Using a U-Net trained on 549 thermo-mechanical simulations, the authors generate full-field residual stress maps of friction-stir processed A380 aluminum from nine sparse point measurements, achieving R2=0.6 against 120 experimental ESPI points.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
A Machine Learning Approach to Generate Residual Stress Distributions using Sparse Characterization Data in Friction-Stir Processed Parts
Using a U-Net trained on 549 thermo-mechanical simulations, the authors generate full-field residual stress maps of friction-stir processed A380 aluminum from nine sparse point measurements, achieving R2=0.6 against 120 experimental ESPI points.