Physics-informed neural network for hydrodynamic lubrication with film thickness discontinuity

IF 6.3 1区 工程技术 Q1 ENGINEERING, MECHANICAL
Bochao Guan, Qiang He, Yang Hu, Zhiyuan Zheng, Weifeng Huang
{"title":"Physics-informed neural network for hydrodynamic lubrication with film thickness discontinuity","authors":"Bochao Guan, Qiang He, Yang Hu, Zhiyuan Zheng, Weifeng Huang","doi":"10.26599/frict.2025.9441035","DOIUrl":null,"url":null,"abstract":" <p>Recent advancements in physics-informed neural networks (PINNs) have shown promise in solving partial differential equations (PDEs), including those in hydrodynamic lubrication. However, PINNs struggle with film thickness discontinuities because of their requirement for continuous differentiability. This paper introduces two novel PINNs models to address this challenge. Model I employs a hyperbolic tangent function to approximate discontinuous film thickness, ensuring differentiability and continuity. The traditional PINNs structure is maintained by adjusting only the film thickness definition in the Reynolds equation. Model II reframes the lubrication problem as an interface issue, introducing a jump equation and an augmented variable to handle discontinuities. It extends the Reynolds equation into a three-dimensional form and includes an interface loss function for accuracy. Numerical experiments demonstrate the effectiveness of both models in handling thickness discontinuities, with the model showing superior computational precision. The model parameters, including the number of sampling points and loss function weights, were optimized for enhanced accuracy. The models were also tested on various groove shapes, confirming their adaptability in resolving discontinuity issues.</p> ","PeriodicalId":12442,"journal":{"name":"Friction","volume":"102 1","pages":""},"PeriodicalIF":6.3000,"publicationDate":"2025-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Friction","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.26599/frict.2025.9441035","RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, MECHANICAL","Score":null,"Total":0}
引用次数: 0

Abstract

Recent advancements in physics-informed neural networks (PINNs) have shown promise in solving partial differential equations (PDEs), including those in hydrodynamic lubrication. However, PINNs struggle with film thickness discontinuities because of their requirement for continuous differentiability. This paper introduces two novel PINNs models to address this challenge. Model I employs a hyperbolic tangent function to approximate discontinuous film thickness, ensuring differentiability and continuity. The traditional PINNs structure is maintained by adjusting only the film thickness definition in the Reynolds equation. Model II reframes the lubrication problem as an interface issue, introducing a jump equation and an augmented variable to handle discontinuities. It extends the Reynolds equation into a three-dimensional form and includes an interface loss function for accuracy. Numerical experiments demonstrate the effectiveness of both models in handling thickness discontinuities, with the model showing superior computational precision. The model parameters, including the number of sampling points and loss function weights, were optimized for enhanced accuracy. The models were also tested on various groove shapes, confirming their adaptability in resolving discontinuity issues.

Abstract Image

膜厚不连续流体动力润滑的物理信息神经网络
近年来,基于物理的神经网络(pinn)在解决偏微分方程(PDEs)方面取得了进展,包括流体动力润滑领域的偏微分方程。然而,由于pinn需要连续可微性,它与薄膜厚度不连续作斗争。本文介绍了两种新的pinn模型来解决这一挑战。模型1采用双曲正切函数近似不连续膜厚,保证了膜的可微性和连续性。传统的pin结构仅通过调整雷诺方程中的膜厚度定义来维持。模型II将润滑问题重新定义为界面问题,引入跳跃方程和增广变量来处理不连续。它将雷诺方程扩展为三维形式,并包括一个界面损失函数以提高精度。数值实验证明了两种模型在处理厚度不连续性方面的有效性,模型具有较高的计算精度。模型参数,包括采样点的数量和损失函数的权重,进行了优化,以提高准确性。该模型还在各种凹槽形状上进行了测试,证实了它们在解决不连续问题方面的适应性。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Friction
Friction Engineering-Mechanical Engineering
CiteScore
12.90
自引率
13.20%
发文量
324
审稿时长
13 weeks
期刊介绍: Friction is a peer-reviewed international journal for the publication of theoretical and experimental research works related to the friction, lubrication and wear. Original, high quality research papers and review articles on all aspects of tribology are welcome, including, but are not limited to, a variety of topics, such as: Friction: Origin of friction, Friction theories, New phenomena of friction, Nano-friction, Ultra-low friction, Molecular friction, Ultra-high friction, Friction at high speed, Friction at high temperature or low temperature, Friction at solid/liquid interfaces, Bio-friction, Adhesion, etc. Lubrication: Superlubricity, Green lubricants, Nano-lubrication, Boundary lubrication, Thin film lubrication, Elastohydrodynamic lubrication, Mixed lubrication, New lubricants, New additives, Gas lubrication, Solid lubrication, etc. Wear: Wear materials, Wear mechanism, Wear models, Wear in severe conditions, Wear measurement, Wear monitoring, etc. Surface Engineering: Surface texturing, Molecular films, Surface coatings, Surface modification, Bionic surfaces, etc. Basic Sciences: Tribology system, Principles of tribology, Thermodynamics of tribo-systems, Micro-fluidics, Thermal stability of tribo-systems, etc. Friction is an open access journal. It is published quarterly by Tsinghua University Press and Springer, and sponsored by the State Key Laboratory of Tribology (TsinghuaUniversity) and the Tribology Institute of Chinese Mechanical Engineering Society.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信