Skip to Main content Skip to Navigation
Theses

Analyse Inverse en Géotechnique: développement d'une méthode à base d'algorithmes génétiques.

Abstract : Most standard geotechnical in situ tests do not allow the direct identification of the constitutive parameters of the soil layers. The use of the finite element method for the designing geotechnical structures is consequently strongly limited by the rough knowledge of the mechanical properties of soil. This fact leads to set the problematics of inverse analysis in geotechnical engineering: What information concerning the constitutive parameters of the soil is it possible to get from in situ measurement ?
This study is dedicated to the identification of parameters of soil constitutive models by inverse analysis from in situ geotechnical measurements. The inverse analysis is based on a genetic algorithm optimization process. Furthermore, to describe the set of solutions identified from a genetic algorithm optimization, a statistical analysis based on Principle Component Analysis is introduced. This analysis provides a set of solutions of the inverse problem. If all solutions are not directly identify by genetic algorithm, a principle component analysis on these solutions permits to define an envelop in which all the parameter values satisfy the inverse problem. These developments are based on some synthetic excavation problems and pressuremeter tests.
Document type :
Theses
Complete list of metadatas

Cited literature [160 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-00185671
Contributor : Séverine Levasseur <>
Submitted on : Tuesday, November 6, 2007 - 4:15:36 PM
Last modification on : Thursday, November 19, 2020 - 3:54:38 PM
Long-term archiving on: : Monday, April 12, 2010 - 1:31:26 AM

Identifiers

  • HAL Id : tel-00185671, version 1

Collections

CNRS | UJF | 3S-R | UGA

Citation

Séverine Levasseur. Analyse Inverse en Géotechnique: développement d'une méthode à base d'algorithmes génétiques.. Sciences de l'ingénieur [physics]. Université Joseph-Fourier - Grenoble I, 2007. Français. ⟨tel-00185671⟩

Share

Metrics

Record views

1236

Files downloads

3020