arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2011.11911physics.flu-dynphysics.comp-ph
收藏

流体流量估计的神经网络推广技术

Generalization techniques of neural networks for fluid flow estimation

Masaki Morimoto, Kai Fukami, Kai Zhang, Koji Fukagata

摘要:

我们展示了几种技术,以鼓励实际使用的神经网络流体流量估计。在本论文中,三个方面的应用机器学习的流体动力学仍然存在的挑战被认为是:1。机器学习结果的可解释性,2.扩充训练数据,以及3.神经网络的泛化能力对于可解释性,我们首先展示了两种方法来观察神经网络的内部过程,即,隐藏层的可视化和梯度加权类激活映射(Grad-CAM)的应用,应用于典型的流体流动估计问题--圆柱体尾流的$(1)$阻力系数估计和来自粒子图像的$(2)$速度估计。举例说明,这两种方法都可以成功地告诉我们基于机器学习的估计的巨大能力的证据。然后,我们利用一些技术批量出训练数据的超分辨率分析和时间预测圆柱尾迹和NOAA海表面温度数据证明,足够的训练,训练数据量有限的神经网络可以实现流体流动问题。通过对训练数据的内插/外推,以及对平行圆柱尾迹的超分辨,讨论了机器学习模型的推广性。我们发现,即使对于有关距离因素的测试配置,也可以很好地重建由两个圆柱体之间复杂相互作用产生的各种流态。本文可以是一个重要的一步,实际使用的神经网络的层流和湍流问题。

英文摘要:

We demonstrate several techniques to encourage practical uses of neural networks for fluid flow estimation. In the present paper, three perspectives which are remaining challenges for applications of machine learning to fluid dynamics are considered: 1. interpretability of machine-learned results, 2. bulking out of training data, and 3. generalizability of neural networks. For the interpretability, we first demonstrate two methods to observe the internal procedure of neural networks, i.e., visualization of hidden layers and application of gradient-weighted class activation mapping (Grad-CAM), applied to canonical fluid flow estimation problems -- $(1)$ drag coefficient estimation of a cylinder wake and $(2)$ velocity estimation from particle images. It is exemplified that both approaches can successfully tell us evidences of the great capability of machine learning-based estimations. We then utilize some techniques to bulk out training data for super-resolution analysis and temporal prediction for cylinder wake and NOAA sea surface temperature data to demonstrate that sufficient training of neural networks with limited amount of training data can be achieved for fluid flow problems. The generalizability of machine learning model is also discussed by accounting for the perspectives of inter/extrapolation of training data, considering super-resolution of wakes behind two parallel cylinders. We find that various flow patterns generated by complex interaction between two cylinders can be reconstructed well, even for the test configurations regarding the distance factor. The present paper can be a significant step toward practical uses of neural networks for both laminar and turbulent flow problems.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑