A study of the use of artificial neural networks to estimate dynamic displacements due to dynamic loads in bridges

S. Ok, W. Son, Yun Mook Lim

Research output: Contribution to journalConference article

5 Citations (Scopus)

Abstract

Measurement of dynamic displacement is one of the most essential aspects of a structural behavior because it portrays history of the global behavior of structure. In general, structural engineers are accepted these response as reliable physical quantities to evaluate the conditions of a structure. The reason is that these physical quantities can easily generate strain as well as stress, velocity and acceleration at the measuring points. However, it is difficult to directly measure the displacement of the bridge due to problems such as test conditions and the limitations of equipment. Therefore, in this study, an artificial neural network (ANN) demonstrates how it could overcome such limitations and utilize the random dynamic load to obtain the reliable estimations. Numerical analysis is conducted to obtain learning data about the axial strain as well as vertical displacement with time frame at multi-points and then applied to the ANN. The scenario centered on a variety of dynamic loads from the analysis of an urban bridge that was selected based on its general volume of traffic. The analysis was performed to estimate its displacement, which corresponds to the strain on the bridge caused by arbitrary loads of leaning results from the ANN. Then, it is confirmed that the estimated displacements of ANN show well agreements with that of an independent set of traffic scenario.

Original languageEnglish
Article number012032
JournalJournal of Physics: Conference Series
Volume382
Issue number1
DOIs
Publication statusPublished - 2012 Jan 1
EventModern Practice in Stress and Vibration Analysis 2012, MPSVA 2012 - Glasgow, United Kingdom
Duration: 2012 Aug 282012 Aug 31

Fingerprint

dynamic loads
estimates
traffic
random loads
axial strain
learning
engineers
numerical analysis
histories

All Science Journal Classification (ASJC) codes

  • Physics and Astronomy(all)

Cite this

@article{10deba8f50524d759c63d66bee666de9,
title = "A study of the use of artificial neural networks to estimate dynamic displacements due to dynamic loads in bridges",
abstract = "Measurement of dynamic displacement is one of the most essential aspects of a structural behavior because it portrays history of the global behavior of structure. In general, structural engineers are accepted these response as reliable physical quantities to evaluate the conditions of a structure. The reason is that these physical quantities can easily generate strain as well as stress, velocity and acceleration at the measuring points. However, it is difficult to directly measure the displacement of the bridge due to problems such as test conditions and the limitations of equipment. Therefore, in this study, an artificial neural network (ANN) demonstrates how it could overcome such limitations and utilize the random dynamic load to obtain the reliable estimations. Numerical analysis is conducted to obtain learning data about the axial strain as well as vertical displacement with time frame at multi-points and then applied to the ANN. The scenario centered on a variety of dynamic loads from the analysis of an urban bridge that was selected based on its general volume of traffic. The analysis was performed to estimate its displacement, which corresponds to the strain on the bridge caused by arbitrary loads of leaning results from the ANN. Then, it is confirmed that the estimated displacements of ANN show well agreements with that of an independent set of traffic scenario.",
author = "S. Ok and W. Son and Lim, {Yun Mook}",
year = "2012",
month = "1",
day = "1",
doi = "10.1088/1742-6596/382/1/012032",
language = "English",
volume = "382",
journal = "Journal of Physics: Conference Series",
issn = "1742-6588",
publisher = "IOP Publishing Ltd.",
number = "1",

}

A study of the use of artificial neural networks to estimate dynamic displacements due to dynamic loads in bridges. / Ok, S.; Son, W.; Lim, Yun Mook.

In: Journal of Physics: Conference Series, Vol. 382, No. 1, 012032, 01.01.2012.

Research output: Contribution to journalConference article

TY - JOUR

T1 - A study of the use of artificial neural networks to estimate dynamic displacements due to dynamic loads in bridges

AU - Ok, S.

AU - Son, W.

AU - Lim, Yun Mook

PY - 2012/1/1

Y1 - 2012/1/1

N2 - Measurement of dynamic displacement is one of the most essential aspects of a structural behavior because it portrays history of the global behavior of structure. In general, structural engineers are accepted these response as reliable physical quantities to evaluate the conditions of a structure. The reason is that these physical quantities can easily generate strain as well as stress, velocity and acceleration at the measuring points. However, it is difficult to directly measure the displacement of the bridge due to problems such as test conditions and the limitations of equipment. Therefore, in this study, an artificial neural network (ANN) demonstrates how it could overcome such limitations and utilize the random dynamic load to obtain the reliable estimations. Numerical analysis is conducted to obtain learning data about the axial strain as well as vertical displacement with time frame at multi-points and then applied to the ANN. The scenario centered on a variety of dynamic loads from the analysis of an urban bridge that was selected based on its general volume of traffic. The analysis was performed to estimate its displacement, which corresponds to the strain on the bridge caused by arbitrary loads of leaning results from the ANN. Then, it is confirmed that the estimated displacements of ANN show well agreements with that of an independent set of traffic scenario.

AB - Measurement of dynamic displacement is one of the most essential aspects of a structural behavior because it portrays history of the global behavior of structure. In general, structural engineers are accepted these response as reliable physical quantities to evaluate the conditions of a structure. The reason is that these physical quantities can easily generate strain as well as stress, velocity and acceleration at the measuring points. However, it is difficult to directly measure the displacement of the bridge due to problems such as test conditions and the limitations of equipment. Therefore, in this study, an artificial neural network (ANN) demonstrates how it could overcome such limitations and utilize the random dynamic load to obtain the reliable estimations. Numerical analysis is conducted to obtain learning data about the axial strain as well as vertical displacement with time frame at multi-points and then applied to the ANN. The scenario centered on a variety of dynamic loads from the analysis of an urban bridge that was selected based on its general volume of traffic. The analysis was performed to estimate its displacement, which corresponds to the strain on the bridge caused by arbitrary loads of leaning results from the ANN. Then, it is confirmed that the estimated displacements of ANN show well agreements with that of an independent set of traffic scenario.

UR - http://www.scopus.com/inward/record.url?scp=84867938980&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84867938980&partnerID=8YFLogxK

U2 - 10.1088/1742-6596/382/1/012032

DO - 10.1088/1742-6596/382/1/012032

M3 - Conference article

VL - 382

JO - Journal of Physics: Conference Series

JF - Journal of Physics: Conference Series

SN - 1742-6588

IS - 1

M1 - 012032

ER -