arXiv 2110.024712021-10physics.flu-dynphysics.comp-ph

用于Turbulence仿真的注意力增强神经网络模型

Attention-Enhanced Neural Network Models for Turbulence Simulation

Wenhui Peng, Zelong Yuan, Jianchun Wang

备注：: 18 pages, 25 figures
DOI：: 10.1063/5.0079302

摘要：

深度神经网络模型在加速流体动力学系统仿真方面显示出巨大的潜力。一旦经过训练，这些模型可以在几秒钟内进行推理，因此非常有效。然而，当流动变得混乱和湍流时，它们遭受泛化问题。其中一个最重要的原因是，现有的模型缺乏处理湍流独特特征的机制：多尺度流动结构是非均匀分布和强非平衡的。在这项工作中，我们解决这个问题的视觉注意力的概念：直观地说，我们希望注意力模块通过自动调整不同区域的权重来捕捉湍流的非平衡。我们基准的性能改进与最先进的神经网络模型，傅立叶神经运算符（FNO），在二维（2D）湍流预测任务。数值实验表明，注意力增强神经网络模型对高雷诺数流动具有较好的泛化能力，能够准确地重建湍流的各种统计特性和瞬时空间结构。注意机制提供了40%的错误减少与1%的参数增加，在相同的计算成本水平。

英文摘要：

Deep neural network models have shown a great potential in accelerating the simulation of fluid dynamic systems. Once trained, these models can make inference within seconds, thus can be extremely efficient. However, they suffer from a generalization problem when the flow becomes chaotic and turbulent. One of the most important reasons is that, existing models lack the mechanism to handle the unique characteristic of turbulent flow: multi-scale flow structures are non-uniformly distributed and strongly nonequilibrium. In this work, we address this issue with the concept of visual attention: intuitively, we expect the attention module to capture the nonequilibrium of turbulence by automatically adjusting weights on different regions. We benchmark the performance improvement with a state of the art neural network model, the Fourier Neural Operator (FNO), on two-dimensional (2D) turbulence prediction task. Numerical experiments show that the attention-enhanced neural network model can generalize well on higher Reynolds numbers flow, and can accurately reconstruct a variety of statistics and instantaneous spatial structures of turbulence. The attention mechanism provides 40% error reduction with 1% increase of parameters, at the same level of computational cost.

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。

摘要：

英文摘要：

相关论文