Accelerating three-dimensional phase-field simulations via deep learning approaches

doi:10.1007/s10853-024-10118-4

IMECH-IR > 非线性力学国家重点实验室

	Accelerating three-dimensional phase-field simulations via deep learning approaches
	Zhou, Xuewei 1,2; Sun, Sheng 4; Cai SL(蔡松林)6 ; Chen, Gongyu 1; Wu, Honghui 3,5; Xiong, Jie 4; Zhu, Jiaming1,2,5
通讯作者	Wu, Honghui([email protected]) ; Xiong, Jie([email protected]) ; Zhu, Jiaming([email protected])
发表期刊	JOURNAL OF MATERIALS SCIENCE
	2024-09-01
卷号	59 期号:33 页码:15727-15737
ISSN	0022-2461
摘要	Phase-field modeling (PFM) is a powerful but computationally expensive technique for simulating three-dimensional (3D) microstructure evolutions. Very recently, integrating machine learning into phase-field simulations provides a promising way to reduce calculation time remarkably. In this study, we propose a deep learning model that combines a convolutional autoencoder with a deep operator network to predict 3D microstructure evolution by using 2D slices of the 3D system. It is found that the deep learning model can shorten the calculation time from 37 min to 3 s after the initial training, while skipping 5-time steps, and reduce the phase-field simulation time by 31% in entire calculation of the evolution process. Interestingly, this model achieves good accuracy in predicting 3D microstructures by utilizing only 2D information. This work demonstrates the efficiency of machine learning in accelerating phase-field simulations while maintaining high accuracy and promotes the application of PFM in fundamental studies.
DOI	10.1007/s10853-024-10118-4
收录类别	SCI ; EI
语种	英语
WOS记录号	WOS:001297590500002
关键词[WOS]	ALLOYS ; MODEL ; CAHN ; RECRYSTALLIZATION ; APPROXIMATION ; NETWORK
WOS研究方向	Materials Science
WOS类目	Materials Science, Multidisciplinary
资助项目	National Natural Science Foundation of China[12372152] ; National Natural Science Foundation of China[52122408] ; National Natural Science Foundation of China[52071023] ; National Natural Science Foundation of China[12072179] ; Qilu Young Talent Program of Shandong University, Zhejiang Lab Open Research Project[K2022PE0AB05] ; Shandong Provincial Natural Science Foundation[ZR2023MA058] ; Guangdong Basic and Applied Basic Research Foundation[2023A1515011819] ; Guangdong Basic and Applied Basic Research Foundation[2024A1515012469]
项目资助者	National Natural Science Foundation of China ; Qilu Young Talent Program of Shandong University, Zhejiang Lab Open Research Project ; Shandong Provincial Natural Science Foundation ; Guangdong Basic and Applied Basic Research Foundation
论文分区	二类
力学所作者排名	3
RpAuthor	Wu, Honghui ; Xiong, Jie ; Zhu, Jiaming
引用统计
文献类型	期刊论文
条目标识符	http://dspace.imech.ac.cn/handle/311007/96385
专题	非线性力学国家重点实验室
作者单位	1.Shandong Univ, Sch Civil Engn, Jinan 250061, Peoples R China; 2.Shandong Univ, Shenzhen Res Inst, Shenzhen 518057, Peoples R China; 3.Univ Sci & Technol Beijing, Inst Carbon Neutral, Beijing Adv Innovat Ctr Mat Genome Engn, Beijing 100083, Peoples R China; 4.Shanghai Univ, Mat Genome Inst, Shanghai 200444, Peoples R China; 5.Liaoning Acad Mat, Inst Mat Intelligent Technol, Shenyang 110004, Peoples R China; 6.Chinese Acad Sci, Inst Mech, State Key Lab Nonlinear Mech, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Zhou, Xuewei,Sun, Sheng,Cai SL,et al. Accelerating three-dimensional phase-field simulations via deep learning approaches[J]. JOURNAL OF MATERIALS SCIENCE,2024,59,33,:15727-15737.Rp_Au:Wu, Honghui, Xiong, Jie, Zhu, Jiaming
APA	Zhou, Xuewei.,Sun, Sheng.,蔡松林.,Chen, Gongyu.,Wu, Honghui.,...&Zhu, Jiaming.(2024).Accelerating three-dimensional phase-field simulations via deep learning approaches.JOURNAL OF MATERIALS SCIENCE,59(33),15727-15737.
MLA	Zhou, Xuewei,et al."Accelerating three-dimensional phase-field simulations via deep learning approaches".JOURNAL OF MATERIALS SCIENCE 59.33(2024):15727-15737.