.png){kind=icon}
返回
Higher order dynamic mode decomposition beyond aerospace engineering
.png){kind=icon}
OA
AI
DOI：10.1016/j.rineng.2023.101471.png)
.png){kind=icon}
.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
It is a well known fact that fluid dynamics play a crucial rule in countless fields in scientific and industrial applications, including nature and medicine (ocean currents, fluid motion around jellyfish, blood circulation...), in energy production (wind turbines, tidal energy, combustion...) and of course in transportation and aerospace (trains, automobiles, aerospace shape optimization, cross-flow instability...). Understanding all these complex fluid dynamics problems is not an easy task and it requires a routinely description and quantification. Data-science and its diverse technique have arisen as a powerful tool to tackle these fluid dynamics problems, techniques such as machine learning and artificial intelligence, data processing and decision making, modeling and simulations are but a few of the promising tools described so far. In this work, we focus on data-driven techniques. In particular, we are introducing the fully data-driven technique, the higher order dynamic mode decomposition (HODMD). This technique was originally developed for the fluid mechanics field, but rapidly grew and covered several other fields and applications. A short review comparing two different applications will be shown in this contribution. We will first present our proposed technique, explain its solid mathematical foundations and highlight its capabilities (robustness, denoising properties, efficiency regardless of the amounts of data...). Next, we show the applicability of the HODMD technique to two, very different and quite complex problems: (i) a compressible, turbulent jet and (ii) pattern identification in medical dataset, consisting of echocardiography images.
Keyword:
Fluid dynamics
Higher order dynamic mode decomposition
Data-driven methods
Aerospace
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
7.9
论文数:
1.2W
被引数:
1.7W
机构
引用论文
An alternative method to study cross-flow instabilities based on high order dynamic mode decomposition.png)
PHYSICS OF FLUIDS
PHYSICS OF FLUIDSIF4.3
PREAI
分享收藏分享收藏
Convolutional neural network based hierarchical autoencoder for nonlinear mode decomposition of fluid field data基于卷积神经网络的分层自动编码器，用于流场数据的非线性模式分解PHYSICS OF FLUIDSOAAI
PHYSICS OF FLUIDSIF4.3
OAAI分享收藏Nonlinear mode decomposition with convolutional neural networks for fluid dynamics用于流体动力学的卷积神经网络的非线性模式分解OAAI
OAAI分享收藏分享收藏
查看更多内容