Direct numerical simulations of three-dimensional two-phase flow using physics-informed neural networks with a distributed parallel training algorithm
文献类型：期刊论文
| 作者 | Qiu RD(丘润荻)5,6; Li, Junzhe2,4; Wang JZ(王静竹)3,6; Fan, Chun1,2; Wang YW(王一伟)3,5,6 |
| 刊名 | JOURNAL OF FLUID MECHANICS
 |
| 出版日期 | 2025-08-15 |
| 卷号 | 1017页码:20 |
| 关键词 | gas/liquid flow machine learning computational methods |
| ISSN号 | 0022-1120 |
| DOI | 10.1017/jfm.2025.10448 |
| 通讯作者 | Wang, Yiwei(wangyw@imech.ac.cn) |
| 英文摘要 | In recent years, integrating physical constraints within deep neural networks has emerged as an effective approach for expediting direct numerical simulations in two-phase flow. This paper introduces physics-informed neural networks (PINNs) that utilise the phase-field method to model three-dimensional two-phase flows. We present a fully connected neural network architecture with residual blocks and spatial parallel training using the overlapping domain decomposition method across multiple graphics processing units to enhance the accuracy and computational efficiency of PINNs for the phase-field method (PF-PINNs). The proposed PINNs framework is applied to a bubble rising scenario in a three-dimensional infinite water tank to quantitatively assess the performance of PF-PINNs. Furthermore, the computational cost and parallel efficiency of the proposed method was evaluated, demonstrating its potential for widespread application in complex training environments. |
| 分类号 | 一类/力学重要期刊 |
| WOS关键词 | DEEP LEARNING FRAMEWORK ; VISCOUS-LIQUIDS ; BUBBLES ; LAYER |
| 资助项目 | National Natural Science Foundation of China[12293003] ; National Natural Science Foundation of China[12525210] ; National Natural Science Foundation of China[12293000] ; National Natural Science Foundation of China[12272382] ; National Natural Science Foundation of China[52441101] ; Youth Innovation Promotion Association CAS[2022019] |
| WOS研究方向 | Mechanics ; Physics |
| 语种 | 英语 |
| WOS记录号 | WOS:001550275800001 |
| 资助机构 | National Natural Science Foundation of China ; Youth Innovation Promotion Association CAS |
| 其他责任者 | 王一伟 |
| 源URL | [http://dspace.imech.ac.cn/handle/311007/103777]  |
| 专题 | 力学研究所_流固耦合系统力学重点实验室(2012-) |
| 作者单位 | 1.Peking Univ, Changsha Inst Comp & Digital Econ, Changsha 410000, Peoples R China 2.Peking Univ, Comp Ctr, Beijing 100871, Peoples R China; 3.Univ Chinese Acad Sci, Sch Engn Sci, Beijing 100049, Peoples R China; 4.Peking Univ, Sch Comp Sci, Beijing 100871, Peoples R China; 5.Univ Chinese Acad Sci, Sch Future Technol, Beijing 100049, Peoples R China; 6.Chinese Acad Sci, Inst Mech, Key Lab Mech Fluid Solid Coupling Syst, Beijing 100190, Peoples R China; |
| 推荐引用方式 GB/T 7714 | Qiu RD,Li, Junzhe,Wang JZ,et al. Direct numerical simulations of three-dimensional two-phase flow using physics-informed neural networks with a distributed parallel training algorithm[J]. JOURNAL OF FLUID MECHANICS,2025,1017:20. |
| APA | 丘润荻,Li, Junzhe,王静竹,Fan, Chun,&王一伟.(2025).Direct numerical simulations of three-dimensional two-phase flow using physics-informed neural networks with a distributed parallel training algorithm.JOURNAL OF FLUID MECHANICS,1017,20. |
| MLA | 丘润荻,et al."Direct numerical simulations of three-dimensional two-phase flow using physics-informed neural networks with a distributed parallel training algorithm".JOURNAL OF FLUID MECHANICS 1017(2025):20. |
入库方式： OAI收割
来源：力学研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。