Physics-informed neural network based on a new adaptive gradient descent algorithm for solving partial differential equations of flow problems

偏微分方程 梯度下降 流量(数学) 应用数学 人工神经网络 数学优化 计算机科学 算法 物理 数学 数学分析 人工智能 机械
作者
Xiaojian Li,Yuhao Liu,Zhengxian Liu
出处
期刊:Physics of Fluids [American Institute of Physics]
卷期号:35 (6) 被引量:13
标识
DOI:10.1063/5.0151244
摘要

Physics-informed neural network (PINN) is an emerging technique for solving partial differential equations (PDEs) of flow problems. Due to the advantage of low computational cost, the gradient descent algorithms coupled with the weighted objectives method are usually used to optimize loss functions in the PINN training. However, the interaction mechanisms between gradients of loss functions are not fully clarified, leading to poor performances in loss functions optimization. For this, an adaptive gradient descent algorithm (AGDA) is proposed based on the interaction mechanisms analyses and then validated by analytical PDEs and flow problems. First, the interaction mechanisms of loss functions gradients in the PINN training based on the traditional Adam optimizer are analyzed. The main factors responsible for the poor performances of the Adam optimizer are identified. Then, a new AGDA optimizer is developed for the PINN training by two modifications: (1) balancing the magnitude difference of loss functions gradients and (2) eliminating the gradient directions conflict. Finally, three types of PDEs (elliptic, hyperbolic, and parabolic) and four viscous incompressible flow problems are selected to validate the proposed algorithm. It is found that to reach the specified accuracy, the required training time of the AGDA optimizer is about 16%–90% of the Adam optimizer and 41%–64% of the PCGrad optimizer, and the demanded number of iterations is about 10%–68% of the Adam optimizer and 38%–77% of the PCGrad optimizer. Therefore, the PINN method coupled with the AGDA optimizer is a more efficient and robust technique for solving partial differential equations of flow problems.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
PDF的下载单位、IP信息已删除 (2025-6-4)

科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
satan9完成签到,获得积分10
1秒前
量子星尘发布了新的文献求助10
1秒前
英俊的铭应助白巧小丸子采纳,获得10
4秒前
雪酪芋泥球完成签到 ,获得积分10
4秒前
4秒前
5秒前
hhh完成签到,获得积分10
5秒前
二冲完成签到,获得积分10
6秒前
打滚完成签到,获得积分10
6秒前
Belinda完成签到 ,获得积分10
7秒前
ST发布了新的文献求助10
7秒前
白巧小丸子完成签到,获得积分10
8秒前
9秒前
小何爱学习完成签到,获得积分10
14秒前
14秒前
15秒前
大模型应助素直采纳,获得10
16秒前
18秒前
19秒前
完美世界应助L912294993采纳,获得10
19秒前
落后的纸鹤完成签到,获得积分10
20秒前
cxdfo完成签到,获得积分10
20秒前
卷卷完成签到,获得积分10
27秒前
27秒前
27秒前
月如钩发布了新的文献求助10
28秒前
30秒前
31秒前
小蘑菇应助忧虑的绮梅采纳,获得10
31秒前
31秒前
dominate发布了新的文献求助10
34秒前
music发布了新的文献求助10
34秒前
rpe发布了新的文献求助10
36秒前
乐乐应助端庄涟妖采纳,获得10
37秒前
yyjm发布了新的文献求助10
38秒前
38秒前
认真草丛完成签到,获得积分10
39秒前
39秒前
yuan发布了新的文献求助10
39秒前
黄74185296完成签到,获得积分10
40秒前
高分求助中
A new approach to the extrapolation of accelerated life test data 1000
Picture Books with Same-sex Parented Families: Unintentional Censorship 700
ACSM’s Guidelines for Exercise Testing and Prescription, 12th edition 500
Nucleophilic substitution in azasydnone-modified dinitroanisoles 500
不知道标题是什么 500
Indomethacinのヒトにおける経皮吸収 400
Phylogenetic study of the order Polydesmida (Myriapoda: Diplopoda) 370
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 遗传学 基因 物理化学 催化作用 冶金 细胞生物学 免疫学
热门帖子
关注 科研通微信公众号,转发送积分 3975543
求助须知:如何正确求助?哪些是违规求助? 3519971
关于积分的说明 11200248
捐赠科研通 3256311
什么是DOI,文献DOI怎么找? 1798213
邀请新用户注册赠送积分活动 877446
科研通“疑难数据库(出版商)”最低求助积分说明 806338