Stable a posteriori LES of 2D turbulence using convolutional neural networks: Backscattering analysis and generalization to higher Re via transfer learning
Stable a posteriori LES of 2D turbulence using convolutional neural networks: Backscattering analysis and generalization to higher Re via transfer learning
复制标题
DOI：
10.1016/j.jcp.2022.111090
复制
发表时间：
2022-03-07
影响因子：
4.1
通讯作者：
Hassanzadeh, Pedram
中科院分区：
文献类型：
--
作者：
Guan, Yifei;Chattopadhyay, Ashesh;Hassanzadeh, Pedram
使用机器学习(ML)开发用于大涡模拟(LES)的数据驱动亚网格尺度(SGS)模型越来越受到人们的关注。在先验(离线)测试中，最近的一些研究发现，基于ML的数据驱动的SGS模型在高保真数据(例如，来自直接数值模拟，DNS)上的训练优于基于基线物理的模型，并准确地捕捉到尺度间的传输，包括前向(扩散)和后向散射。虽然后验(在线)测试的不稳定性和不能推广到不同的流动(例如，具有更高的雷诺数，Re)仍然是扩大这种数据驱动的SGS模型应用的主要障碍。例如，许多相同的前述研究都发现了不稳定性，需要经常特别的补救措施来稳定LES，但代价是降低准确性。本文以二维衰变湍流为实验平台，证明了深度卷积神经网络(CNN)在先验检验中能够准确地预测SGS强迫项和尺度间转移，如果训练足够多的样本，则可以得到稳定和准确的后验LES-CNN。进一步的分析将上述不稳定性归因于当训练集较小时CNN在捕获后向散射(反扩散)方面的准确性不成比例地低。我们还表明，转移学习包括用来自新流的少量数据(例如，1%)重新训练CNN，对于Re高16倍(以及如果需要的话，还可以有更高的网格分辨率)的流，可以实现准确和稳定的后验LES-CNN。这些结果显示了CNN与迁移学习相结合的前景，可以为实际应用提供稳定、准确和可推广的LES。(C)2022 Elsevier Inc.保留所有权利。
There is a growing interest in developing data-driven subgrid-scale (SGS) models for large eddy simulation (LES) using machine learning (ML). In a priori (offline) tests, some recent studies have found ML-based data-driven SGS models that are trained on high-fidelity data (e.g., from direct numerical simulation, DNS) to outperform baseline physics-based models and accurately capture the inter-scale transfers, both forward (diffusion) and backscatter. While promising, instabilities in a posteriori (online) tests and inabilities to generalize to a different flow (e.g., with a higher Reynolds number, Re) remain as major obstacles in broadening the applications of such data-driven SGS models. For example, many of the same aforementioned studies have found instabilities that required often ad-hoc remedies to stabilize the LES at the expense of reducing accuracy. Here, using 2D decaying turbulence as the testbed, we show that deep convolutional neural networks (CNNs) can accurately predict the SGS forcing terms and the inter-scale transfers in a priori tests, and if trained with enough samples, lead to stable and accurate a posteriori LES-CNN. Further analysis attributes aforementioned instabilities to the disproportionately lower accuracy of the CNNs in capturing backscattering (anti-diffusion) when the training set is small. We also show that transfer learning, which involves re-training the CNN with a small amount of data (e.g., 1%) from the new flow, enables accurate and stable a posteriori LES-CNN for flows with 16x higher Re (as well as higher grid resolution if needed). These results show the promise of CNNs with transfer learning to provide stable, accurate, and generalizable LES for practical use. (c) 2022 Elsevier Inc. All rights reserved.