用空间相关聚合方法增强非侵入式降阶模型

IF 2.3 3区 工程技术 Q2 MECHANICS
Anna Ivagnes, Niccolò Tonicello, Paola Cinnella, Gianluigi Rozza
{"title":"用空间相关聚合方法增强非侵入式降阶模型","authors":"Anna Ivagnes, Niccolò Tonicello, Paola Cinnella, Gianluigi Rozza","doi":"10.1007/s00707-024-04007-9","DOIUrl":null,"url":null,"abstract":"<p>In this manuscript, we combine non-intrusive reduced-order models (ROMs) with space-dependent aggregation techniques to build a <i>mixed-ROM</i>, able to accurately capture the flow dynamics in different physical settings. The flow prediction obtained using the <i>mixed</i> formulation is derived from a convex combination of the predictions of several previously trained reduced-order models (ROMs), with each model assigned a space-dependent weight. The ROMs incorporated in the <i>mixed</i> model utilize different <i>reduction</i> methods, such as proper orthogonal decomposition and autoencoders, and various <i>approximation</i> techniques, including radial basis function interpolation (RBF), Gaussian process regression, and feed-forward artificial neural networks. Each model’s contribution is given higher weights in regions where it performs best and lower weights where its accuracy is lower compared to the other models. Additionally, a random forest regression technique is used to determine the weights for previously unseen conditions. The performance of the aggregated model is assessed through two test cases: the 2D flow past a NACA 4412 airfoil at a 5-degree angle of attack, with the Reynolds number ranging between <span>\\(1 \\times 10^{5}\\)</span> and <span>\\(1 \\times 10^{6}\\)</span>, and a transonic flow over a NACA 0012 airfoil, with the angle of attack as the varying parameter. In both scenarios, the mixed-ROM demonstrated improved accuracy compared to each individual ROM technique, while providing an estimate for the predictive uncertainty.</p>","PeriodicalId":456,"journal":{"name":"Acta Mechanica","volume":"4 1","pages":""},"PeriodicalIF":2.3000,"publicationDate":"2024-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Enhancing non-intrusive reduced-order models with space-dependent aggregation methods\",\"authors\":\"Anna Ivagnes, Niccolò Tonicello, Paola Cinnella, Gianluigi Rozza\",\"doi\":\"10.1007/s00707-024-04007-9\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p>In this manuscript, we combine non-intrusive reduced-order models (ROMs) with space-dependent aggregation techniques to build a <i>mixed-ROM</i>, able to accurately capture the flow dynamics in different physical settings. The flow prediction obtained using the <i>mixed</i> formulation is derived from a convex combination of the predictions of several previously trained reduced-order models (ROMs), with each model assigned a space-dependent weight. The ROMs incorporated in the <i>mixed</i> model utilize different <i>reduction</i> methods, such as proper orthogonal decomposition and autoencoders, and various <i>approximation</i> techniques, including radial basis function interpolation (RBF), Gaussian process regression, and feed-forward artificial neural networks. Each model’s contribution is given higher weights in regions where it performs best and lower weights where its accuracy is lower compared to the other models. Additionally, a random forest regression technique is used to determine the weights for previously unseen conditions. The performance of the aggregated model is assessed through two test cases: the 2D flow past a NACA 4412 airfoil at a 5-degree angle of attack, with the Reynolds number ranging between <span>\\\\(1 \\\\times 10^{5}\\\\)</span> and <span>\\\\(1 \\\\times 10^{6}\\\\)</span>, and a transonic flow over a NACA 0012 airfoil, with the angle of attack as the varying parameter. In both scenarios, the mixed-ROM demonstrated improved accuracy compared to each individual ROM technique, while providing an estimate for the predictive uncertainty.</p>\",\"PeriodicalId\":456,\"journal\":{\"name\":\"Acta Mechanica\",\"volume\":\"4 1\",\"pages\":\"\"},\"PeriodicalIF\":2.3000,\"publicationDate\":\"2024-07-22\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Acta Mechanica\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://doi.org/10.1007/s00707-024-04007-9\",\"RegionNum\":3,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"MECHANICS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Acta Mechanica","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1007/s00707-024-04007-9","RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"MECHANICS","Score":null,"Total":0}
引用次数: 0

摘要

在本手稿中,我们将非侵入式降阶模型(ROM)与空间相关聚合技术相结合,建立了一种混合 ROM,能够准确捕捉不同物理环境下的流动动态。使用混合模型获得的水流预测结果是由多个先前训练过的降阶模型(ROMs)的预测结果凸组合而成的,每个模型都有一个与空间相关的权重。混合模型中的 ROM 采用了不同的简化方法,如适当的正交分解和自动编码器,以及各种近似技术,包括径向基函数插值(RBF)、高斯过程回归和前馈人工神经网络。与其他模型相比,每个模型的贡献在其表现最佳的区域被赋予较高的权重,而在其准确性较低的区域则被赋予较低的权重。此外,还使用随机森林回归技术来确定先前未见条件的权重。通过两个测试案例评估了聚合模型的性能:以 5 度攻角流过 NACA 4412 翼面的二维气流,雷诺数介于(1 /times 10^{5}\) 和(1 /times 10^{6}\) 之间;以及以攻角作为变化参数流过 NACA 0012 翼面的跨音速气流。在这两种情况下,与每种单独的 ROM 技术相比,混合 ROM 的精度都有所提高,同时提供了预测不确定性的估计值。
本文章由计算机程序翻译,如有差异,请以英文原文为准。

Enhancing non-intrusive reduced-order models with space-dependent aggregation methods

Enhancing non-intrusive reduced-order models with space-dependent aggregation methods

In this manuscript, we combine non-intrusive reduced-order models (ROMs) with space-dependent aggregation techniques to build a mixed-ROM, able to accurately capture the flow dynamics in different physical settings. The flow prediction obtained using the mixed formulation is derived from a convex combination of the predictions of several previously trained reduced-order models (ROMs), with each model assigned a space-dependent weight. The ROMs incorporated in the mixed model utilize different reduction methods, such as proper orthogonal decomposition and autoencoders, and various approximation techniques, including radial basis function interpolation (RBF), Gaussian process regression, and feed-forward artificial neural networks. Each model’s contribution is given higher weights in regions where it performs best and lower weights where its accuracy is lower compared to the other models. Additionally, a random forest regression technique is used to determine the weights for previously unseen conditions. The performance of the aggregated model is assessed through two test cases: the 2D flow past a NACA 4412 airfoil at a 5-degree angle of attack, with the Reynolds number ranging between \(1 \times 10^{5}\) and \(1 \times 10^{6}\), and a transonic flow over a NACA 0012 airfoil, with the angle of attack as the varying parameter. In both scenarios, the mixed-ROM demonstrated improved accuracy compared to each individual ROM technique, while providing an estimate for the predictive uncertainty.

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
Acta Mechanica
Acta Mechanica 物理-力学
CiteScore
4.30
自引率
14.80%
发文量
292
审稿时长
6.9 months
期刊介绍: Since 1965, the international journal Acta Mechanica has been among the leading journals in the field of theoretical and applied mechanics. In addition to the classical fields such as elasticity, plasticity, vibrations, rigid body dynamics, hydrodynamics, and gasdynamics, it also gives special attention to recently developed areas such as non-Newtonian fluid dynamics, micro/nano mechanics, smart materials and structures, and issues at the interface of mechanics and materials. The journal further publishes papers in such related fields as rheology, thermodynamics, and electromagnetic interactions with fluids and solids. In addition, articles in applied mathematics dealing with significant mechanics problems are also welcome.
×
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