Thermomechanical fatigue life due to scatter in constitutive parameters

Leidermark, Daniel; Rouse, James; Engel, Benedikt; Hyde, Christopher; Pattison, Stephen; Stekovic, Svjetlana

Abstract

For critical component application, such as aerospace turbine rotors, it is imperative to be able to make accurate in-service material behaviour and component life predictions for both design and monitoring of component life. The development of such predictive capability is dependent on the quality of the experimental data from which the material parameters are derived. This paper shows the effect that scatter which may be present within experimental data, manifesting itself within the constitutive parameters derived from this data, has on the resulting fatigue crack initiation life of the nickel-based superalloy RR1000. Industrial relevance was added to this investigation by the use of flight representative thermomechanical fatigue loading cycles and state of the art material behaviour and fatigue crack initiation models used within the finite element simulations conducted. The effect of the ‘scatter in’ to the modelling approach on the outcoming predictions is made via a Monte-Carlo analysis. This analysis consisted of running the same simulation several times, but with the experimentally determined and validated ‘baseline’ constitutive parameters varied via correction factors built into the model, for each run via a singular value decomposition procedure. It was found that small ‘scatter in’ has only a very localised ‘scatter out’ effect on the crack initiation predictions under the flight representative loading.

Citation

Leidermark, D., Rouse, J., Engel, B., Hyde, C., Pattison, S., & Stekovic, S. (2021). Thermomechanical fatigue life due to scatter in constitutive parameters. Computational Materials Science, 198, Article 110690. https://doi.org/10.1016/j.commatsci.2021.110690

• Journal Article Type: Article
• Acceptance Date: Jun 25, 2021
• Online Publication Date: Jul 8, 2021
• Publication Date: 2021-10
• Deposit Date: Nov 10, 2021
• Publicly Available Date: Nov 18, 2021
• Journal: Computational Materials Science
• Print ISSN: 0927-0256
• Publisher: Elsevier
• Peer Reviewed: Peer Reviewed
• Volume: 198
• Article Number: 110690
• DOI: https://doi.org/10.1016/j.commatsci.2021.110690
• Public URL: https://nottingham-repository.worktribe.com/output/5792653
• Publisher URL: https://www.sciencedirect.com/science/article/pii/S0927025621004171?via%3Dihub