lambda-Mu 15 /Lille.
  3. Grall, A., Berenguer, C., and Dieulle, L. (2002). A condition-based maintenance policy for stochastically deteriorating systems. Reliability Engineering and System Safety, 76(2):167-180.
  4. Kullback, S. and Leibler, R. A. (1951). On information and sufficiency. Annal of Mathematical Statistics,22:79- 86.
  5. Lassagne, M. (2000). Applying a decision-analysis-based method to the evaluation of potential risk-reducing measures : The case of a floating production storage and offloading unit in the gulf of mexico. SPE annual technical conference, Dallas TX , USA.
  6. Lecoeuche, S., Lurette, C., and Lalot, S. (2004). New supervision architecture based on on-line modelling of non-stationary data. Neural Computing and Applications Journal, 13:323-338.
  7. Lurette, C. and Lecoeuche, S. (2003). Unsupervised and auto-adaptive neural architecture for on-line monitoring. application to a hydraulic process. Engineering Applications of Artificial Intelligence, 16:441-451.
  8. Mouchawed, S. M. and Billaudel, P. (2002). Influence of the choice of histogram parameters at fuzzy pattern matching performance, int. journal of wseas transactions on system. WSEAS Transactions on Systems, 1:260-266.
  9. Muller, A., Suhner, M.-C., Iung, B., and Morel, G. (2004). Prognosis-based maintenance decision-making for industrial process performance optimisation. In 7th IFAC Symposium on Cost Oriented Automation (COA2004). Gatineau/Ottawa Canada.
  10. Vesely, W. E., Goldberg, F. F., Robert, N. H., and Haasl, D. F. (1981). Fault Tree Handbook. US nuclear Regulatory Commission, Washington D.C., USA.
Download


Paper Citation


in Harvard Style

Traore M., Duviella E. and Lecoeuche S. (2009). DYNAMICAL CLUSTERING TECHNIQUE TO ESTIMATE THE PROBABILITY OF THE FAILURE OCCURRENCE OF PROCESS SUBJECTED TO SLOW DEGRADATION . In Proceedings of the 6th International Conference on Informatics in Control, Automation and Robotics - Volume 1: ICINCO, ISBN 978-989-674-000-9, pages 360-365. DOI: 10.5220/0002250003600365


in Bibtex Style

@conference{icinco09,
author={M. Traore and E. Duviella and S. Lecoeuche},
title={DYNAMICAL CLUSTERING TECHNIQUE TO ESTIMATE THE PROBABILITY OF THE FAILURE OCCURRENCE OF PROCESS SUBJECTED TO SLOW DEGRADATION},
booktitle={Proceedings of the 6th International Conference on Informatics in Control, Automation and Robotics - Volume 1: ICINCO,},
year={2009},
pages={360-365},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0002250003600365},
isbn={978-989-674-000-9},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 6th International Conference on Informatics in Control, Automation and Robotics - Volume 1: ICINCO,
TI - DYNAMICAL CLUSTERING TECHNIQUE TO ESTIMATE THE PROBABILITY OF THE FAILURE OCCURRENCE OF PROCESS SUBJECTED TO SLOW DEGRADATION
SN - 978-989-674-000-9
AU - Traore M.
AU - Duviella E.
AU - Lecoeuche S.
PY - 2009
SP - 360
EP - 365
DO - 10.5220/0002250003600365