Attention-enhanced neural network models for turbulence simulation

Peng，Wenhui1,2,3,4; Yuan，Zelong1,2,3; Wang，Jianchun1,2,3

2022-02-01

10.1063/5.0079302

PHYSICS OF FLUIDS   影响因子和分区

1070-6631

1089-7666

Deep neural network models have shown great potential in accelerating the simulation of fluid dynamic systems. Once trained, these models can make inferences within seconds, thus can be extremely efficient. However, it becomes more difficult for neural networks to make accurate predictions when the flow becomes more chaotic and turbulent at higher Reynolds numbers. One of the most important reasons is that existing models lack the mechanism to handle the unique characteristic of high-Reynolds-number turbulent flow; multi-scale flow structures are nonuniformly 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 compare the model performance against a state-of-the-art neural network model as the baseline, the Fourier neural operator, on a two-dimensional turbulence prediction task. Numerical experiments show that the attention-enhanced neural network model outperforms existing state-of-the-art baselines, and can accurately reconstruct a variety of statistics and instantaneous spatial structures of turbulence at high Reynolds numbers. Furthermore, the attention mechanism provides 40% error reduction with 1% increase in parameters, at the same level of computational cost.

SCI ; EI

英语

第一 ; 通讯

National Natural Science Foundation of China[12172161];National Natural Science Foundation of China[91752201];National Natural Science Foundation of China[91952104];National Natural Science Foundation of China[92052301];

WOS:000753470400012

20220811680908

Behavioral research ; Deep neural networks ; Reynolds number

Ergonomics and Human Factors Engineering:461.4 ; Fluid Flow, General:631.1 ; Social Sciences:971

PHYSICS

2-s2.0-85124707240

Scopus

1.Department of Mechanics and Aerospace Engineering,Southern University of Science and Technology,Shenzhen,518055,China
2.Southern Marine Science and Engineering Guangdong Laboratory (Guangzhou),Guangzhou,511458,China
3.Guangdong-Hong Kong-Macao Joint Laboratory for Data-Driven Fluid Mechanics and Engineering Applications,Southern University of Science and Technology,Shenzhen,518055,China
4.Department of Computer Engineering,Polytechnique Montreal,Montreal,H3T1J4,Canada

Peng，Wenhui,Yuan，Zelong,Wang，Jianchun. Attention-enhanced neural network models for turbulence simulation[J]. PHYSICS OF FLUIDS,2022,34(2).

Peng，Wenhui,Yuan，Zelong,&Wang，Jianchun.(2022).Attention-enhanced neural network models for turbulence simulation.PHYSICS OF FLUIDS,34(2).

Peng，Wenhui,et al."Attention-enhanced neural network models for turbulence simulation".PHYSICS OF FLUIDS 34.2(2022).