Fatigue life prediction using recurrent neural networks and critical plane approach

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21220%2F25%3A00362459" target="_blank" >RIV/68407700:21220/25:00362459 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
Fatigue life prediction using recurrent neural networks and critical plane approach
Original language description
Lately the machine learning methodology has been used in fatigue life prediction models as an attempt to overcome limitations of the classical approach. In this paper, a fatigue life prediction method for components under cyclic plane stress loading is proposed. This method combines the critical plane approach with the capabilities of artificial neural networks. Eleven different cases of cyclic axial-torsion loadings of the 42CrMo4 steel are utilized for the demonstration purpose in this study. The model achieved great prediction accuracy on this data set, however its ability to extrapolate is very unstable.
Czech name
—
Czech description
—

Project
<a href="/en/project/GA21-06645S" target="_blank" >GA21-06645S: Life assessment of mechanical components under multiaxial thermo-mechanical loading with variable amplitude</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Publication year
2025
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection
Proceedings of the 60th Annual Conference on Experimental Stress Analysis
ISBN
978-80-01-07400-8
ISSN
—
e-ISSN
—
Number of pages
6
Pages from-to
101-106
Publisher name
ČVUT v Praze, Fakulta strojní, Ústav mechaniky, biomechaniky a mechatroniky
Place of publication
Praha
Event location
Praha
Event date
Jun 6, 2022
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—

Similar results(10)