Applications of scientific machine learning for the analysis of functionally graded porous beams

Research output: Contribution to journalArticleResearchpeer review

Authors

  • Mohammad Sadegh Eshaghi
  • Mostafa Bamdad
  • Cosmin Anitescu
  • Yizheng Wang
  • Xiaoying Zhuang
  • Timon Rabczuk

Research Organisations

External Research Organisations

  • Bauhaus-Universität Weimar
  • Tsinghua University
  • Tongji University
View graph of relations

Details

Original languageEnglish
Article number129119
Number of pages15
JournalNEUROCOMPUTING
Volume619
Early online date7 Dec 2024
Publication statusE-pub ahead of print - 7 Dec 2024

Abstract

This study help investigates different Scientific Machine Learning (SciML) approaches for the analysis of functionally graded (FG) porous beams and compares them under a new framework. The beam material properties are assumed to vary as an arbitrary continuous function. The methods consider the output of a neural network/operator as an approximation to the displacement fields and derive the equations governing beam behavior based on the continuum formulation. They are implemented in the framework and formulated by three approaches: (a) the vector approach leads to a Physics-Informed Neural Network (PINN), (b) the energy approach brings about the Deep Energy Method (DEM), and (c) the data-driven approach, which results in a class of Neural Operator methods. Finally, a neural operator has been trained to predict the response of the porous beam with functionally graded material under any porosity distribution pattern and any arbitrary traction condition. The results are validated with analytical and numerical reference solutions. The data and code accompanying this manuscript will be publicly available at https://github.com/eshaghi-ms/DeepNetBeam.

Keywords

    Deep energy method, Fourier neural operator, Functionally graded material, Physics-informed neural network, Porous beam, Scientific machine learning

ASJC Scopus subject areas

Cite this

Applications of scientific machine learning for the analysis of functionally graded porous beams. / Eshaghi, Mohammad Sadegh; Bamdad, Mostafa; Anitescu, Cosmin et al.
In: NEUROCOMPUTING, Vol. 619, 129119, 28.02.2025.

Research output: Contribution to journalArticleResearchpeer review

Eshaghi, M. S., Bamdad, M., Anitescu, C., Wang, Y., Zhuang, X., & Rabczuk, T. (2025). Applications of scientific machine learning for the analysis of functionally graded porous beams. NEUROCOMPUTING, 619, Article 129119. Advance online publication. https://doi.org/10.48550/arXiv.2408.02698, https://doi.org/10.1016/j.neucom.2024.129119
Eshaghi MS, Bamdad M, Anitescu C, Wang Y, Zhuang X, Rabczuk T. Applications of scientific machine learning for the analysis of functionally graded porous beams. NEUROCOMPUTING. 2025 Feb 28;619:129119. Epub 2024 Dec 7. doi: 10.48550/arXiv.2408.02698, 10.1016/j.neucom.2024.129119
Eshaghi, Mohammad Sadegh ; Bamdad, Mostafa ; Anitescu, Cosmin et al. / Applications of scientific machine learning for the analysis of functionally graded porous beams. In: NEUROCOMPUTING. 2025 ; Vol. 619.
Download
@article{01497b314c774992ad1b5c3679252603,
title = "Applications of scientific machine learning for the analysis of functionally graded porous beams",
abstract = "This study help investigates different Scientific Machine Learning (SciML) approaches for the analysis of functionally graded (FG) porous beams and compares them under a new framework. The beam material properties are assumed to vary as an arbitrary continuous function. The methods consider the output of a neural network/operator as an approximation to the displacement fields and derive the equations governing beam behavior based on the continuum formulation. They are implemented in the framework and formulated by three approaches: (a) the vector approach leads to a Physics-Informed Neural Network (PINN), (b) the energy approach brings about the Deep Energy Method (DEM), and (c) the data-driven approach, which results in a class of Neural Operator methods. Finally, a neural operator has been trained to predict the response of the porous beam with functionally graded material under any porosity distribution pattern and any arbitrary traction condition. The results are validated with analytical and numerical reference solutions. The data and code accompanying this manuscript will be publicly available at https://github.com/eshaghi-ms/DeepNetBeam.",
keywords = "Deep energy method, Fourier neural operator, Functionally graded material, Physics-informed neural network, Porous beam, Scientific machine learning",
author = "Eshaghi, {Mohammad Sadegh} and Mostafa Bamdad and Cosmin Anitescu and Yizheng Wang and Xiaoying Zhuang and Timon Rabczuk",
note = "Publisher Copyright: {\textcopyright} 2024 The Authors",
year = "2024",
month = dec,
day = "7",
doi = "10.48550/arXiv.2408.02698",
language = "English",
volume = "619",
journal = "NEUROCOMPUTING",
issn = "0925-2312",
publisher = "Elsevier",

}

Download

TY - JOUR

T1 - Applications of scientific machine learning for the analysis of functionally graded porous beams

AU - Eshaghi, Mohammad Sadegh

AU - Bamdad, Mostafa

AU - Anitescu, Cosmin

AU - Wang, Yizheng

AU - Zhuang, Xiaoying

AU - Rabczuk, Timon

N1 - Publisher Copyright: © 2024 The Authors

PY - 2024/12/7

Y1 - 2024/12/7

N2 - This study help investigates different Scientific Machine Learning (SciML) approaches for the analysis of functionally graded (FG) porous beams and compares them under a new framework. The beam material properties are assumed to vary as an arbitrary continuous function. The methods consider the output of a neural network/operator as an approximation to the displacement fields and derive the equations governing beam behavior based on the continuum formulation. They are implemented in the framework and formulated by three approaches: (a) the vector approach leads to a Physics-Informed Neural Network (PINN), (b) the energy approach brings about the Deep Energy Method (DEM), and (c) the data-driven approach, which results in a class of Neural Operator methods. Finally, a neural operator has been trained to predict the response of the porous beam with functionally graded material under any porosity distribution pattern and any arbitrary traction condition. The results are validated with analytical and numerical reference solutions. The data and code accompanying this manuscript will be publicly available at https://github.com/eshaghi-ms/DeepNetBeam.

AB - This study help investigates different Scientific Machine Learning (SciML) approaches for the analysis of functionally graded (FG) porous beams and compares them under a new framework. The beam material properties are assumed to vary as an arbitrary continuous function. The methods consider the output of a neural network/operator as an approximation to the displacement fields and derive the equations governing beam behavior based on the continuum formulation. They are implemented in the framework and formulated by three approaches: (a) the vector approach leads to a Physics-Informed Neural Network (PINN), (b) the energy approach brings about the Deep Energy Method (DEM), and (c) the data-driven approach, which results in a class of Neural Operator methods. Finally, a neural operator has been trained to predict the response of the porous beam with functionally graded material under any porosity distribution pattern and any arbitrary traction condition. The results are validated with analytical and numerical reference solutions. The data and code accompanying this manuscript will be publicly available at https://github.com/eshaghi-ms/DeepNetBeam.

KW - Deep energy method

KW - Fourier neural operator

KW - Functionally graded material

KW - Physics-informed neural network

KW - Porous beam

KW - Scientific machine learning

UR - http://www.scopus.com/inward/record.url?scp=85212570136&partnerID=8YFLogxK

U2 - 10.48550/arXiv.2408.02698

DO - 10.48550/arXiv.2408.02698

M3 - Article

AN - SCOPUS:85212570136

VL - 619

JO - NEUROCOMPUTING

JF - NEUROCOMPUTING

SN - 0925-2312

M1 - 129119

ER -