论文标题
关于使用复发性神经网络预测湍流
On the use of recurrent neural networks for predictions of turbulent flows
论文作者
论文摘要
在本文中，Moehlis {\ it等人}（new J. Phys。{\ bf 6}，56，2004）在低阶模型中评估了复发性神经网络的预测能力。我们的结果表明，有可能通过经过适当训练的长期记忆（LSTM）网络来获得有关湍流统计数据和流动的动态行为的出色预测，从而导致平均值的相对误差和低于$ 1 \％$的波动。我们还观察到，仅根据流量的瞬时预测使用损失函数可能不会导致湍流统计的最佳预测，并且有必要根据计算的统计数据来定义停止标准。此外，更复杂的损失功能，不仅包括瞬时预测，而且还包括流动的平均行为，可能会导致更快的神经网络训练。
In this paper, the prediction capabilities of recurrent neural networks are assessed in the low-order model of near-wall turbulence by Moehlis {\it et al.} (New J. Phys. {\bf 6}, 56, 2004). Our results show that it is possible to obtain excellent predictions of the turbulence statistics and the dynamic behavior of the flow with properly trained long short-term memory (LSTM) networks, leading to relative errors in the mean and the fluctuations below $1\%$. We also observe that using a loss function based only on the instantaneous predictions of the flow may not lead to the best predictions in terms of turbulence statistics, and it is necessary to define a stopping criterion based on the computed statistics. Furthermore, more sophisticated loss functions, including not only the instantaneous predictions but also the averaged behavior of the flow, may lead to much faster neural network training.
