Skip to Main content Skip to Navigation
Theses

Remémoration guidée par l'adaptation et maintenance des systèmes de diagnostic industriel par l'approche du raisonnement à partir de cas.

Abstract : The development of new technologies of the various products and components made the nature of systems more and more complex. This complexity is reverberated on the good functioning of the equipment with the appearance of new failures and the increasing involved costs. Maintenance became an indispensable element to maintain any equipment in operational condition regardless of its nature. In this context we are interested in corrective maintenance and more particularly in fault diagnostic of the industrial equipment. We have developed a method based on case-based reasoning (CBR), this latter being widely used in the industrial diagnostic domain. The CBR is an approach for problem solving and learning. In diagnostic, several CBR systems, ranging from classification problems “casebase mining” to knowledge-based systems, have proven their effectiveness. We have deployed in the first type of system a method insuring its maintenance where the case formalization is trivial. The whole system's maintenance requires the maintenance of the case-base which represents the heart of CBR systems. This method of maintenance is composed of a structuring step and an auto-increment step of the case-base, both aim at ensuring the quality of the system throughout its evolution. As for the second type of system, we have implemented a system which relies on knowledge models associated with different handling phases of the CBR cycle. We have proposed an adaptation-guided retrieval method based on two measures, one of similarity and one of adaptation, and an adaptation algorithm dedicated to the industrial diagnostic field. Our proposals have been implemented and validated on a GaMA-Frame (Global asset MAintenance) e-maintenance platform. This platform integrates our CBR diagnostic module and different knowledge models to diagnose faults on SISTRE (Supervised Industrial System of pallets TRansfEr) equipment.
Document type :
Theses
Complete list of metadata

https://tel.archives-ouvertes.fr/tel-00466560
Contributor : Martine Azema Connect in order to contact the contributor
Submitted on : Wednesday, March 24, 2010 - 10:51:38 AM
Last modification on : Thursday, January 13, 2022 - 12:00:14 PM
Long-term archiving on: : Monday, June 28, 2010 - 4:33:56 PM

Identifiers

  • HAL Id : tel-00466560, version 1

Citation

Mohamed Karim Haouchine. Remémoration guidée par l'adaptation et maintenance des systèmes de diagnostic industriel par l'approche du raisonnement à partir de cas.. Automatique / Robotique. Université de Franche-Comté, 2009. Français. ⟨tel-00466560⟩

Share

Metrics

Record views

404

Files downloads

985