Forecasting FSW Material’s Behavior using an Artificial Intelligence-Driven Approach

Abdelhakim Dorbane, Fouzi Harrou, Ying Sun

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Scopus citations

Abstract

A flexible data-driven methodology was developed to forecast the mechanical behavior of an aluminum alloy, namely Al6061-T6, in the case of friction stir welding. Specifically, Gated recurrent unit (GRU), a deep learning model, was investigated in this study. This is the first time the GRU model has been used to forecast the stress-strain curve of a material. The major features of the GRU consist in its ability to model time-series data and rely only on historical and actual data from the investigated material. The performance of the GRU model has been demonstrated based on actual data collected by conducting uniaxial tensile testing on the base material, and friction stirred welded, both tested at a deformation speed of 10 −3 s −1 . Forecasting tensile tests results showed promising and accurate results of the GRU-driven forecasting.
Original languageEnglish (US)
Title of host publication2022 International Conference on Decision Aid Sciences and Applications (DASA)
PublisherIEEE
ISBN (Print)978-1-6654-9502-8
DOIs
StatePublished - May 2 2022

Bibliographical note

KAUST Repository Item: Exported on 2022-05-10
Acknowledged KAUST grant number(s): OSR-2019-CRG7-3800
Acknowledgements: Supported by King Abdullah University of Science and Technology (KAUST), Office of Sponsored Research (OSR) under Award No: OSR-2019-CRG7-3800.

Fingerprint

Dive into the research topics of 'Forecasting FSW Material’s Behavior using an Artificial Intelligence-Driven Approach'. Together they form a unique fingerprint.

Cite this