Enabling High-Degree-of-Freedom Thermal Engineering Calculations via Lightweight Machine Learning

IF 4.3 3区 材料科学 Q1 ENGINEERING, ELECTRICAL & ELECTRONIC
Yajing Tian, Yuyang Wang, Shasha Yin, Jia Lu, Yu Hu
{"title":"Enabling High-Degree-of-Freedom Thermal Engineering Calculations via Lightweight Machine Learning","authors":"Yajing Tian, Yuyang Wang, Shasha Yin, Jia Lu, Yu Hu","doi":"10.3390/en17163916","DOIUrl":null,"url":null,"abstract":"U-tube steam generators (UTSGs) are crucial in nuclear power plants, serving as the interface between the primary and secondary coolant loops. UTSGs ensure efficient heat exchange, operational stability, and safety, directly impacting the plant’s efficiency and reliability. Existing UTSG models have fixed structures, which can only be used when certain parameters are given as model input. Such constraints hinder their ability to accommodate the diverse operating conditions, where input and output parameters can vary significantly. To address this challenge, we propose a machine learning-based method for developing a high-degree-of-freedom UTSG thermal model. The most notable feature of this approach is its capacity to flexibly interchange input and output parameters. By adopting comprehensive parameter sensitivity analysis, the most efficient method for training dataset generation is determined. Leveraging a lightweight machine learning method, the prediction accuracy for all UTSG parameters is improved to within 2.1%. The flexibility of the proposed machine learning approach ensures that the UTSG model can accommodate any type of parameter input without extensive reconfiguration of the model structure, thereby enhancing its applicability and robustness in real-world applications.","PeriodicalId":3,"journal":{"name":"ACS Applied Electronic Materials","volume":"10 3","pages":""},"PeriodicalIF":4.3000,"publicationDate":"2024-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"ACS Applied Electronic Materials","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.3390/en17163916","RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}
引用次数: 0

Abstract

U-tube steam generators (UTSGs) are crucial in nuclear power plants, serving as the interface between the primary and secondary coolant loops. UTSGs ensure efficient heat exchange, operational stability, and safety, directly impacting the plant’s efficiency and reliability. Existing UTSG models have fixed structures, which can only be used when certain parameters are given as model input. Such constraints hinder their ability to accommodate the diverse operating conditions, where input and output parameters can vary significantly. To address this challenge, we propose a machine learning-based method for developing a high-degree-of-freedom UTSG thermal model. The most notable feature of this approach is its capacity to flexibly interchange input and output parameters. By adopting comprehensive parameter sensitivity analysis, the most efficient method for training dataset generation is determined. Leveraging a lightweight machine learning method, the prediction accuracy for all UTSG parameters is improved to within 2.1%. The flexibility of the proposed machine learning approach ensures that the UTSG model can accommodate any type of parameter input without extensive reconfiguration of the model structure, thereby enhancing its applicability and robustness in real-world applications.
通过轻量级机器学习实现高自由度热工计算
U 型管蒸汽发生器(UTSG)是核电站的关键设备,是一次冷却剂回路和二次冷却剂回路之间的接口。UTSG 确保高效的热交换、运行稳定性和安全性,直接影响核电站的效率和可靠性。现有的 UTSG 模型结构固定,只有在输入特定参数时才能使用。这种限制妨碍了它们适应不同运行条件的能力,因为在这些条件下,输入和输出参数可能会有很大的变化。为了应对这一挑战,我们提出了一种基于机器学习的方法,用于开发高自由度 UTSG 热模型。这种方法最显著的特点是能够灵活地互换输入和输出参数。通过全面的参数敏感性分析,确定了最有效的训练数据集生成方法。利用轻量级机器学习方法,UTSG 所有参数的预测精度提高到了 2.1% 以内。所提出的机器学习方法的灵活性确保了UTSG 模型能够适应任何类型的参数输入,而无需对模型结构进行大量重新配置,从而增强了其在实际应用中的适用性和稳健性。
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
7.20
自引率
4.30%
发文量
567
期刊介绍: ACS Applied Electronic Materials is an interdisciplinary journal publishing original research covering all aspects of electronic materials. The journal is devoted to reports of new and original experimental and theoretical research of an applied nature that integrate knowledge in the areas of materials science, engineering, optics, physics, and chemistry into important applications of electronic materials. Sample research topics that span the journal's scope are inorganic, organic, ionic and polymeric materials with properties that include conducting, semiconducting, superconducting, insulating, dielectric, magnetic, optoelectronic, piezoelectric, ferroelectric and thermoelectric. Indexed/​Abstracted: Web of Science SCIE Scopus CAS INSPEC Portico
×
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