Prediction of buckling coefficient of stiffened plate girders using deep learning algorithm

George Papazafeiropoulos; Quang Viet Vu; Viet Hung Truong; Minh Chinh Luong; Van Trung Pham

doi:10.1007/978-981-15-0802-8_183

Prediction of buckling coefficient of stiffened plate girders using deep learning algorithm

George Papazafeiropoulos
, Quang Viet Vu^*
, Viet Hung Truong
, Minh Chinh Luong
, Van Trung Pham

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Chapter › peer-review

8 Scopus citations

Abstract

This paper aims at introducing a new method to determine the buckling coefficient k_b of the stiffened plate girders under pure bending using deep learning, one of the most powerful algorithms in machine learning. Firstly, output data k_b is generated from eigenvalue buckling analyses based on input data (various geometric dimensions of the girder). This procedure is implemented by using the Abaqus2Matlab toolbox, which allows the transfer of data between Matlab and Abaqus and vice versa. After that, 2,200 training data are used to build the model for predicting k_b using deep learning. Finally, 200 test data are used to evaluate the accuracy of the model. The results obtained from this model are also compared with analogous results of previous works with a good agreement.

Original language	English
Title of host publication	Lecture Notes in Civil Engineering
Publisher	Springer
Pages	1143-1148
Number of pages	6
DOIs	https://doi.org/10.1007/978-981-15-0802-8_183
State	Published - 2020
Externally published	Yes

Publication series

Name	Lecture Notes in Civil Engineering
Volume	54
ISSN (Print)	2366-2557
ISSN (Electronic)	2366-2565

Bibliographical note

Publisher Copyright:
© Springer Nature Singapore Pte Ltd. 2020.

Keywords

Abaqus2Matlab
Deep learning
Stiffened plate girders
Stiffeners

ASJC Scopus subject areas

Civil and Structural Engineering

Access to Document

10.1007/978-981-15-0802-8_183

Cite this

@inbook{6dd5602cee81499281b6ce6fa38962f7,

title = "Prediction of buckling coefficient of stiffened plate girders using deep learning algorithm",

abstract = "This paper aims at introducing a new method to determine the buckling coefficient kb of the stiffened plate girders under pure bending using deep learning, one of the most powerful algorithms in machine learning. Firstly, output data kb is generated from eigenvalue buckling analyses based on input data (various geometric dimensions of the girder). This procedure is implemented by using the Abaqus2Matlab toolbox, which allows the transfer of data between Matlab and Abaqus and vice versa. After that, 2,200 training data are used to build the model for predicting kb using deep learning. Finally, 200 test data are used to evaluate the accuracy of the model. The results obtained from this model are also compared with analogous results of previous works with a good agreement.",

keywords = "Abaqus2Matlab, Deep learning, Stiffened plate girders, Stiffeners",

author = "George Papazafeiropoulos and Vu, \{Quang Viet\} and Truong, \{Viet Hung\} and Luong, \{Minh Chinh\} and Pham, \{Van Trung\}",

note = "Publisher Copyright: {\textcopyright} Springer Nature Singapore Pte Ltd. 2020.",

year = "2020",

doi = "10.1007/978-981-15-0802-8\_183",

language = "English",

series = "Lecture Notes in Civil Engineering",

publisher = "Springer",

pages = "1143--1148",

booktitle = "Lecture Notes in Civil Engineering",

address = "Germany",

}

TY - CHAP

T1 - Prediction of buckling coefficient of stiffened plate girders using deep learning algorithm

AU - Papazafeiropoulos, George

AU - Vu, Quang Viet

AU - Truong, Viet Hung

AU - Luong, Minh Chinh

AU - Pham, Van Trung

PY - 2020

Y1 - 2020

N2 - This paper aims at introducing a new method to determine the buckling coefficient kb of the stiffened plate girders under pure bending using deep learning, one of the most powerful algorithms in machine learning. Firstly, output data kb is generated from eigenvalue buckling analyses based on input data (various geometric dimensions of the girder). This procedure is implemented by using the Abaqus2Matlab toolbox, which allows the transfer of data between Matlab and Abaqus and vice versa. After that, 2,200 training data are used to build the model for predicting kb using deep learning. Finally, 200 test data are used to evaluate the accuracy of the model. The results obtained from this model are also compared with analogous results of previous works with a good agreement.

AB - This paper aims at introducing a new method to determine the buckling coefficient kb of the stiffened plate girders under pure bending using deep learning, one of the most powerful algorithms in machine learning. Firstly, output data kb is generated from eigenvalue buckling analyses based on input data (various geometric dimensions of the girder). This procedure is implemented by using the Abaqus2Matlab toolbox, which allows the transfer of data between Matlab and Abaqus and vice versa. After that, 2,200 training data are used to build the model for predicting kb using deep learning. Finally, 200 test data are used to evaluate the accuracy of the model. The results obtained from this model are also compared with analogous results of previous works with a good agreement.

KW - Abaqus2Matlab

KW - Deep learning

KW - Stiffened plate girders

KW - Stiffeners

UR - https://www.scopus.com/pages/publications/85073681528

U2 - 10.1007/978-981-15-0802-8_183

DO - 10.1007/978-981-15-0802-8_183

M3 - Chapter

AN - SCOPUS:85073681528

T3 - Lecture Notes in Civil Engineering

SP - 1143

EP - 1148

BT - Lecture Notes in Civil Engineering

PB - Springer

ER -

Prediction of buckling coefficient of stiffened plate girders using deep learning algorithm

Abstract

Publication series

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this