层合复合材料的原位疲劳预测:使用自温升数据的机器学习方法

IF 12.7 1区 材料科学 Q1 ENGINEERING, MULTIDISCIPLINARY
A.H. Mirzaei, P. Haghi
{"title":"层合复合材料的原位疲劳预测:使用自温升数据的机器学习方法","authors":"A.H. Mirzaei,&nbsp;P. Haghi","doi":"10.1016/j.compositesb.2025.112537","DOIUrl":null,"url":null,"abstract":"<div><div>This study presents a new methodology for fatigue prognosis in laminated composites at early stages of loading using in-situ thermal data. To this end, a Non-Dominated Sorting Genetic Algorithm II was coupled with an Artificial Neural Network model to optimize the model's hyperparameters. The model was then trained on carbon/epoxy laminate data, considering various stress concentration factors, loading levels, and stacking sequences, with a focus on self-temperature rise data as key input features. To enhance the training dataset, two different data augmentation methods were employed. Also, the performance of the developed model was compared to conventional regression-based machine learning algorithms, including Decision Tree and Gradient Boosting. Results showed that the model predicted the applied force on the composites with a mean absolute percentage error of 1.34 and achieved an R<sup>2</sup> score of 0.91 for fatigue life prediction on validation datasets.</div></div>","PeriodicalId":10660,"journal":{"name":"Composites Part B: Engineering","volume":"301 ","pages":"Article 112537"},"PeriodicalIF":12.7000,"publicationDate":"2025-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"In-situ fatigue prognosis in laminated composites: A machine learning approach using self-temperature rise data\",\"authors\":\"A.H. Mirzaei,&nbsp;P. Haghi\",\"doi\":\"10.1016/j.compositesb.2025.112537\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>This study presents a new methodology for fatigue prognosis in laminated composites at early stages of loading using in-situ thermal data. To this end, a Non-Dominated Sorting Genetic Algorithm II was coupled with an Artificial Neural Network model to optimize the model's hyperparameters. The model was then trained on carbon/epoxy laminate data, considering various stress concentration factors, loading levels, and stacking sequences, with a focus on self-temperature rise data as key input features. To enhance the training dataset, two different data augmentation methods were employed. Also, the performance of the developed model was compared to conventional regression-based machine learning algorithms, including Decision Tree and Gradient Boosting. Results showed that the model predicted the applied force on the composites with a mean absolute percentage error of 1.34 and achieved an R<sup>2</sup> score of 0.91 for fatigue life prediction on validation datasets.</div></div>\",\"PeriodicalId\":10660,\"journal\":{\"name\":\"Composites Part B: Engineering\",\"volume\":\"301 \",\"pages\":\"Article 112537\"},\"PeriodicalIF\":12.7000,\"publicationDate\":\"2025-04-18\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Composites Part B: Engineering\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S135983682500438X\",\"RegionNum\":1,\"RegionCategory\":\"材料科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"ENGINEERING, MULTIDISCIPLINARY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Composites Part B: Engineering","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S135983682500438X","RegionNum":1,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 0

摘要

本研究提出了一种利用原位热数据预测加载初期层合复合材料疲劳的新方法。为此,将非支配排序遗传算法II与人工神经网络模型相结合,对模型的超参数进行优化。然后对碳/环氧层压板数据进行训练,考虑各种应力集中因素、加载水平和堆叠顺序,重点关注自温升数据作为关键输入特征。为了增强训练数据集,采用了两种不同的数据增强方法。此外,将所开发模型的性能与传统的基于回归的机器学习算法(包括决策树和梯度增强)进行了比较。结果表明,该模型预测复合材料所受作用力的平均绝对百分比误差为1.34,在验证数据集上的疲劳寿命预测R2得分为0.91。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
In-situ fatigue prognosis in laminated composites: A machine learning approach using self-temperature rise data
This study presents a new methodology for fatigue prognosis in laminated composites at early stages of loading using in-situ thermal data. To this end, a Non-Dominated Sorting Genetic Algorithm II was coupled with an Artificial Neural Network model to optimize the model's hyperparameters. The model was then trained on carbon/epoxy laminate data, considering various stress concentration factors, loading levels, and stacking sequences, with a focus on self-temperature rise data as key input features. To enhance the training dataset, two different data augmentation methods were employed. Also, the performance of the developed model was compared to conventional regression-based machine learning algorithms, including Decision Tree and Gradient Boosting. Results showed that the model predicted the applied force on the composites with a mean absolute percentage error of 1.34 and achieved an R2 score of 0.91 for fatigue life prediction on validation datasets.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
Composites Part B: Engineering
Composites Part B: Engineering 工程技术-材料科学:复合
CiteScore
24.40
自引率
11.50%
发文量
784
审稿时长
21 days
期刊介绍: Composites Part B: Engineering is a journal that publishes impactful research of high quality on composite materials. This research is supported by fundamental mechanics and materials science and engineering approaches. The targeted research can cover a wide range of length scales, ranging from nano to micro and meso, and even to the full product and structure level. The journal specifically focuses on engineering applications that involve high performance composites. These applications can range from low volume and high cost to high volume and low cost composite development. The main goal of the journal is to provide a platform for the prompt publication of original and high quality research. The emphasis is on design, development, modeling, validation, and manufacturing of engineering details and concepts. The journal welcomes both basic research papers and proposals for review articles. Authors are encouraged to address challenges across various application areas. These areas include, but are not limited to, aerospace, automotive, and other surface transportation. The journal also covers energy-related applications, with a focus on renewable energy. Other application areas include infrastructure, off-shore and maritime projects, health care technology, and recreational products.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信