WaveNets: physics-informed neural networks for full-field recovery of rotational flow beneath large-amplitude periodic water waves
WaveNets: physics-informed neural networks for full-field recovery of rotational flow beneath large-amplitude periodic water waves
复制标题
DOI：
10.1007/s00366-024-01944-w
复制
发表时间：
2024-02
影响因子：
8.7
通讯作者：
Lin Chen;Ben Li;Chenyi Luo;Xiao-Yong Lei
中科院分区：
文献类型：
--
作者：
Lin Chen;Ben Li;Chenyi Luo;Xiao-Yong Lei
我们制定物理信息神经网络（PINNs）的旋转流下的非线性周期性水波使用少量的测量数据，创造WaveNets的全场重建。WaveNets有两个神经网络，分别预测水面和速度/压力场。在WaveNets损失函数中包含了欧拉方程和波动问题的其他先验知识。我们还提出了一种新的方法来动态更新采样点在残差评估的自由表面逐渐形成在模型训练。高保真度的数据集，获得使用的数值延拓方法，能够解决非线性波接近最大高度。对单层和双层旋转流场的模型训练和验证结果表明，WaveNets能够以较少的表面数据和流场数据重建波面和流场。根据涡度分布的先验信息，增加冗余的物理约束，可以提高涡度估计的精度。
We formulate physics-informed neural networks (PINNs) for full-field reconstruction of rotational flow beneath nonlinear periodic water waves using a small amount of measurement data, coined WaveNets. The WaveNets have two NNs to, respectively, predict the water surface, and velocity/pressure fields. The Euler equation and other prior knowledge of the wave problem are included in WaveNets loss function. We also propose a novel method to dynamically update the sampling points in residual evaluation as the free surface is gradually formed during model training. High-fidelity data sets are obtained using the numerical continuation method which is able to solve nonlinear waves close to the largest height. Model training and validation results in cases of both one-layer and two-layer rotational flows show that WaveNets can reconstruct wave surface and flow field with few data either on the surface or in the flow. Accuracy in vorticity estimate can be improved by adding a redundant physical constraint according to the prior information on the vorticity distribution.