Augmentation method of fatigue data of welded structures based on physics-informed CTGAN

Variable amplitude loading is frequently applied to welded structures in practical engineering, and fatigue failure is a prevalent problem. In recent years, machine learning is a useful technique for predicting fatigue life. However, it is challenging to acquire a sufficient number of reliable train...

Full description

Saved in:
Bibliographic Details
Main Authors: Xinyu Cao, Li Zou, Chen Lu
Format: Article
Language:English
Published: Gruppo Italiano Frattura 2025-04-01
Series:Fracture and Structural Integrity
Subjects:
Online Access:https://www.fracturae.com/index.php/fis/article/view/5332/4195
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849728186879836160
author Xinyu Cao
Li Zou
Chen Lu
author_facet Xinyu Cao
Li Zou
Chen Lu
author_sort Xinyu Cao
collection DOAJ
description Variable amplitude loading is frequently applied to welded structures in practical engineering, and fatigue failure is a prevalent problem. In recent years, machine learning is a useful technique for predicting fatigue life. However, it is challenging to acquire a sufficient number of reliable training samples for fatigue tests under variable amplitude loading. The machine learning models' accuracy and generalization capabilities are impacted by this. This work introduces a novel data augmentation approach utilizing physics-informed Generative Adversarial Networks (GAN). Data augmentation is accomplished by incorporating the traditional damage model - Ye model as constraints within the loss function of the Conditional Tabular GAN (CTGAN). The method combines physical laws of damage with CTGAN, which makes generated fatigue data conform to physical characteristics under two-step loading. Then the impact of generated data on model performance is evaluated on four machine learning models and compared to traditional damage models. The experimental results show that generated fatigue data helps machine learning models to get better prediction results compared with traditional models and unaugmented machine learning models, which significantly enhancing the precision of fatigue life predictions.
format Article
id doaj-art-73b42708bd074bb3990aa3d8e6d97592
institution DOAJ
issn 1971-8993
language English
publishDate 2025-04-01
publisher Gruppo Italiano Frattura
record_format Article
series Fracture and Structural Integrity
spelling doaj-art-73b42708bd074bb3990aa3d8e6d975922025-08-20T03:09:38ZengGruppo Italiano FratturaFracture and Structural Integrity1971-89932025-04-01197216217810.3221/IGF-ESIS.72.1210.3221/IGF-ESIS.72.12Augmentation method of fatigue data of welded structures based on physics-informed CTGANXinyu CaoLi ZouChen LuVariable amplitude loading is frequently applied to welded structures in practical engineering, and fatigue failure is a prevalent problem. In recent years, machine learning is a useful technique for predicting fatigue life. However, it is challenging to acquire a sufficient number of reliable training samples for fatigue tests under variable amplitude loading. The machine learning models' accuracy and generalization capabilities are impacted by this. This work introduces a novel data augmentation approach utilizing physics-informed Generative Adversarial Networks (GAN). Data augmentation is accomplished by incorporating the traditional damage model - Ye model as constraints within the loss function of the Conditional Tabular GAN (CTGAN). The method combines physical laws of damage with CTGAN, which makes generated fatigue data conform to physical characteristics under two-step loading. Then the impact of generated data on model performance is evaluated on four machine learning models and compared to traditional damage models. The experimental results show that generated fatigue data helps machine learning models to get better prediction results compared with traditional models and unaugmented machine learning models, which significantly enhancing the precision of fatigue life predictions.https://www.fracturae.com/index.php/fis/article/view/5332/4195fatigue life predictiondata augmentationctganvariable amplitude loading
spellingShingle Xinyu Cao
Li Zou
Chen Lu
Augmentation method of fatigue data of welded structures based on physics-informed CTGAN
Fracture and Structural Integrity
fatigue life prediction
data augmentation
ctgan
variable amplitude loading
title Augmentation method of fatigue data of welded structures based on physics-informed CTGAN
title_full Augmentation method of fatigue data of welded structures based on physics-informed CTGAN
title_fullStr Augmentation method of fatigue data of welded structures based on physics-informed CTGAN
title_full_unstemmed Augmentation method of fatigue data of welded structures based on physics-informed CTGAN
title_short Augmentation method of fatigue data of welded structures based on physics-informed CTGAN
title_sort augmentation method of fatigue data of welded structures based on physics informed ctgan
topic fatigue life prediction
data augmentation
ctgan
variable amplitude loading
url https://www.fracturae.com/index.php/fis/article/view/5332/4195
work_keys_str_mv AT xinyucao augmentationmethodoffatiguedataofweldedstructuresbasedonphysicsinformedctgan
AT lizou augmentationmethodoffatiguedataofweldedstructuresbasedonphysicsinformedctgan
AT chenlu augmentationmethodoffatiguedataofweldedstructuresbasedonphysicsinformedctgan