Skip to main content
QUICK REVIEW

[论文解读] Convolutional causal learning for aerodynamic flows

Ryo Koshikawa, Ryo Araki|arXiv (Cornell University)|Jan 27, 2026
Model Reduction and Neural Networks被引用 0
一句话总结

该论文提出一个信息理论卷积学习框架,将流场分解为时变的、与未来升力因果相关的有信息涌动模态,从而实现低秩表示和数据驱动的因果分析用于非定常空气动力学。

ABSTRACT

This study considers capturing aerodynamic causality from snapshot data with a time-varying mode decomposition technique referred to as information-theoretic machine learning. The current approach extracts time-dependent informative vortical structures, contributing to the future evolution of the aerodynamic coefficients. The present decomposition is employed with a convolutional neural network, enabling the identification of the spatial continuous mode. In addition, a low-order representation, characterizing the informative vortical structures and their corresponding aerodynamic coefficients, can also be identified by considering autoencoder-based data compression. The present technique is applied to a range of aerodynamic examples, including extreme vortex-gust airfoil interactions, experimentally measured transverse jet-wing interaction, and a turbulent separated wake. For the cases of gust-wing interaction, the time-varying gust effect on the lift response is extracted in an interpretable manner. With the example of a turbulent wake, the relationship between large-scale vortical motion and lift force is identified without any spatial length-scale information. The proposed approach could serve as a foundation for data-driven causal modeling and control for a range of unsteady flows.

研究动机与目标

  • 激励并量化涡性结构如何因果影响未来的气动力量(力)
  • 开发一个信息理论的卷积深度学习方法以提取有信息的模态
  • 提供能够捕捉因果相关流场特征与升力动力学的低维表示

提出的方法

  • 将 q 定义为流场状态,将 λ 定义为未来的升力系数,并将 q 分解为信息性部分与残差部分(q = q_I + q_R)
  • 使用香农熵和互信息来识别使 I(q_I; λ) 最大化且 H(λ|q_I)=0 与 I(q_R; q_I)=0 的 q_I
  • 实现信息模态提取器 F,作为卷积自编码器或基于 CNN 的模型,权重为非负、激活函数为双射以确保变换为双射
  • 通过一个损失函数进行训练,损失包含回归误差 ||q − q_I||_2 和互信息正则化 β||I(q_R; q_I)||_2
  • 在多样化的气动流动上评估该方法,包括极端涡旋-乱流相互作用、横向射流与翼面的相互作用,以及湍流尾流
Figure 1: An example of the given state $\bm{q}$ and the informative component $\bm{q}_{I}$ decomposed by a data-driven technique.
Figure 1: An example of the given state $\bm{q}$ and the informative component $\bm{q}_{I}$ decomposed by a data-driven technique.

实验结果

研究问题

  • RQ1信息理论因果分解是否能够识别对未来升力具有决定性影响的涡性结构?
  • RQ2卷积结构在提取时变信息模态时如何保持空间一致性?
  • RQ3信息模态对时间窗口 Δt 与正则化参数 β 的依赖性如何?
  • RQ4该方法是否能够在实验或湍流流动数据中揭示仍然可解释的低维表示?

主要发现

  • 该方法能够产生时变的有信息模态,捕捉瞬态涡-翼相互作用中的乱流与升力因果关系。
  • 有信息模态识别出在某些湍流尾流中不依赖纯粹空间尺度信息也能贡献升力的大尺度涡结构。
  • 一个低维潜在表示能够反映乱流贡献与过渡,潜在轨迹随 Δt 变化,且在几何约束下可能形成圆形流形。
  • 该方法对实验噪声具有鲁棒性,能够在跨翼向和三维分离流中恢复有信息的流动特征。
  • 与常规自编码器的比较表明互信息项在揭示具有因果相关性的结构方面具有额外价值。
Figure 2: Informative mode extractor $\mathcal{F}$ based on ( $a$ ) convolutional autoencoder and ( $b$ ) convolutional neural network.
Figure 2: Informative mode extractor $\mathcal{F}$ based on ( $a$ ) convolutional autoencoder and ( $b$ ) convolutional neural network.

更好的研究,从现在开始

从论文设计到论文写作,大幅缩短您的研究时间。

无需绑定信用卡

本解读由 AI 生成,并经人工编辑审核。