Fault detection on non linear systems described by multiple linearized models

Laurent Gross, Hassan Noura, Didier Theilliol

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper deals with the use of adaptive Kalman filters for fault diagnosis on non linear systems described by multiple models. The matrices Q and R used to build up the gain of Kalman filters are considered here as synthesis parameters for these filters. Adequate variations of these matrices values allow to perform fault detection during the transition phase between two models.

Original languageEnglish
Title of host publicationEuropean Control Conference, ECC 1999 - Conference Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1219-1225
Number of pages7
ISBN (Electronic)9783952417355
DOIs
Publication statusPublished - Mar 24 2015
Event1999 European Control Conference, ECC 1999 - Karlsruhe, Germany
Duration: Aug 31 1999Sept 3 1999

Publication series

NameEuropean Control Conference, ECC 1999 - Conference Proceedings

Other

Other1999 European Control Conference, ECC 1999
Country/TerritoryGermany
CityKarlsruhe
Period8/31/999/3/99

Keywords

  • Adaptation strategy
  • Adaptive Kalman filters
  • Fault detection
  • Multiple models
  • Switch strategy

ASJC Scopus subject areas

  • Control and Systems Engineering

Fingerprint

Dive into the research topics of 'Fault detection on non linear systems described by multiple linearized models'. Together they form a unique fingerprint.

Cite this