A data-driven health assessment method for electromechanical actuation systems

Aitor Isturiz, Jose Ignacio Aizpurua, Fidel E. Hernández, Aitzol Iturrospe, Eñaut Muxika, Javier Viñals

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

44 Downloads (Pure)

Abstract

The design of health assessment applications for the electromechanical actuation system of the aircraft is a challenging task. Physics-of-failure models involve non-linear complex equations which are further complicated at the system-level. Data-driven techniques require run-to-failure tests to predict the remaining useful life. However, components are not allowed to run until failure in the aerospace engineering arena. Besides, when adding new monitoring elements for an improved health assessment, the airliner sets constraints due to the increased cost and weight. In this context, the health assessment of the electromechanical actuation system is a challenging task. In this paper we propose a data-driven approach which estimates the health state of the system without run-to-failure data and limited health information. The approach combines basic reliability theory with Bayesian concepts and obtained results show the feasibility of the technique for asset health assessment.
Original languageEnglish
Title of host publicationProceedings of the Third European Conference of the Prognostics and Health Management Society 2016
EditorsIoana Eballard, Anibal Bregon
Place of PublicationBilbao
Pages686-694
Number of pages8
Publication statusPublished - 8 Jul 2016
EventThird European Conference of the Prognostics and Health Management Society 2016 - Bilbao, Spain
Duration: 5 Jul 20168 Jul 2016

Conference

ConferenceThird European Conference of the Prognostics and Health Management Society 2016
CountrySpain
CityBilbao
Period5/07/168/07/16

Keywords

  • health assessment applications
  • data-driven techniques
  • reliability theory
  • Bayesian

Fingerprint Dive into the research topics of 'A data-driven health assessment method for electromechanical actuation systems'. Together they form a unique fingerprint.

Cite this