Efficient prediction of aerodynamic forces in rarefied flow using convolutional neural network based multi-process method

The direct simulation Monte Carlo (DSMC) is a widely used approach for studying aerodynamics effects of rarefied flows, but it is highly time-consuming and may exhibit statistical fluctuations. In this study, we propose an efficient aerodynamic prediction method based on convolutional neural network...

Full description

Saved in:
Bibliographic Details
Main Authors: Haifeng Huang, Guobiao Cai, Chuanfeng Wei, Baiyi Zhang, Xiang Cui, Yongjia Zhao, Huiyan Weng, Weizong Wang, Lihui Liu, Bijiao He
Format: Article
Language:English
Published: IOP Publishing 2025-01-01
Series:Machine Learning: Science and Technology
Subjects:
Online Access:https://doi.org/10.1088/2632-2153/addf10
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849469618374049792
author Haifeng Huang
Guobiao Cai
Chuanfeng Wei
Baiyi Zhang
Xiang Cui
Yongjia Zhao
Huiyan Weng
Weizong Wang
Lihui Liu
Bijiao He
author_facet Haifeng Huang
Guobiao Cai
Chuanfeng Wei
Baiyi Zhang
Xiang Cui
Yongjia Zhao
Huiyan Weng
Weizong Wang
Lihui Liu
Bijiao He
author_sort Haifeng Huang
collection DOAJ
description The direct simulation Monte Carlo (DSMC) is a widely used approach for studying aerodynamics effects of rarefied flows, but it is highly time-consuming and may exhibit statistical fluctuations. In this study, we propose an efficient aerodynamic prediction method based on convolutional neural networks (CNNs) to further explore the application of deep learning in improving the efficiency of DSMC for calculating the aerodynamics of rarefied flows. The method includes centroid aerodynamics forces prediction (CFP) and surface aerodynamic forces distribution prediction (SFP), both of which are trained using a dataset of free molecular flow around obstacles derived from DSMC simulations. The SFP is designed to bridge the gap between flow field and surface forces, with two characteristics extraction methods developed specifically for this purpose. Additionally, two data preprocessing methods are designed to suppress the statistical noise inherent in DSMC simulations. Both CFP and SFP have demonstrated optimal performance in terms of accuracy and resistance to overfitting, achieving considerable predictive accuracy. The SFP exhibits a significant speedup, enabling real-time prediction of aerodynamic distributions from flow field. The results demonstrate that the proposed CNN-based approach offers a promising solution for the efficient calculation of aerodynamic forces in rarefied flows, and provide a robust foundation for ongoing development.
format Article
id doaj-art-14028f58dd2d4fbaaed64a3a94f89b1c
institution Kabale University
issn 2632-2153
language English
publishDate 2025-01-01
publisher IOP Publishing
record_format Article
series Machine Learning: Science and Technology
spelling doaj-art-14028f58dd2d4fbaaed64a3a94f89b1c2025-08-20T03:25:26ZengIOP PublishingMachine Learning: Science and Technology2632-21532025-01-016202506010.1088/2632-2153/addf10Efficient prediction of aerodynamic forces in rarefied flow using convolutional neural network based multi-process methodHaifeng Huang0https://orcid.org/0009-0005-9120-7959Guobiao Cai1Chuanfeng Wei2Baiyi Zhang3Xiang Cui4Yongjia Zhao5Huiyan Weng6Weizong Wang7https://orcid.org/0000-0002-6022-1441Lihui Liu8https://orcid.org/0000-0002-3752-038XBijiao He9https://orcid.org/0000-0002-6346-7981School of Astronautics, Beihang University , Beijing 100191, People’s Republic of China; Key Laboratory of Spacecraft Design Optimization and Dynamic Simulation Technologies, Ministry of Education , Beijing 100191, People’s Republic of ChinaSchool of Astronautics, Beihang University , Beijing 100191, People’s Republic of China; Key Laboratory of Spacecraft Design Optimization and Dynamic Simulation Technologies, Ministry of Education , Beijing 100191, People’s Republic of ChinaChina Satellite Network System Co., Ltd , Beijing 100029, People’s Republic of ChinaSchool of Astronautics, Beihang University , Beijing 100191, People’s Republic of China; Key Laboratory of Spacecraft Design Optimization and Dynamic Simulation Technologies, Ministry of Education , Beijing 100191, People’s Republic of ChinaChina Satellite Network System Co., Ltd , Beijing 100029, People’s Republic of ChinaChina Satellite Network System Co., Ltd , Beijing 100029, People’s Republic of ChinaSchool of Astronautics, Beihang University , Beijing 100191, People’s Republic of China; Key Laboratory of Spacecraft Design Optimization and Dynamic Simulation Technologies, Ministry of Education , Beijing 100191, People’s Republic of ChinaSchool of Astronautics, Beihang University , Beijing 100191, People’s Republic of China; Key Laboratory of Spacecraft Design Optimization and Dynamic Simulation Technologies, Ministry of Education , Beijing 100191, People’s Republic of China; Aircraft and Propulsion Laboratory, Ningbo Institute of Technology, Beihang University , Ningbo 315832, People’s Republic of ChinaSchool of Astronautics, Beihang University , Beijing 100191, People’s Republic of China; Key Laboratory of Spacecraft Design Optimization and Dynamic Simulation Technologies, Ministry of Education , Beijing 100191, People’s Republic of ChinaSchool of Astronautics, Beihang University , Beijing 100191, People’s Republic of China; Key Laboratory of Spacecraft Design Optimization and Dynamic Simulation Technologies, Ministry of Education , Beijing 100191, People’s Republic of ChinaThe direct simulation Monte Carlo (DSMC) is a widely used approach for studying aerodynamics effects of rarefied flows, but it is highly time-consuming and may exhibit statistical fluctuations. In this study, we propose an efficient aerodynamic prediction method based on convolutional neural networks (CNNs) to further explore the application of deep learning in improving the efficiency of DSMC for calculating the aerodynamics of rarefied flows. The method includes centroid aerodynamics forces prediction (CFP) and surface aerodynamic forces distribution prediction (SFP), both of which are trained using a dataset of free molecular flow around obstacles derived from DSMC simulations. The SFP is designed to bridge the gap between flow field and surface forces, with two characteristics extraction methods developed specifically for this purpose. Additionally, two data preprocessing methods are designed to suppress the statistical noise inherent in DSMC simulations. Both CFP and SFP have demonstrated optimal performance in terms of accuracy and resistance to overfitting, achieving considerable predictive accuracy. The SFP exhibits a significant speedup, enabling real-time prediction of aerodynamic distributions from flow field. The results demonstrate that the proposed CNN-based approach offers a promising solution for the efficient calculation of aerodynamic forces in rarefied flows, and provide a robust foundation for ongoing development.https://doi.org/10.1088/2632-2153/addf10rarefied flowdirect simulation Monte Carloaerodynamic forcesdeep learningconvolutional neural networks
spellingShingle Haifeng Huang
Guobiao Cai
Chuanfeng Wei
Baiyi Zhang
Xiang Cui
Yongjia Zhao
Huiyan Weng
Weizong Wang
Lihui Liu
Bijiao He
Efficient prediction of aerodynamic forces in rarefied flow using convolutional neural network based multi-process method
Machine Learning: Science and Technology
rarefied flow
direct simulation Monte Carlo
aerodynamic forces
deep learning
convolutional neural networks
title Efficient prediction of aerodynamic forces in rarefied flow using convolutional neural network based multi-process method
title_full Efficient prediction of aerodynamic forces in rarefied flow using convolutional neural network based multi-process method
title_fullStr Efficient prediction of aerodynamic forces in rarefied flow using convolutional neural network based multi-process method
title_full_unstemmed Efficient prediction of aerodynamic forces in rarefied flow using convolutional neural network based multi-process method
title_short Efficient prediction of aerodynamic forces in rarefied flow using convolutional neural network based multi-process method
title_sort efficient prediction of aerodynamic forces in rarefied flow using convolutional neural network based multi process method
topic rarefied flow
direct simulation Monte Carlo
aerodynamic forces
deep learning
convolutional neural networks
url https://doi.org/10.1088/2632-2153/addf10
work_keys_str_mv AT haifenghuang efficientpredictionofaerodynamicforcesinrarefiedflowusingconvolutionalneuralnetworkbasedmultiprocessmethod
AT guobiaocai efficientpredictionofaerodynamicforcesinrarefiedflowusingconvolutionalneuralnetworkbasedmultiprocessmethod
AT chuanfengwei efficientpredictionofaerodynamicforcesinrarefiedflowusingconvolutionalneuralnetworkbasedmultiprocessmethod
AT baiyizhang efficientpredictionofaerodynamicforcesinrarefiedflowusingconvolutionalneuralnetworkbasedmultiprocessmethod
AT xiangcui efficientpredictionofaerodynamicforcesinrarefiedflowusingconvolutionalneuralnetworkbasedmultiprocessmethod
AT yongjiazhao efficientpredictionofaerodynamicforcesinrarefiedflowusingconvolutionalneuralnetworkbasedmultiprocessmethod
AT huiyanweng efficientpredictionofaerodynamicforcesinrarefiedflowusingconvolutionalneuralnetworkbasedmultiprocessmethod
AT weizongwang efficientpredictionofaerodynamicforcesinrarefiedflowusingconvolutionalneuralnetworkbasedmultiprocessmethod
AT lihuiliu efficientpredictionofaerodynamicforcesinrarefiedflowusingconvolutionalneuralnetworkbasedmultiprocessmethod
AT bijiaohe efficientpredictionofaerodynamicforcesinrarefiedflowusingconvolutionalneuralnetworkbasedmultiprocessmethod