Artificial neural network models of burst strength for thin-wall pipelines

IF 4.8 Q2 ENERGY & FUELS
Xian-Kui Zhu, William R. Johnson, Robert Sindelar, Bruce Wiersma
{"title":"Artificial neural network models of burst strength for thin-wall pipelines","authors":"Xian-Kui Zhu,&nbsp;William R. Johnson,&nbsp;Robert Sindelar,&nbsp;Bruce Wiersma","doi":"10.1016/j.jpse.2022.100090","DOIUrl":null,"url":null,"abstract":"<div><p>Burst strength is critical to pipeline design, operation, and integrity management. The Barlow formula with the ultimate tensile stress (UTS) was often used to estimate burst strength of line pipes. To consider the plastic flow effect of pipeline steels, an average shear stress yield criterion was proposed, and the associated Zhu-Leis solution of burst strength was obtained for defect-free pipelines in term of UTS and strain hardening exponent, <em>n</em>, of materials. The Zhu-Leis solution was validated by more than 100 burst tests for various pipeline steels. The Zhu-Leis solution, when normalized by the Barlow strength, is a function of strain hardening rate, <em>n</em>, only. In contrast, experimental burst test data, when normalized by the Barlow strength, are a weak function of UTS and pipe diameter to thickness ratio <em>D</em>/<em>t</em>, in addition to be the function of <em>n</em>. Due to the difficulty of obtaining a closed-from solution using three-parameter regression, machine learning technology is adopted to develop alternative models of burst strength based on a large database of full-scale burst tests. In comparing to the regression, the machine learning method works well for both single and multiple parameters by introducing an artificial neural network (ANN), activation functions and learning algorithm for the network to learn from training data and then make predictions. Three ANN models were developed in this paper for predicting the burst strength of defect-free pipelines. Model 1 has one input variable and one hidden layer with three neurons; Model 2 has three input variables and one hidden layer with five neurons; and Model 3 has three input variables and two hidden layers with three neurons for the first hidden layer and two neurons for the second hidden layer. These three ANN models were then validated by the full-scale test data and evaluated through comparison with the Zhu-Leis solution and the linear regression results. On this basis, the best ANN model was recommended.</p></div>","PeriodicalId":100824,"journal":{"name":"Journal of Pipeline Science and Engineering","volume":"2 4","pages":"Article 100090"},"PeriodicalIF":4.8000,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S2667143322000622/pdfft?md5=11c58be76c0df0bbd93eb84625a6c0e6&pid=1-s2.0-S2667143322000622-main.pdf","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Pipeline Science and Engineering","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2667143322000622","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENERGY & FUELS","Score":null,"Total":0}
引用次数: 3

Abstract

Burst strength is critical to pipeline design, operation, and integrity management. The Barlow formula with the ultimate tensile stress (UTS) was often used to estimate burst strength of line pipes. To consider the plastic flow effect of pipeline steels, an average shear stress yield criterion was proposed, and the associated Zhu-Leis solution of burst strength was obtained for defect-free pipelines in term of UTS and strain hardening exponent, n, of materials. The Zhu-Leis solution was validated by more than 100 burst tests for various pipeline steels. The Zhu-Leis solution, when normalized by the Barlow strength, is a function of strain hardening rate, n, only. In contrast, experimental burst test data, when normalized by the Barlow strength, are a weak function of UTS and pipe diameter to thickness ratio D/t, in addition to be the function of n. Due to the difficulty of obtaining a closed-from solution using three-parameter regression, machine learning technology is adopted to develop alternative models of burst strength based on a large database of full-scale burst tests. In comparing to the regression, the machine learning method works well for both single and multiple parameters by introducing an artificial neural network (ANN), activation functions and learning algorithm for the network to learn from training data and then make predictions. Three ANN models were developed in this paper for predicting the burst strength of defect-free pipelines. Model 1 has one input variable and one hidden layer with three neurons; Model 2 has three input variables and one hidden layer with five neurons; and Model 3 has three input variables and two hidden layers with three neurons for the first hidden layer and two neurons for the second hidden layer. These three ANN models were then validated by the full-scale test data and evaluated through comparison with the Zhu-Leis solution and the linear regression results. On this basis, the best ANN model was recommended.

薄壁管道爆破强度的人工神经网络模型
爆裂强度对管道设计、运行和完整性管理至关重要。带极限拉应力(UTS)的巴洛公式常用于估算管线的爆破强度。为考虑管道钢的塑性流动效应,提出了平均剪切应力屈服准则,并以材料的应变硬化指数n为变量,建立了无缺陷管道爆裂强度的相关Zhu-Leis解。通过对各种管线钢进行100多次爆破试验,验证了Zhu-Leis解决方案的有效性。当用Barlow强度归一化时,Zhu-Leis解仅是应变硬化速率n的函数。相比之下,实验爆破试验数据经Barlow强度归一化后,除了是n的函数外,还是UTS和管径厚比D/t的弱函数。由于使用三参数回归难以获得封闭解,因此采用机器学习技术,基于大型全尺寸爆破试验数据库,开发了爆破强度的替代模型。与回归方法相比,机器学习方法通过引入人工神经网络(ANN)、激活函数和学习算法,使网络从训练数据中学习,然后做出预测,在单参数和多参数下都能很好地工作。本文建立了三种人工神经网络模型,用于预测无缺陷管道的破裂强度。模型1有一个输入变量和一个包含三个神经元的隐藏层;模型2有3个输入变量和1个包含5个神经元的隐藏层;模型3有三个输入变量和两个隐藏层,第一隐藏层有三个神经元,第二隐藏层有两个神经元。通过全尺寸试验数据对这三种人工神经网络模型进行验证,并与朱磊解和线性回归结果进行对比评价。在此基础上,推荐了最佳的人工神经网络模型。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
CiteScore
7.50
自引率
0.00%
发文量
0
×
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