A Physics-Informed Neural Operator for the Simulation of Surface Waves

被引：0

作者：

Mathias M.S. ^{[1
]}

Netto C.F.D. ^{[1
]}

Moreno F.M. ^{[1
]}

Coelho J.F. ^{[1
]}

de Freitas L.P. ^{[1
]}

de Barros M.R. ^{[1
]}

de Mello P.C. ^{[1
]}

Dottori M. ^{[2
]}

Cozman F.G. ^{[1
]}

Costa A.H.R. ^{[1
]}

Nogueira A.C., Jr. ^{[3
]}

Gomi E.S. ^{[1
]}

Tannuri E.A. ^{[1
]}

机构：

[1] Escola Politećnica, Universidade de São Paulo, São Paulo

[2] Instituto Oceanografíco, Universidade de São Paulo, São Paulo

[3] IBM Research, São Paulo

来源：

Journal of Offshore Mechanics and Arctic Engineering | 2024年 / 146卷 / 06期

基金：

巴西圣保罗研究基金会;

关键词：

computational fluid dynamics; machine learning; neural networks; wave modeling;

D O I：

10.1115/1.4064676

中图分类号：

学科分类号：

摘要：

We develop and implement a neural operator (NOp) to predict the evolution of waves on the surface of water. The NOp uses a graph neural network (GNN) to connect randomly sampled points on the water surface and exchange information between them to make the prediction. Our main contribution is adding physical knowledge to the implementation, which allows the model to be more general and able to be used in domains of different geometries with no retraining. Our implementation also takes advantage of the fact that the governing equations are independent of rotation and translation to make training easier. In this work, the model is trained with data from a single domain with fixed dimensions and evaluated in domains of different dimensions with little impact to performance. Copyright © 2024 by ASME.

引用

共 50 条

[21] Exactly conservative physics-informed neural networks and deep operator networks for dynamical systems
Cardoso-Bihlo, Elsa
Bihlo, Alex
Neural Networks, 2025, 181
[22] A fast general thermal simulation model based on Multi-Branch Physics-Informed deep operator neural network
Lu, Zibo
Zhou, Yuanye
Zhang, Yanbo
Hu, Xiaoguang
Zhao, Qiao
Hu, Xuyang
PHYSICS OF FLUIDS, 2024, 36 (03)
[23] Enhanced physics-informed neural networks for hyperelasticity
Abueidda, Diab W.
Koric, Seid
Guleryuz, Erman
Sobh, Nahil A.
INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN ENGINEERING, 2023, 124 (07) : 1585 - 1601
[24] SOBOLEV TRAINING FOR PHYSICS-INFORMED NEURAL NETWORKS
Son, Hwijae
Jang, Jin woo
Han, Woo jin
Hwang, Hyung ju
COMMUNICATIONS IN MATHEMATICAL SCIENCES, 2023, 21 (06) : 1679 - 1705
[25] Physics-informed neural networks for diffraction tomography
Saba, Amirhossein
Gigli, Carlo
Ayoub, Ahmed B.
Psaltis, Demetri
ADVANCED PHOTONICS, 2022, 4 (06):
[26] Is L2 Physics-Informed Loss Always Suitable for Training Physics-Informed Neural Network?
Wang, Chuwei
Li, Shanda
He, Di
Wang, Liwei
ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 35 (NEURIPS 2022), 2022,
[27] Physics-Informed Neural Differential Equation Model
Chen, Haowei
Guo, Yu
Yuan, Zhaolin
Wang, Baojie
Ban, Xiaojuan
Beijing Youdian Daxue Xuebao/Journal of Beijing University of Posts and Telecommunications, 2024, 47 (04): : 90 - 97
[28] Physics-informed neural networks for consolidation of soils
Zhang, Sheng
Lan, Peng
Li, Hai-Chao
Tong, Chen-Xi
Sheng, Daichao
ENGINEERING COMPUTATIONS, 2022, 39 (07) : 2845 - 2865
[29] Optimizing Physics-Informed Neural Network in Dynamic System Simulation and Learning of Parameters
Oluwasakin, Ebenezer O.
Khaliq, Abdul Q. M.
ALGORITHMS, 2023, 16 (12)
[30] PINNProv: Provenance for Physics-Informed Neural Networks
de Oliveira, Lyncoln S.
Kunstmann, Liliane
Pina, Debora
de Oliveira, Daniel
Mattoso, Marta
2023 INTERNATIONAL SYMPOSIUM ON COMPUTER ARCHITECTURE AND HIGH PERFORMANCE COMPUTING WORKSHOPS, SBAC-PADW, 2023, : 16 - 23

← 1 2 3 4 5 →