RUL Prediction for Turbine Disc Based on High-Order Particle Filtering

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

Turbine disc is one of the most crucial parts in an aero-engine and it is also regarded as the vulnerable part which suffers from damage frequently. Remaining useful life (RUL) prediction technology is a useful tool to provide failure warnings and increase safety. For turbine disc, however, it is still a great challenge. This paper proposes a high-order particle filtering (HOPF)-based approach for RUL prediction of turbine disc. First, a turbine disc health indicator is constructed based on the unbalance vibration analysis. Then, an improved double exponential model is developed for turbine disc degradation modeling. Next, model updating and RUL prediction are carried out by the HOPF algorithm. The proposed approach is finally validated using the experimental data of a turbine disc. Satisfactory results demonstrate that the proposed approach performs well on turbine disc RUL prediction.

Original languageEnglish
Title of host publicationInternational Conference on Sensing, Measurement and Data Analytics in the Era of Artificial Intelligence, ICSMD 2020 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages215-220
Number of pages6
ISBN (Electronic)9781728192772
DOIs
StatePublished - 15 Oct 2020
Event1st International Conference on Sensing, Measurement and Data Analytics in the Era of Artificial Intelligence, ICSMD 2020 - Xi'an, China
Duration: 15 Oct 202017 Oct 2020

Publication series

NameInternational Conference on Sensing, Measurement and Data Analytics in the Era of Artificial Intelligence, ICSMD 2020 - Proceedings

Conference

Conference1st International Conference on Sensing, Measurement and Data Analytics in the Era of Artificial Intelligence, ICSMD 2020
Country/TerritoryChina
CityXi'an
Period15/10/2017/10/20

Keywords

  • degradation modeling
  • high-order particle filtering
  • remaining useful life
  • turbine disc

Fingerprint

Dive into the research topics of 'RUL Prediction for Turbine Disc Based on High-Order Particle Filtering'. Together they form a unique fingerprint.

Cite this