An AK‐MCS‐based probabilistic fatigue life prediction framework for turbine disc with a mean stress correction model

The turbine disc plays a crucial role in aerospace engine, significantly influencing both their fatigue life and overall reliability. Due to the substantial uncertainties present, devising a method for the probabilistic analysis of turbine disc fatigue life becomes crucial. This paper introduces a n...

Full description

Saved in:
Bibliographic Details
Published inQuality and reliability engineering international Vol. 40; no. 6; pp. 3238 - 3252
Main Authors Wang, Yun, Li, Yan‐Feng, Huang, Hong‐Zhong, Bai, Song
Format Journal Article
LanguageEnglish
Published Bognor Regis Wiley Subscription Services, Inc 01.10.2024
Subjects
Aerospace engines
AK‐MCS
Algorithms
Fatigue life
fatigue life prediction
Finite element method
Kriging model
Life prediction
Monte Carlo simulation
Probabilistic analysis
reliability
turbine disc
Turbine disks
Uncertainty analysis
Online AccessGet full text

Cover

More Information
Summary:The turbine disc plays a crucial role in aerospace engine, significantly influencing both their fatigue life and overall reliability. Due to the substantial uncertainties present, devising a method for the probabilistic analysis of turbine disc fatigue life becomes crucial. This paper introduces a novel framework that combines adaptive Kriging Monte Carlo simulation (AK‐MCS) with a mean stress correction model to evaluate fatigue life distribution effectively. AK‐MCS, stands out as an algorithm that constructs a Kriging model, effectively reducing the need for an extensive finite element analysis. Furthermore, a mean stress correction model is proposed for the Masson‐Coffin equation tailored for fatigue life prediction of GH4133 disc material. This model has been empirically validated to be effective. The presented framework for probabilistic fatigue life analysis not only holds considerable engineering value but also offers innovative approaches for tackling analogous challenges in the field.
ISSN:0748-8017
1099-1638
DOI:10.1002/qre.3556