Learning-Based High-Performance Algorithm for Long-Term Motion Prediction of Fluid Flows

Authors

Jingyuan Zhu, 1.Department of Electronic Engineering, Tsinghua University, Beijing 100084, China;Follow
Huimin Ma, 2.School of Computer and Communication Engineering, University of Science and Technology Beijing, Beijing 100083, China;Follow
Jian Yuan, 1.Department of Electronic Engineering, Tsinghua University, Beijing 100084, China;

Abstract

Abstract: Simulating the dynamics of fluid flows accurately and efficiently remains a challenging task nowadays, and traditional fluid simulation methods consume large computational resources to obtain accurate results. Deep learning methods have developed rapidly, which makes data-based fluid simulation and generation possible. In this paper, a motion prediction algorithm for long-term fluid simulation is proposed, which is based on a density field with a single frame and a previous velocity field of a sequence. The model focuses on matching the velocity and density fields predicted by the neural network with the simulated data based on the Navier-Stokes equation at a macroscopic level. With the help of fully convolutional U-Net-based autoencoders and LSTM-based time series prediction subnetworks, the model better maintains the visual macroscopic similarity during temporal evolutions and significantly improves computation speed. As a result, the proposed method achieves accurate and rapid long-term motion prediction for the macroscopic distributions of flow field evolution. In addition, the paper demonstrates the effectiveness and efficiency of the proposed algorithm on a series of benchmark tests based on two-dimensional (2D) and three-dimension (3D) simulation data.

Recommended Citation

Zhu, Jingyuan; Ma, Huimin; and Yuan, Jian (2023) "Learning-Based High-Performance Algorithm for Long-Term Motion Prediction of Fluid Flows," Journal of System Simulation: Vol. 35: Iss. 3, Article 1.
DOI: 10.16182/j.issn1004731x.joss.22-1507
Available at: https://dc-china-simulation.researchcommons.org/journal/vol35/iss3/1

First Page

435

DOI Link

https://doi.org/10.16182/j.issn1004731x.joss.22-1507

Last Page

453

CLC

TP391.9

Recommended Citation

Jingyuan Zhu, Huimin Ma, Jian Yuan. Learning-Based High-Performance Algorithm for Long-Term Motion Prediction of Fluid Flows[J]. Journal of System Simulation, 2023, 35(3): 435-453.

Corresponding Author

Huimin Ma,mhmpub@ustb.edu.cn

DOI

10.16182/j.issn1004731x.joss.22-1507