Small strain stiffness within logarithmic contractancy model for structured anisotropic clay

N Sivasithamparam,M D’Ignazio,C Madshus,J Castro,A B Tsegaye

doi:10.1088/1755-1315/710/1/012042

Abstract

Stiffness of soils in the small strain region is high and it decays nonlinearly with increasing shear strains or with mobilization of shear stresses. However, the commonly used critical state based constitutive models use a simple elastic formulation at small strains that falls short in the prediction of the small strain nonlinearity and anisotropy. This paper proposes a simple way for rendering the existing constitutive models with the capability to capture the small strain behaviour of soils. This is illustrated by proposing a new model for structured anisotropic clay extending an existing model that uses the framework of logarithmic contractancy called E-SCLAY1S. The proposed model is implemented into a Finite Element program as a user-defined soil model. The model predictions are compared with experimental data for various clays. Furthermore, the effect of nonlinearity is investigated for an excavation in soft clay.

Highlights

For robust design of retaining wall system in the serviceability limit state, traditional design methods such as limit equilibrium or empirical and semi empirical methods are not suitable
This paper proposes a simple way for rendering the existing constitutive models with the capability to capture the small strain behaviour of soils
This is to be attributed to the shear strain induced by the excavation, which are in the order of 10-3

Summary

Introduction

For robust design of retaining wall system in the serviceability limit state, traditional design methods such as limit equilibrium or empirical and semi empirical methods are not suitable. Ground settlement induced by excavation is more difficult to accurately predict than wall deflection as it requires proper modelling of soil behaviour at the level of small strains ([1], [2]). Given these issues, the use of finite element (FE) analysis with an accurate constitutive model that can capture small strain behaviour can be a solution. The results are compared with that of the reference model and differences and gained benefits are discussed

Literature review

The E-SCLAY1S model

Model performance in a FE boundary problem