Seismic Response Meta-model of High-Rise Fame Structure Based on Time-Delay Neural Network

Research output: Contribution to journalArticleResearchpeer review

Research Organisations

External Research Organisations

  • Yangzhou University
  • University of Liverpool
  • Tongji University
View graph of relations

Details

Original languageEnglish
Pages (from-to)2273-2294
Number of pages22
JournalKSCE journal of civil engineering
Volume26
Issue number5
Early online date18 Feb 2022
Publication statusPublished - May 2022

Abstract

To make structural seismic response simulation more efficient, a meta-model method which is based on the time delay neural network is proposed. And an accuracy evaluation method that considers the drift peak amplitudes and maximum amplitudes in each intensity as performance parameters is also proposed, this method can make a balance between accuracy and training time. Exampled by 4 frame structures which are all 20 stories, and accuracy evaluating results show that more than 80% of samples, which include training models and testing models of these performance parameters can be explained by meta models’ fitting. The average time to simulate by this method is 0.08s and faster than the finite element method which spends 24 min averagely.

Keywords

    Frame structure, Metamodel, Neural network, Seismic, SRC-RC frame

ASJC Scopus subject areas

Cite this

Seismic Response Meta-model of High-Rise Fame Structure Based on Time-Delay Neural Network. / Zhang, He; Bittner, Marius; Beer, Michael.
In: KSCE journal of civil engineering, Vol. 26, No. 5, 05.2022, p. 2273-2294.

Research output: Contribution to journalArticleResearchpeer review

Download
@article{299cc13fa6ec42d08f18c02ea9011b1f,
title = "Seismic Response Meta-model of High-Rise Fame Structure Based on Time-Delay Neural Network",
abstract = "To make structural seismic response simulation more efficient, a meta-model method which is based on the time delay neural network is proposed. And an accuracy evaluation method that considers the drift peak amplitudes and maximum amplitudes in each intensity as performance parameters is also proposed, this method can make a balance between accuracy and training time. Exampled by 4 frame structures which are all 20 stories, and accuracy evaluating results show that more than 80% of samples, which include training models and testing models of these performance parameters can be explained by meta models{\textquoteright} fitting. The average time to simulate by this method is 0.08s and faster than the finite element method which spends 24 min averagely.",
keywords = "Frame structure, Metamodel, Neural network, Seismic, SRC-RC frame",
author = "He Zhang and Marius Bittner and Michael Beer",
note = "Funding Information: This research was funded by Junior Researcher Grant Yangzhou University (Grant No. 137012122), and this study was financially supported by the China Scholarship Council (CSC) (Grant No. 20180670149). Natural earthquake data were downloaded from PEER Strong Ground Motion Databases ( https://peer.berkeley.edu/peer-strong-ground-motion-databases ). And the calculation is supported by Institute for Risk and Reliability, University Hannover.",
year = "2022",
month = may,
doi = "10.1007/s12205-022-0878-7",
language = "English",
volume = "26",
pages = "2273--2294",
journal = "KSCE journal of civil engineering",
issn = "1226-7988",
publisher = "Korean Society of Civil Engineers",
number = "5",

}

Download

TY - JOUR

T1 - Seismic Response Meta-model of High-Rise Fame Structure Based on Time-Delay Neural Network

AU - Zhang, He

AU - Bittner, Marius

AU - Beer, Michael

N1 - Funding Information: This research was funded by Junior Researcher Grant Yangzhou University (Grant No. 137012122), and this study was financially supported by the China Scholarship Council (CSC) (Grant No. 20180670149). Natural earthquake data were downloaded from PEER Strong Ground Motion Databases ( https://peer.berkeley.edu/peer-strong-ground-motion-databases ). And the calculation is supported by Institute for Risk and Reliability, University Hannover.

PY - 2022/5

Y1 - 2022/5

N2 - To make structural seismic response simulation more efficient, a meta-model method which is based on the time delay neural network is proposed. And an accuracy evaluation method that considers the drift peak amplitudes and maximum amplitudes in each intensity as performance parameters is also proposed, this method can make a balance between accuracy and training time. Exampled by 4 frame structures which are all 20 stories, and accuracy evaluating results show that more than 80% of samples, which include training models and testing models of these performance parameters can be explained by meta models’ fitting. The average time to simulate by this method is 0.08s and faster than the finite element method which spends 24 min averagely.

AB - To make structural seismic response simulation more efficient, a meta-model method which is based on the time delay neural network is proposed. And an accuracy evaluation method that considers the drift peak amplitudes and maximum amplitudes in each intensity as performance parameters is also proposed, this method can make a balance between accuracy and training time. Exampled by 4 frame structures which are all 20 stories, and accuracy evaluating results show that more than 80% of samples, which include training models and testing models of these performance parameters can be explained by meta models’ fitting. The average time to simulate by this method is 0.08s and faster than the finite element method which spends 24 min averagely.

KW - Frame structure

KW - Metamodel

KW - Neural network

KW - Seismic

KW - SRC-RC frame

UR - http://www.scopus.com/inward/record.url?scp=85124759645&partnerID=8YFLogxK

U2 - 10.1007/s12205-022-0878-7

DO - 10.1007/s12205-022-0878-7

M3 - Article

AN - SCOPUS:85124759645

VL - 26

SP - 2273

EP - 2294

JO - KSCE journal of civil engineering

JF - KSCE journal of civil engineering

SN - 1226-7988

IS - 5

ER -