Using a Simple Soil Spring Model and Support Vector Machine to Determine Bridge Scour Depth and Bridge Safety

Chung-Wei Feng, Shen-Haw Ju, Hsun Yi Huang

研究成果: Article

3 引文 (Scopus)

摘要

Scouring around bridge piers is an important safety issue of bridge management since it can lead to bridge slant and collapse. Therefore, bridge scour depth is an important index to determine a bridge's safety condition. However, directly measuring the scour depth is difficult due to the water surrounds bridge piers. Several researchers have discovered the bridge natural frequency, a proxy measurement that can be used to determine the scour depth of the bridge pier. In this study, two finite-element models including the complicated finite-element model and the simple soil spring model are introduced to compute the natural frequency of a bridge and the result is verified with field experiments. In addition, a series of simulations is then implemented to analyze the factors impacting scouring depth. Seven factors related to bridge substructure and soil property are identified. A support vector machine (SVM) is further employed to understand how those factors impact the safety of the bridge. The results show that by hybridizing the simple soil spring model and SVM, the bridge's safety level can be efficiently established.

原文English
文章編號04015088
期刊Journal of Performance of Constructed Facilities
30
發行號4
DOIs
出版狀態Published - 2016 八月 1

指紋

Scour
Support vector machines
Soils
Bridge piers
Natural frequencies

All Science Journal Classification (ASJC) codes

  • Civil and Structural Engineering
  • Building and Construction
  • Safety, Risk, Reliability and Quality

引用此文

@article{a1c7d227b6b84d409f5caefdb5f0d6fe,
title = "Using a Simple Soil Spring Model and Support Vector Machine to Determine Bridge Scour Depth and Bridge Safety",
abstract = "Scouring around bridge piers is an important safety issue of bridge management since it can lead to bridge slant and collapse. Therefore, bridge scour depth is an important index to determine a bridge's safety condition. However, directly measuring the scour depth is difficult due to the water surrounds bridge piers. Several researchers have discovered the bridge natural frequency, a proxy measurement that can be used to determine the scour depth of the bridge pier. In this study, two finite-element models including the complicated finite-element model and the simple soil spring model are introduced to compute the natural frequency of a bridge and the result is verified with field experiments. In addition, a series of simulations is then implemented to analyze the factors impacting scouring depth. Seven factors related to bridge substructure and soil property are identified. A support vector machine (SVM) is further employed to understand how those factors impact the safety of the bridge. The results show that by hybridizing the simple soil spring model and SVM, the bridge's safety level can be efficiently established.",
author = "Chung-Wei Feng and Shen-Haw Ju and Huang, {Hsun Yi}",
year = "2016",
month = "8",
day = "1",
doi = "10.1061/(ASCE)CF.1943-5509.0000837",
language = "English",
volume = "30",
journal = "Journal of Performance of Constructed Facilities",
issn = "0887-3828",
publisher = "American Society of Civil Engineers (ASCE)",
number = "4",

}

TY - JOUR

T1 - Using a Simple Soil Spring Model and Support Vector Machine to Determine Bridge Scour Depth and Bridge Safety

AU - Feng, Chung-Wei

AU - Ju, Shen-Haw

AU - Huang, Hsun Yi

PY - 2016/8/1

Y1 - 2016/8/1

N2 - Scouring around bridge piers is an important safety issue of bridge management since it can lead to bridge slant and collapse. Therefore, bridge scour depth is an important index to determine a bridge's safety condition. However, directly measuring the scour depth is difficult due to the water surrounds bridge piers. Several researchers have discovered the bridge natural frequency, a proxy measurement that can be used to determine the scour depth of the bridge pier. In this study, two finite-element models including the complicated finite-element model and the simple soil spring model are introduced to compute the natural frequency of a bridge and the result is verified with field experiments. In addition, a series of simulations is then implemented to analyze the factors impacting scouring depth. Seven factors related to bridge substructure and soil property are identified. A support vector machine (SVM) is further employed to understand how those factors impact the safety of the bridge. The results show that by hybridizing the simple soil spring model and SVM, the bridge's safety level can be efficiently established.

AB - Scouring around bridge piers is an important safety issue of bridge management since it can lead to bridge slant and collapse. Therefore, bridge scour depth is an important index to determine a bridge's safety condition. However, directly measuring the scour depth is difficult due to the water surrounds bridge piers. Several researchers have discovered the bridge natural frequency, a proxy measurement that can be used to determine the scour depth of the bridge pier. In this study, two finite-element models including the complicated finite-element model and the simple soil spring model are introduced to compute the natural frequency of a bridge and the result is verified with field experiments. In addition, a series of simulations is then implemented to analyze the factors impacting scouring depth. Seven factors related to bridge substructure and soil property are identified. A support vector machine (SVM) is further employed to understand how those factors impact the safety of the bridge. The results show that by hybridizing the simple soil spring model and SVM, the bridge's safety level can be efficiently established.

UR - http://www.scopus.com/inward/record.url?scp=84979025661&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84979025661&partnerID=8YFLogxK

U2 - 10.1061/(ASCE)CF.1943-5509.0000837

DO - 10.1061/(ASCE)CF.1943-5509.0000837

M3 - Article

AN - SCOPUS:84979025661

VL - 30

JO - Journal of Performance of Constructed Facilities

JF - Journal of Performance of Constructed Facilities

SN - 0887-3828

IS - 4

M1 - 04015088

ER -