Predicting tunnel convergence using Multivariate Adaptive Regression Spline and Artificial Neural Network

Amoussou Coffi Adoko, Yu Yong Jiao, Li Wu, Hao Wang, Zi Hao Wang

Research output: Contribution to journalArticle

36 Citations (Scopus)

Abstract

Determining the tunnel convergence is an indispensable task in tunneling, especially when adopting the New Austrian Tunneling Method. The interpretation of the monitoring allows adjusting the construction methods in order to achieve more effective tunneling conditions and to avoid problems like rock collapse, trapping and jamming of boring machine, delay of the project or even geological disasters. In this research, a model capable of predicting the diameter convergence of a high-speed railway tunnel in weak rock was established based on two approaches: Multivariate Adaptive Regression Spline (MARS) and Artificial Neural Network (ANN). A tunnel construction project located in Hunan province (China) was used as case study. The input parameters included the class index of the surrounding rock mass, angle of internal friction, cohesion, Young's modulus, rock density, tunnel overburden, distance between the monitoring station and the tunnel heading face and the elapsed monitoring time. The performance of the models was evaluated by comparing the predicted convergence to the measured data using several performance indices. Overall, the results showed high accuracy of the model predictability of tunnel convergence with MARS showing a light lesser accuracy. However, MARS was more flexible and computationally efficient. It is concluded that MARS can constitute a reliable alternative to ANN in modeling nonlinear geo-engineering problem such as the tunnel convergence.

Original languageEnglish
Pages (from-to)368-376
Number of pages9
JournalTunnelling and Underground Space Technology
Volume38
DOIs
Publication statusPublished - Sep 1 2013

Fingerprint

Splines
artificial neural network
Tunnels
tunnel
Neural networks
Rocks
Monitoring
rock
NATM
weak rock
construction method
Young modulus
Internal friction
Jamming
monitoring
boring
cohesion
overburden
Disasters
trapping

All Science Journal Classification (ASJC) codes

  • Building and Construction
  • Geotechnical Engineering and Engineering Geology

Cite this

@article{09841e6dce764131a5117d402204d35a,
title = "Predicting tunnel convergence using Multivariate Adaptive Regression Spline and Artificial Neural Network",
abstract = "Determining the tunnel convergence is an indispensable task in tunneling, especially when adopting the New Austrian Tunneling Method. The interpretation of the monitoring allows adjusting the construction methods in order to achieve more effective tunneling conditions and to avoid problems like rock collapse, trapping and jamming of boring machine, delay of the project or even geological disasters. In this research, a model capable of predicting the diameter convergence of a high-speed railway tunnel in weak rock was established based on two approaches: Multivariate Adaptive Regression Spline (MARS) and Artificial Neural Network (ANN). A tunnel construction project located in Hunan province (China) was used as case study. The input parameters included the class index of the surrounding rock mass, angle of internal friction, cohesion, Young's modulus, rock density, tunnel overburden, distance between the monitoring station and the tunnel heading face and the elapsed monitoring time. The performance of the models was evaluated by comparing the predicted convergence to the measured data using several performance indices. Overall, the results showed high accuracy of the model predictability of tunnel convergence with MARS showing a light lesser accuracy. However, MARS was more flexible and computationally efficient. It is concluded that MARS can constitute a reliable alternative to ANN in modeling nonlinear geo-engineering problem such as the tunnel convergence.",
author = "Adoko, {Amoussou Coffi} and Jiao, {Yu Yong} and Li Wu and Hao Wang and Wang, {Zi Hao}",
year = "2013",
month = "9",
day = "1",
doi = "10.1016/j.tust.2013.07.023",
language = "English",
volume = "38",
pages = "368--376",
journal = "Tunnelling and Underground Space Technology",
issn = "0886-7798",
publisher = "Elsevier Limited",

}

Predicting tunnel convergence using Multivariate Adaptive Regression Spline and Artificial Neural Network. / Adoko, Amoussou Coffi; Jiao, Yu Yong; Wu, Li; Wang, Hao; Wang, Zi Hao.

In: Tunnelling and Underground Space Technology, Vol. 38, 01.09.2013, p. 368-376.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Predicting tunnel convergence using Multivariate Adaptive Regression Spline and Artificial Neural Network

AU - Adoko, Amoussou Coffi

AU - Jiao, Yu Yong

AU - Wu, Li

AU - Wang, Hao

AU - Wang, Zi Hao

PY - 2013/9/1

Y1 - 2013/9/1

N2 - Determining the tunnel convergence is an indispensable task in tunneling, especially when adopting the New Austrian Tunneling Method. The interpretation of the monitoring allows adjusting the construction methods in order to achieve more effective tunneling conditions and to avoid problems like rock collapse, trapping and jamming of boring machine, delay of the project or even geological disasters. In this research, a model capable of predicting the diameter convergence of a high-speed railway tunnel in weak rock was established based on two approaches: Multivariate Adaptive Regression Spline (MARS) and Artificial Neural Network (ANN). A tunnel construction project located in Hunan province (China) was used as case study. The input parameters included the class index of the surrounding rock mass, angle of internal friction, cohesion, Young's modulus, rock density, tunnel overburden, distance between the monitoring station and the tunnel heading face and the elapsed monitoring time. The performance of the models was evaluated by comparing the predicted convergence to the measured data using several performance indices. Overall, the results showed high accuracy of the model predictability of tunnel convergence with MARS showing a light lesser accuracy. However, MARS was more flexible and computationally efficient. It is concluded that MARS can constitute a reliable alternative to ANN in modeling nonlinear geo-engineering problem such as the tunnel convergence.

AB - Determining the tunnel convergence is an indispensable task in tunneling, especially when adopting the New Austrian Tunneling Method. The interpretation of the monitoring allows adjusting the construction methods in order to achieve more effective tunneling conditions and to avoid problems like rock collapse, trapping and jamming of boring machine, delay of the project or even geological disasters. In this research, a model capable of predicting the diameter convergence of a high-speed railway tunnel in weak rock was established based on two approaches: Multivariate Adaptive Regression Spline (MARS) and Artificial Neural Network (ANN). A tunnel construction project located in Hunan province (China) was used as case study. The input parameters included the class index of the surrounding rock mass, angle of internal friction, cohesion, Young's modulus, rock density, tunnel overburden, distance between the monitoring station and the tunnel heading face and the elapsed monitoring time. The performance of the models was evaluated by comparing the predicted convergence to the measured data using several performance indices. Overall, the results showed high accuracy of the model predictability of tunnel convergence with MARS showing a light lesser accuracy. However, MARS was more flexible and computationally efficient. It is concluded that MARS can constitute a reliable alternative to ANN in modeling nonlinear geo-engineering problem such as the tunnel convergence.

UR - http://www.scopus.com/inward/record.url?scp=84882785468&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84882785468&partnerID=8YFLogxK

U2 - 10.1016/j.tust.2013.07.023

DO - 10.1016/j.tust.2013.07.023

M3 - Article

VL - 38

SP - 368

EP - 376

JO - Tunnelling and Underground Space Technology

JF - Tunnelling and Underground Space Technology

SN - 0886-7798

ER -