|Title:||Development of a new three-dimensional anisotropic Elastic Visco-Plastic model for natural soft soils and applications in deformation analysis|
|Subject:||Hong Kong Polytechnic University -- Dissertations|
|Department:||Department of Civil and Structural Engineering|
|Pages:||xxviii, 280 leaves : ill. ; 30 cm|
|Abstract:||This study focuses on three interrelated aspects: (a) development of anisotropic Elastic Visco-Plastic (EVP) constitutive models; (b) verification of the EVP models by comparing simulation results with measured results from four types of tests on a Hong Kong Marine Clay (HKMC), and (c) application of the constitutive models in fully coupled consolidation analyses of case histories. The four types of tests are (i) standard oedometer compression/creep tests, (ii) oedometer tests compressed at a single stage of constant strain rate or stepped changed strain rates, (iii) triaxial K0 -consolidation tests, and (iv) K0 -consolidated undrained triaxial compression and extension tests at multi-stage constant strain rates. The case histories include a test embankment at the new Chek Lap Kok International Airport in Hong Kong, Haarajoki test embankment in Finland and a small-scale physical model soft clay foundation improved by Deep Cement Mixed (DCM) soil column. First, a new general one-dimensional (1-D) EVP constitutive model using a non-linear creep function is developed and presented for the time-dependent stress-strain behaviour of clays under a K0 -stress condition including loading, unloading and reloading. This new 1-D EVP model is an extension of the 1-D EVP model proposed by Yin and Graham (1989, 1994) based on the 'equivalent time' concept. The new elements in the newly developed 1-D EVP model are that (a) a non-linear creep function with a limit for the creep volumetric strain is adopted and (b) a 'limit time' line is used. A method for determining all model parameters is presented using data from one conventional standard oedometer compression/creep test on natural HKMC. The l-D EVP model is verified by comparing modeling results with measured results from (a) conventional standard oedometer compression/creep tests, (b) oedometer tests compressed at a single stage of constant strain rate or stepped changed strain rate and (c) triaxial K0 -consolidation/compression tests on the HKMC. A three dimensional (3-D) anisotropic EVP model for normally consolidated and overconsolidated clays is then developed on the basis of the 1-D EVP constitutive relationship. This 1-D EVP relationship is used to determine the visco-plastic scaling factor and is taken as a strain hardening law similar to that used in the Modified Cam-Clay model. Yin and Graham's 'equivalent time', 'limit time line' concepts and a non-linear creep function with a limit for the creep volumetric strain are also incorporated in the 3-D EVP model. A symmetrical elliptical loading locus along an 帢K0 -line in the p - q plane is incorporated in the 3-D anisotropic EVP model to initiate the field K0 - stress condition. Based on the visco-plasticity theory and limit time line concept, the nonlinear time-dependent stress-strain relationship under loading or unloading condition, can be modeled using the EVP model. The particular strain softening and pore water pressure decreasing behavior in the multi-stage constant strain rate triaxial tests can be modeled using a dilation stress ratio Md instead of the critical state strength parameter M as used in the Modified Cam-Clay model. Most model parameters in the 3-D EVP model are the same as the 1-D EVP model determined using data from a set of aforementioned conventional oedometer compression and creep test. The critical state strength parameter M is determined using consolidated undrained triaxial compression tests on undisturbed soft HKMC. The 3-D anisotropic EVP model is verified by comparing modeling results with measured results from undrained triaxial compression and extension tests on K0 -consolidated HKMC at multi-stage constant stain rates, which are from +2 %/hr to +0.2 %/hr, +20 %/hr, -2 %/hr (unloading) and +2 %/hr (reloading) for compression tests; or from -2 %/hr to -0.2 %/hr, -20 %/hr, +2 %/hr (unloading) and -2 %/hr (reloading) for extension tests. The highly non-linear 3-D anisotropic EVP model is incorporated in a finite element code based on Biot's fully coupled consolidation theory. An elastic visco-plastic finite element analysis is carried out to model consolidation of the foundation soils with drains under a test embankment at the new Chek Lap Kok International Airport in Hong Kong. The EVP finite element analysis is conducted on one unit cell around a drain. The EVP model takes account of the inherent K0 -stress anisotropy, subsequent stress induced anisotropy and coupling mechanism between creep deformation of the clay skeleton and consolidation of the clay layer. The 3-D anisotropic EVP model is used for two layers of soft marine clay and two layers of alluvium clay at the site. The material parameters are derived from results in the original site investigation report. The nonlinear nature of hydraulic permeability of soft soils is taken into consideration. Numerical results are in good agreement with settlements and pore water pressures measured under the test embankment. The comparison demonstrates that the finite element approach using the 3-D EVP model is appropriate for analyzing the consolidation of soft soils exhibiting creep with vertical drains. The highly non-linear 3-D EVP model in combination with the Modified Cam-Clay model is also incorporated in a fully coupled consolidation analysis of the soils underneath the Haarajoki test embankment in Finland. The 3-D EVP model is used for two layers of soft clay at the site, while the Modified Cam-Clay model is used for an upper layer of stiff weathered clay and a lower level of silt. Vertical drains are modeled as an equivalent plane strain boundary problem in the elastic visco-plastic finite element analysis. Consolidation analyses are also conducted for the clay layers underneath the Haarajoki Test Embankment in Finland using a soft soil model (the Modified Cam-Clay model) in a finite element program PLAXIS. Elastic visco-plastic numerical results are in better agreement with settlements and pore water pressures measured under the test embankment than PLAXIS does using the soft soil model. The comparison demonstrates that the finite element approach using the 3-D EVP and Modified Cam Clay models and the equivalent plane strain modeling method for vertical drains is more appropriate for analyzing the consolidation of soft soils exhibiting creep both with vertical drains. The analysis also investigates the effects of the top crust of stiff clay, and the spacing and penetrating depth of the vertical drains. The method of Deep Cement Mixing (DCM) is a deep in-situ stabilization technique of mixing cement power or slurry with soft soils. The DCM method has been widely applied to improve soft soils in many countries and regions. A few researches have been performed on the strength, stiffness and stress-strain behavior of the cement mixed soil in recent decades. However, few studies have been conducted on the consolidation behavior of soft clay treated by DCM soil columns. In order to study the consolidation behavior of the DCM soil column improved soft clay, a small-scale physical model test was carried out on a DCM soil composite foundation. Elastic visco-plastic finite element analysis incorporated with the aforementioned highly non-linear 3-D anisotropic EVP model is performed on the small-scale physical model DCM soil composite foundation. The 3-D anisotropic EVP model is used for the Unimproved Soft Clay (USC), and a hyperbolic model is used for DCM soil column before and after local failure. A simple yet workable method is presented for finite element solution to local failure induced strain softening of DCM soil column. The nonlinear nature of hydraulic permeability of USC is taken into consideration. The predicted and measured data of settlement and excess pore water pressure are compared, interpreted and discussed. Numerical results are in good agreement with measured data. The comparison demonstrates that the finite element approach using the 3-D EVP model and the hyperbolic model is appropriate for analyzing the consolidation of soft soils exhibiting creep improved by DCM soil column. In the appendix, visco-elastic plastic analytical solutions to the hollow cylindrical infinite boundary problems are also derived. Some natural soils can be simplified as ideal visco-elastic-brittle-plastic and linear dilative strain softening materials in order to derive analytical solutions. Based on the axisymmetric equilibrium equation, geometric strain equation and the Mohr-Coulomb failure criterion, the visco-elastic plastic analytical solutions to the hollow cylindrical infinite boundary problems are derived for natural soils using Kelvin's and Maxwell's viscous elastic models and a linear dilative relationship. The effects of different initial stress, damage ratio (or residual frictional angle) and dilative deformation behavior, on the development of strain softening zones and time dependent strain softening deformation and strain, are investigated using the analytical solutions. These solutions may be useful reference for study of long-term tunnelling excavation, open-end caisson construction and pressuremeter test data interpretation.|
|Rights:||All rights reserved|
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item: