Deep learned finite elements

Jaeho Jung; Kyungho Yoon; Phill Seung Lee

doi:10.1016/j.cma.2020.113401

Deep learned finite elements

Jaeho Jung, Kyungho Yoon, Phill Seung Lee

School of Mathematics and Computing

Research output: Contribution to journal › Article › peer-review

19 Citations (Scopus)

Abstract

In this paper, we propose a method that employs deep learning, an artificial intelligence technique, to generate stiffness matrices of finite elements. The proposed method is used to develop 4- and 8-node 2D solid finite elements. The deep learned finite elements practically pass the patch tests and the zero energy mode tests. Through various numerical examples, the performance of the developed elements is investigated and compared with those of existing elements. Computation efficiency is also studied. It was confirmed that the deep learned finite elements can potentially outperform existing finite elements. The proposed method can be applied to generate various types of finite elements in the future.

Original language	English
Article number	113401
Journal	Computer Methods in Applied Mechanics and Engineering
Volume	372
DOIs	https://doi.org/10.1016/j.cma.2020.113401
Publication status	Published - 2020 Dec 1

Bibliographical note

Funding Information:
This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (No. NRF-2018R1A2B3005328 ). This work was also supported by the “Human Resources Program in Energy Technology” of the Korea Institute of Energy Technology Evaluation and Planning (KETEP), granted financial resource from the Ministry of Trade, Industry & Energy, Republic of Korea (No. 20184030202000 ). This research was the result of a study on the “HPC Support” Project, supported by the Ministry of Science and ICT (MSIT) and National IT Industry Promotion Agency (NIPA) . We also thank Dr. Yonggyun Yu at Korea Atomic Energy Research Institute (KAERI) for his valuable comments.

Funding Information:
This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (No. NRF-2018R1A2B3005328). This work was also supported by the ?Human Resources Program in Energy Technology? of the Korea Institute of Energy Technology Evaluation and Planning (KETEP), granted financial resource from the Ministry of Trade, Industry & Energy, Republic of Korea (No. 20184030202000). This research was the result of a study on the ?HPC Support? Project, supported by the Ministry of Science and ICT (MSIT) andNational IT Industry Promotion Agency (NIPA). We also thank Dr. Yonggyun Yu at Korea Atomic Energy Research Institute (KAERI) for his valuable comments.

Publisher Copyright:
© 2020 Elsevier B.V.

All Science Journal Classification (ASJC) codes

Computational Mechanics
Mechanics of Materials
Mechanical Engineering
Physics and Astronomy(all)
Computer Science Applications

Access to Document

10.1016/j.cma.2020.113401

Cite this

@article{dc3d9a79b3424a69b085f0ae7990d710,

title = "Deep learned finite elements",

abstract = "In this paper, we propose a method that employs deep learning, an artificial intelligence technique, to generate stiffness matrices of finite elements. The proposed method is used to develop 4- and 8-node 2D solid finite elements. The deep learned finite elements practically pass the patch tests and the zero energy mode tests. Through various numerical examples, the performance of the developed elements is investigated and compared with those of existing elements. Computation efficiency is also studied. It was confirmed that the deep learned finite elements can potentially outperform existing finite elements. The proposed method can be applied to generate various types of finite elements in the future.",

author = "Jaeho Jung and Kyungho Yoon and Lee, {Phill Seung}",

note = "Funding Information: This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (No. NRF-2018R1A2B3005328 ). This work was also supported by the “Human Resources Program in Energy Technology” of the Korea Institute of Energy Technology Evaluation and Planning (KETEP), granted financial resource from the Ministry of Trade, Industry & Energy, Republic of Korea (No. 20184030202000 ). This research was the result of a study on the “HPC Support” Project, supported by the Ministry of Science and ICT (MSIT) and National IT Industry Promotion Agency (NIPA) . We also thank Dr. Yonggyun Yu at Korea Atomic Energy Research Institute (KAERI) for his valuable comments. Funding Information: This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (No. NRF-2018R1A2B3005328). This work was also supported by the ?Human Resources Program in Energy Technology? of the Korea Institute of Energy Technology Evaluation and Planning (KETEP), granted financial resource from the Ministry of Trade, Industry & Energy, Republic of Korea (No. 20184030202000). This research was the result of a study on the ?HPC Support? Project, supported by the Ministry of Science and ICT (MSIT) andNational IT Industry Promotion Agency (NIPA). We also thank Dr. Yonggyun Yu at Korea Atomic Energy Research Institute (KAERI) for his valuable comments. Publisher Copyright: {\textcopyright} 2020 Elsevier B.V.",

year = "2020",

month = dec,

day = "1",

doi = "10.1016/j.cma.2020.113401",

language = "English",

volume = "372",

journal = "Computer Methods in Applied Mechanics and Engineering",

issn = "0374-2830",

publisher = "Elsevier",

}

TY - JOUR

T1 - Deep learned finite elements

AU - Jung, Jaeho

AU - Yoon, Kyungho

AU - Lee, Phill Seung

N1 - Funding Information: This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (No. NRF-2018R1A2B3005328 ). This work was also supported by the “Human Resources Program in Energy Technology” of the Korea Institute of Energy Technology Evaluation and Planning (KETEP), granted financial resource from the Ministry of Trade, Industry & Energy, Republic of Korea (No. 20184030202000 ). This research was the result of a study on the “HPC Support” Project, supported by the Ministry of Science and ICT (MSIT) and National IT Industry Promotion Agency (NIPA) . We also thank Dr. Yonggyun Yu at Korea Atomic Energy Research Institute (KAERI) for his valuable comments. Funding Information: This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (No. NRF-2018R1A2B3005328). This work was also supported by the ?Human Resources Program in Energy Technology? of the Korea Institute of Energy Technology Evaluation and Planning (KETEP), granted financial resource from the Ministry of Trade, Industry & Energy, Republic of Korea (No. 20184030202000). This research was the result of a study on the ?HPC Support? Project, supported by the Ministry of Science and ICT (MSIT) andNational IT Industry Promotion Agency (NIPA). We also thank Dr. Yonggyun Yu at Korea Atomic Energy Research Institute (KAERI) for his valuable comments. Publisher Copyright: © 2020 Elsevier B.V.

PY - 2020/12/1

Y1 - 2020/12/1

N2 - In this paper, we propose a method that employs deep learning, an artificial intelligence technique, to generate stiffness matrices of finite elements. The proposed method is used to develop 4- and 8-node 2D solid finite elements. The deep learned finite elements practically pass the patch tests and the zero energy mode tests. Through various numerical examples, the performance of the developed elements is investigated and compared with those of existing elements. Computation efficiency is also studied. It was confirmed that the deep learned finite elements can potentially outperform existing finite elements. The proposed method can be applied to generate various types of finite elements in the future.

AB - In this paper, we propose a method that employs deep learning, an artificial intelligence technique, to generate stiffness matrices of finite elements. The proposed method is used to develop 4- and 8-node 2D solid finite elements. The deep learned finite elements practically pass the patch tests and the zero energy mode tests. Through various numerical examples, the performance of the developed elements is investigated and compared with those of existing elements. Computation efficiency is also studied. It was confirmed that the deep learned finite elements can potentially outperform existing finite elements. The proposed method can be applied to generate various types of finite elements in the future.

UR - http://www.scopus.com/inward/record.url?scp=85090402886&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85090402886&partnerID=8YFLogxK

U2 - 10.1016/j.cma.2020.113401

DO - 10.1016/j.cma.2020.113401

M3 - Article

AN - SCOPUS:85090402886

SN - 0374-2830

VL - 372

JO - Computer Methods in Applied Mechanics and Engineering

JF - Computer Methods in Applied Mechanics and Engineering

M1 - 113401

ER -

Deep learned finite elements

Abstract

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this