Examination for predicting ground settlement based on measurement records by using a neural network model

M. Kanayama, Y. Okamura, A. Rohe, Leon van Paassen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Earth-fill structures such as embankments, which are constructed for the preservation of land and infrastructure, show significant amount of settlement during and after construction in lowland areas. The long term settlement of those structures is often measured. In this paper, we examined the applicability of a neural network model for settlement prediction using measurements in the early stage after construction. Simulations using a basic network model showed that when the measurement data used for teaching the neural network accumulated, the prediction was in good agreement with the measurement data, and the variance of predicted values was low. However, the basic model could not predict the settlement behaviour precisely, when the amount of teach data was limited as would be in the early stage after construction. Some improvement was necessary for this model to conduct early settlement prediction. To achieve a higher accuracy in long term settlement prediction from early stage measurements, several improvements to the model are proposed, which generate additional data points and improve the prediction accuracy. Firstly, a cubic spline interpolation technique is used to generate additional data between measurements and regulate the input to constant time intervals. Secondly, statistical techniques are used to weed out predicted data points that are outside preset parameters. Short-term predicted values that satisfy with clear statistical criteria (low co-variance or low standard deviation) are added to the network teach data. The improved network model simulations showed that the accuracy of settlement prediction based on early stage measurements improved significantly.

Original languageEnglish (US)
Title of host publicationISRM International Symposium - 8th Asian Rock Mechanics Symposium, ARMS 2014
Editors Kaneko, Kodama, Shimizu
PublisherInternational Society for Rock Mechanics
Pages2337-2345
Number of pages9
ISBN (Electronic)9784907430030
StatePublished - Jan 1 2014
Externally publishedYes
Event8th Asian Rock Mechanics Symposium, ARMS 2014 - Sapporo, Japan
Duration: Oct 14 2014Oct 16 2014

Other

Other8th Asian Rock Mechanics Symposium, ARMS 2014
CountryJapan
CitySapporo
Period10/14/1410/16/14

Fingerprint

ground settlement
examination
Neural networks
predictions
prediction
Settlement of structures
Embankments
splines
Splines
time constant
interpolation
standard deviation
embankment
Interpolation
Teaching
education
teaching
simulation
weed
Earth (planet)

Keywords

  • Measurement record
  • Neural network
  • Observational method
  • Settlement prediction
  • Soft ground

ASJC Scopus subject areas

  • Geophysics
  • Geochemistry and Petrology

Cite this

Kanayama, M., Okamura, Y., Rohe, A., & van Paassen, L. (2014). Examination for predicting ground settlement based on measurement records by using a neural network model. In Kaneko, Kodama, & Shimizu (Eds.), ISRM International Symposium - 8th Asian Rock Mechanics Symposium, ARMS 2014 (pp. 2337-2345). International Society for Rock Mechanics.

Examination for predicting ground settlement based on measurement records by using a neural network model. / Kanayama, M.; Okamura, Y.; Rohe, A.; van Paassen, Leon.

ISRM International Symposium - 8th Asian Rock Mechanics Symposium, ARMS 2014. ed. / Kaneko; Kodama; Shimizu. International Society for Rock Mechanics, 2014. p. 2337-2345.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kanayama, M, Okamura, Y, Rohe, A & van Paassen, L 2014, Examination for predicting ground settlement based on measurement records by using a neural network model. in Kaneko, Kodama & Shimizu (eds), ISRM International Symposium - 8th Asian Rock Mechanics Symposium, ARMS 2014. International Society for Rock Mechanics, pp. 2337-2345, 8th Asian Rock Mechanics Symposium, ARMS 2014, Sapporo, Japan, 10/14/14.
Kanayama M, Okamura Y, Rohe A, van Paassen L. Examination for predicting ground settlement based on measurement records by using a neural network model. In Kaneko, Kodama, Shimizu, editors, ISRM International Symposium - 8th Asian Rock Mechanics Symposium, ARMS 2014. International Society for Rock Mechanics. 2014. p. 2337-2345
Kanayama, M. ; Okamura, Y. ; Rohe, A. ; van Paassen, Leon. / Examination for predicting ground settlement based on measurement records by using a neural network model. ISRM International Symposium - 8th Asian Rock Mechanics Symposium, ARMS 2014. editor / Kaneko ; Kodama ; Shimizu. International Society for Rock Mechanics, 2014. pp. 2337-2345
@inproceedings{16cd171be61648b6947ee3a394a7f081,
title = "Examination for predicting ground settlement based on measurement records by using a neural network model",
abstract = "Earth-fill structures such as embankments, which are constructed for the preservation of land and infrastructure, show significant amount of settlement during and after construction in lowland areas. The long term settlement of those structures is often measured. In this paper, we examined the applicability of a neural network model for settlement prediction using measurements in the early stage after construction. Simulations using a basic network model showed that when the measurement data used for teaching the neural network accumulated, the prediction was in good agreement with the measurement data, and the variance of predicted values was low. However, the basic model could not predict the settlement behaviour precisely, when the amount of teach data was limited as would be in the early stage after construction. Some improvement was necessary for this model to conduct early settlement prediction. To achieve a higher accuracy in long term settlement prediction from early stage measurements, several improvements to the model are proposed, which generate additional data points and improve the prediction accuracy. Firstly, a cubic spline interpolation technique is used to generate additional data between measurements and regulate the input to constant time intervals. Secondly, statistical techniques are used to weed out predicted data points that are outside preset parameters. Short-term predicted values that satisfy with clear statistical criteria (low co-variance or low standard deviation) are added to the network teach data. The improved network model simulations showed that the accuracy of settlement prediction based on early stage measurements improved significantly.",
keywords = "Measurement record, Neural network, Observational method, Settlement prediction, Soft ground",
author = "M. Kanayama and Y. Okamura and A. Rohe and {van Paassen}, Leon",
year = "2014",
month = "1",
day = "1",
language = "English (US)",
pages = "2337--2345",
editor = "Kaneko and Kodama and Shimizu",
booktitle = "ISRM International Symposium - 8th Asian Rock Mechanics Symposium, ARMS 2014",
publisher = "International Society for Rock Mechanics",

}

TY - GEN

T1 - Examination for predicting ground settlement based on measurement records by using a neural network model

AU - Kanayama, M.

AU - Okamura, Y.

AU - Rohe, A.

AU - van Paassen, Leon

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Earth-fill structures such as embankments, which are constructed for the preservation of land and infrastructure, show significant amount of settlement during and after construction in lowland areas. The long term settlement of those structures is often measured. In this paper, we examined the applicability of a neural network model for settlement prediction using measurements in the early stage after construction. Simulations using a basic network model showed that when the measurement data used for teaching the neural network accumulated, the prediction was in good agreement with the measurement data, and the variance of predicted values was low. However, the basic model could not predict the settlement behaviour precisely, when the amount of teach data was limited as would be in the early stage after construction. Some improvement was necessary for this model to conduct early settlement prediction. To achieve a higher accuracy in long term settlement prediction from early stage measurements, several improvements to the model are proposed, which generate additional data points and improve the prediction accuracy. Firstly, a cubic spline interpolation technique is used to generate additional data between measurements and regulate the input to constant time intervals. Secondly, statistical techniques are used to weed out predicted data points that are outside preset parameters. Short-term predicted values that satisfy with clear statistical criteria (low co-variance or low standard deviation) are added to the network teach data. The improved network model simulations showed that the accuracy of settlement prediction based on early stage measurements improved significantly.

AB - Earth-fill structures such as embankments, which are constructed for the preservation of land and infrastructure, show significant amount of settlement during and after construction in lowland areas. The long term settlement of those structures is often measured. In this paper, we examined the applicability of a neural network model for settlement prediction using measurements in the early stage after construction. Simulations using a basic network model showed that when the measurement data used for teaching the neural network accumulated, the prediction was in good agreement with the measurement data, and the variance of predicted values was low. However, the basic model could not predict the settlement behaviour precisely, when the amount of teach data was limited as would be in the early stage after construction. Some improvement was necessary for this model to conduct early settlement prediction. To achieve a higher accuracy in long term settlement prediction from early stage measurements, several improvements to the model are proposed, which generate additional data points and improve the prediction accuracy. Firstly, a cubic spline interpolation technique is used to generate additional data between measurements and regulate the input to constant time intervals. Secondly, statistical techniques are used to weed out predicted data points that are outside preset parameters. Short-term predicted values that satisfy with clear statistical criteria (low co-variance or low standard deviation) are added to the network teach data. The improved network model simulations showed that the accuracy of settlement prediction based on early stage measurements improved significantly.

KW - Measurement record

KW - Neural network

KW - Observational method

KW - Settlement prediction

KW - Soft ground

UR - http://www.scopus.com/inward/record.url?scp=84962415127&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84962415127&partnerID=8YFLogxK

M3 - Conference contribution

SP - 2337

EP - 2345

BT - ISRM International Symposium - 8th Asian Rock Mechanics Symposium, ARMS 2014

A2 - Kaneko, null

A2 - Kodama, null

A2 - Shimizu, null

PB - International Society for Rock Mechanics

ER -