• Online only
  • New
First page ICMFF14-2025-51
search
  • First page ICMFF14-2025-51

On the life estimation of wires using non-local machine learning methodology

This work aims to evaluate the generalization capability of a fretting fatigue life prediction methodology based on a hybrid data-driven non-local multiaxial approach. More specifically, this model is tested using data from overhead conductor wires subjected to fretting fatigue. These wires were obtained from ACSR Tern (1120 Al alloy) and AAAC 900 (6201 Al alloy) conductors and subjected to a normal constant load and constant amplitude bulk fatigue load and also to a variable amplitude bulk fatigue load. The finite element method is used to compute the time varying stress field under the contact. The fatigue modelling challenge proposed in this work is not only related to the new material and complex loading conditions, never seen by the model, but also to the geometric effects, which…

€25.00
VAT included
Quantity

 

Datenschutzbedingungen (bearbeiten im Modul "Kundenvorteile")

 

Lieferbedingungen (bearbeiten im Modul "Kundenvorteile")

 

Rücksendebedingungen (bearbeiten im Modul "Kundenvorteile")

This work aims to evaluate the generalization capability of a fretting fatigue life prediction methodology based on a hybrid data-driven non-local multiaxial approach. More specifically, this model is tested using data from overhead conductor wires subjected to fretting fatigue. These wires were obtained from ACSR Tern (1120 Al alloy) and AAAC 900 (6201 Al alloy) conductors and subjected to a normal constant load and constant amplitude bulk fatigue load and also to a variable amplitude bulk fatigue load. The finite element method is used to compute the time varying stress field under the contact. The fatigue modelling challenge proposed in this work is not only related to the new material and complex loading conditions, never seen by the model, but also to the geometric effects, which have completely different scales and contact shapes from those the model was trained on. The results obtained showed that the proposed data-driven model was successful, providing life estimates with a factor of three band for the majority of the available data.

Reference
ICMFF14-2025-51

Title
On the life estimation of wires using non-local machine learning methodology
Author(s)
G. A. Brito Oliveira, I. M. Matos, R. A. Cardoso, J. A. Araújo
DOI
10.48447/ICMFF14-2025-51
Event
14. International Conference on Multiaxial Fatigue and Fracture 2025
Year of publication
2025
Publication type
conference paper (PDF)
Language
English
Keywords
Fretting fatigue,Data-driven models,Overhead conductors,Life prediction