Fatigue life prediction of composite materials using strain distribution images and a deep convolution neural network

Abstract The damage process of composite materials, such as short fiber-reinforced plastics (SFRP), is complex. Therefore, it is necessary to accurately represent the damage process in fatigue life prediction. Herein, fatigue life prediction was conducted by combining the digital image correlation m...

Full description

Saved in:
Bibliographic Details
Main Authors: Yuta Mizuno, Atsushi Hosoi, Hiroyuki Koshita, Dai Tsunoda, Hiroyuki Kawada
Format: Article
Language:English
Published: Nature Portfolio 2024-10-01
Series:Scientific Reports
Online Access:https://doi.org/10.1038/s41598-024-75884-2
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850204455666974720
author Yuta Mizuno
Atsushi Hosoi
Hiroyuki Koshita
Dai Tsunoda
Hiroyuki Kawada
author_facet Yuta Mizuno
Atsushi Hosoi
Hiroyuki Koshita
Dai Tsunoda
Hiroyuki Kawada
author_sort Yuta Mizuno
collection DOAJ
description Abstract The damage process of composite materials, such as short fiber-reinforced plastics (SFRP), is complex. Therefore, it is necessary to accurately represent the damage process in fatigue life prediction. Herein, fatigue life prediction was conducted by combining the digital image correlation method, which is a non-destructive testing technique, with a convolutional neural network (CNN), using Xception as the network architecture. High prediction accuracy was obtained when training and testing were performed on the same SFRP specimens. In contrast, using different specimens for training and testing resulted in lower accuracy. This issue may be improved by increasing the number of specimens. The regions of interest in the model were visualized by Gradient-weighted Class Activation Mapping. Notably, the model indicated the breaking point as the region of interest from the early stages of the test. The breaking point was identified at an earlier stage by the CNN than by visual inspection, demonstrating the potential for a new method of damage observation.
format Article
id doaj-art-bfeed30e5eca46b0a82a061accc0e979
institution OA Journals
issn 2045-2322
language English
publishDate 2024-10-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-bfeed30e5eca46b0a82a061accc0e9792025-08-20T02:11:17ZengNature PortfolioScientific Reports2045-23222024-10-0114111410.1038/s41598-024-75884-2Fatigue life prediction of composite materials using strain distribution images and a deep convolution neural networkYuta Mizuno0Atsushi Hosoi1Hiroyuki Koshita2Dai Tsunoda3Hiroyuki Kawada4Department of Applied Mechanics and Aerospace Engineering, Waseda UniversityDepartment of Applied Mechanics and Aerospace Engineering, Waseda UniversityMobility Business Headquarters Mobility R&D Center, Resonac CorporationMobility Business Headquarters Mobility R&D Center, Resonac CorporationDepartment of Applied Mechanics and Aerospace Engineering, Waseda UniversityAbstract The damage process of composite materials, such as short fiber-reinforced plastics (SFRP), is complex. Therefore, it is necessary to accurately represent the damage process in fatigue life prediction. Herein, fatigue life prediction was conducted by combining the digital image correlation method, which is a non-destructive testing technique, with a convolutional neural network (CNN), using Xception as the network architecture. High prediction accuracy was obtained when training and testing were performed on the same SFRP specimens. In contrast, using different specimens for training and testing resulted in lower accuracy. This issue may be improved by increasing the number of specimens. The regions of interest in the model were visualized by Gradient-weighted Class Activation Mapping. Notably, the model indicated the breaking point as the region of interest from the early stages of the test. The breaking point was identified at an earlier stage by the CNN than by visual inspection, demonstrating the potential for a new method of damage observation.https://doi.org/10.1038/s41598-024-75884-2
spellingShingle Yuta Mizuno
Atsushi Hosoi
Hiroyuki Koshita
Dai Tsunoda
Hiroyuki Kawada
Fatigue life prediction of composite materials using strain distribution images and a deep convolution neural network
Scientific Reports
title Fatigue life prediction of composite materials using strain distribution images and a deep convolution neural network
title_full Fatigue life prediction of composite materials using strain distribution images and a deep convolution neural network
title_fullStr Fatigue life prediction of composite materials using strain distribution images and a deep convolution neural network
title_full_unstemmed Fatigue life prediction of composite materials using strain distribution images and a deep convolution neural network
title_short Fatigue life prediction of composite materials using strain distribution images and a deep convolution neural network
title_sort fatigue life prediction of composite materials using strain distribution images and a deep convolution neural network
url https://doi.org/10.1038/s41598-024-75884-2
work_keys_str_mv AT yutamizuno fatiguelifepredictionofcompositematerialsusingstraindistributionimagesandadeepconvolutionneuralnetwork
AT atsushihosoi fatiguelifepredictionofcompositematerialsusingstraindistributionimagesandadeepconvolutionneuralnetwork
AT hiroyukikoshita fatiguelifepredictionofcompositematerialsusingstraindistributionimagesandadeepconvolutionneuralnetwork
AT daitsunoda fatiguelifepredictionofcompositematerialsusingstraindistributionimagesandadeepconvolutionneuralnetwork
AT hiroyukikawada fatiguelifepredictionofcompositematerialsusingstraindistributionimagesandadeepconvolutionneuralnetwork