Iceberg-seabed interaction evaluation in clay seabed using tree-based machine learning algorithms

IF 4.8 Q2 ENERGY & FUELS
Hamed Azimi , Hodjat Shiri , Masoud Mahdianpari
{"title":"Iceberg-seabed interaction evaluation in clay seabed using tree-based machine learning algorithms","authors":"Hamed Azimi ,&nbsp;Hodjat Shiri ,&nbsp;Masoud Mahdianpari","doi":"10.1016/j.jpse.2022.100075","DOIUrl":null,"url":null,"abstract":"<div><p>In Arctic offshore regions, the oil and gas hydrocarbons are transferred to the onshore basins through the subsea pipelines. However, the operational integrity of the subsea pipeline may be at risk of collision with traveling icebergs, which gouge the seabed in the Arctic shallow waters. Even though these sea bottom-founded structures are buried at a secure depth below the seafloor, the pipeline is still threatened by the ice scouring event extended much deeper than the ice tip due to the shear resistance of the seabed soil. Modeling the sub-gouge soil characteristics is a challenging problem that requires costly experimental and long-running finite element (FE) simulations. To overcome these challenges, in this paper, the reaction forces and sub-gouge soil deformations in clay were modeled using an advanced extra tree regression (ETR) algorithm, as a quick and cost-effective alternative for the early design phases of pipeline engineering projects. Eight ETR models, ETR 1 to ETR 8, were developed by using the input parameters governing the iceberg-seabed interaction problem. The collected data were randomly split into 70% for training the machine learning (ML) models and 30% for testing purposes. The most accurate ETR models and the most significant input parameters were identified by performing a sensitivity analysis. The comparison of the most accurate ETR models and decision tree regression (DTR), random forest regression (RFR), and gradient boosting regression (GBR) algorithms proved that the ETR models had better performance to simulate the ice keel seabed interaction in clay.</p></div>","PeriodicalId":100824,"journal":{"name":"Journal of Pipeline Science and Engineering","volume":"2 4","pages":"Article 100075"},"PeriodicalIF":4.8000,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S2667143322000476/pdfft?md5=d6a69ba337789792bf15a7b1d087c26d&pid=1-s2.0-S2667143322000476-main.pdf","citationCount":"4","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Pipeline Science and Engineering","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2667143322000476","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: 4

Abstract

In Arctic offshore regions, the oil and gas hydrocarbons are transferred to the onshore basins through the subsea pipelines. However, the operational integrity of the subsea pipeline may be at risk of collision with traveling icebergs, which gouge the seabed in the Arctic shallow waters. Even though these sea bottom-founded structures are buried at a secure depth below the seafloor, the pipeline is still threatened by the ice scouring event extended much deeper than the ice tip due to the shear resistance of the seabed soil. Modeling the sub-gouge soil characteristics is a challenging problem that requires costly experimental and long-running finite element (FE) simulations. To overcome these challenges, in this paper, the reaction forces and sub-gouge soil deformations in clay were modeled using an advanced extra tree regression (ETR) algorithm, as a quick and cost-effective alternative for the early design phases of pipeline engineering projects. Eight ETR models, ETR 1 to ETR 8, were developed by using the input parameters governing the iceberg-seabed interaction problem. The collected data were randomly split into 70% for training the machine learning (ML) models and 30% for testing purposes. The most accurate ETR models and the most significant input parameters were identified by performing a sensitivity analysis. The comparison of the most accurate ETR models and decision tree regression (DTR), random forest regression (RFR), and gradient boosting regression (GBR) algorithms proved that the ETR models had better performance to simulate the ice keel seabed interaction in clay.

Abstract Image

基于树的机器学习算法的粘土海底冰山-海底相互作用评价
在北极近海地区,石油和天然气碳氢化合物通过海底管道输送到陆上盆地。然而,海底管道的运行完整性可能会面临与移动冰山碰撞的风险,这些冰山会在北极浅水区凿出海底。尽管这些海底基础结构被埋在海底以下的安全深度,但由于海底土壤的剪切阻力,管道仍然受到比冰尖深得多的冰冲刷事件的威胁。模拟潜泥土的特性是一个具有挑战性的问题,需要昂贵的实验和长时间的有限元模拟。为了克服这些挑战,本文使用先进的额外树回归(ETR)算法对粘土中的反作用力和潜泥土变形进行了建模,作为管道工程项目早期设计阶段快速且经济有效的替代方案。利用控制冰山-海底相互作用问题的输入参数,建立了ETR 1 ~ ETR 8个ETR模型。收集的数据被随机分成70%用于训练机器学习(ML)模型,30%用于测试目的。通过灵敏度分析确定了最准确的ETR模型和最重要的输入参数。将最精确的ETR模型与决策树回归(DTR)、随机森林回归(RFR)和梯度增强回归(GBR)算法进行比较,证明ETR模型在模拟粘土冰龙骨海底相互作用方面具有更好的性能。
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