Fook-Hou Lee

ELASTO-PLASTIC CONSOLIDATION ANALYSIS FOR STRUTTED EXCAVATION IN CLAY

Abstract Ground movements and strut loads in strutted excavations in clay have been observed to change with time. In this paper, the time-dependent behaviour of excavation support system is studied by comparing the results of undrained and consolidation analyses with data from an instrumented excavation project. Dissipation of excess pore pressure is modelled using a fully coupled consolidation analysis while the soil is assumed to be an elastic-perfectly plastic material obeying the Mohr-Coulomb yield criterion. The results of the study show that the undrained analysis underestimates the sheet pile wall movement and fail to reflect the progressive movement of the sheet pile. In contrast, these effects are well-predicted by the consolidation analysis, thereby indicating that dissipation of excess negative pore pressure can indeed account for much of the observed progressive ground movement and build-up of strut loads with time. The elasto-plastic consolidation model can also simulate excavation sequence including uneven excavation and time delays in excavation and strutting.

Modified isotropic compression relationship for cement-Admixed marine clay at low confining stress

This technical note proposes a modified log-linear compression relationship for the pre-yield isotropic compression behavior of cement-admixed marine clay. The difference between this modified relationship and the conventional log-linear relationship is the use of the tensile strength as an added stress to the mean effective stress. This allows the cement-admixed clay to retain a finite specific volume as the mean effective stress approaches zero, which is more consistent with the fact that cement-admixed clays remain intact even under zero effective stress condition. Comparison with data from isotropic compression test shows that the addition of the tensile strength to the mean effective stress leads to a remarkable improvement in linearity, indicating a much-improved fit to the log-linear relationship. The resulting pre-yield compression index implies a non-zero equivalent effective bulk modulus, which is also intuitively more reasonable. At effective stress levels much lower than the tensile strength, the bulk modulus is approximately constant; this is also consistent with the notion of linear elastic behavior at low stress level.

A generalised Jacobi preconditioner for finite element solution of large-scale consolidation problems

Publisher Summary Finite element simulations of very large-scale soil–structure interaction problems typically involve the solution of a very large, ill-conditioned, and indefinite Biot system of equations. The traditional preconditioned conjugate gradient solver coupled with the standard Jacobi (SJ) preconditioner is inefficient for this class of problems. This chapter presents a robust generalized Jacobi (GJ) preconditioner that is extremely effective for solving very large-scale Biots finite element equations using the symmetric quasi-minimal residual method. It was derived as a diagonal approximation to a theoretical form, which can be proven mathematically to possess an attractive eigenvalue clustering property. The GJ preconditioner is formed, inverted, and implemented within an element-by-element framework as readily as the SJ preconditioner. Significantly better result can be achieved by applying a small negative scalar to the block of the generalized Jacobi preconditioner.

Failure Modes of Tunnels with Improved Soil Surrounds

AbstractThis paper presents the study of possible failure mechanisms and stability of large-diameter tunnels in cement-treated soil surrounds. The failure modes and critical tunnel support pressure...

A Design Framework for Spatial Variability in Cement-Treated Soft Clay in Deep Excavations and Underground Constructions

This paper summarizes some of the recent works on spatial variability of cement-treated clay for underground construction. The studies were conducted using deterministic and random finite element analyses, calibrated and validated using centrifuge model test data. An approach to estimating a ‘design strength’ of an equivalent homogeneous material is proposed, based on the notion of reducing the mean strength of the spatially variable soil domain by a multiple, termed reduction factor, of its standard deviation. The results suggest that the value of the reduction factor depends upon the stresses arising from external loading. For highly uniform stress situations such as that in a cement-treated soil slab, the reduction factor is relatively low. On the other hand, for non-uniform stress situations such as those in cement-treated soil rings and heading, the reduction factor is substantially higher. This implies that the non-uniform stress distribution can aggravate the effect of material spatial variability, and vice versa. An issue which is still receiving on-going attention is the interaction between material model and spatial variability. Results to-date indicate that the Mohr–Coulomb model tends to err on the unsafe side. If this is so, then Mohr–Coulomb parameters may need to be factored down accordingly.

A NEW FINITE ELEMENT MODEL FOR PILE-SOIL INTERACTION

This paper describes the formulation and validation of a new finite element model for 3-D analysis of pile-soil interaction problems. The proposed element is constructed by wrapping 4 slip elements around a 2-nodes flexural element. The 4 slip elements are to model the interface between soil and pile and the 2-node flexural element simulates the pile segment. This element will lead to an ease of use in the 3-D numerical analysis of soil-pile interaction problems with slip elements being built in one element and a significant reduction of numbers of system Degree of Freedom comparing to the conventional way of modeling pile with 20-node brick elements. The element stiffness formulation is based on the weak formulation of virtual work principle with the internal equilibrium between pile and soil being governed by Eulers beam theory in lateral reaction. The FEM program using the proposed model was developed and verified and validated with analytical solutions and a simple case study.

Large-scale finite element analyses of intersecting tunnels using PCs

Publisher Summary This chapter presents the settlement analyses of intersecting tunnels using a 3-D finite element method. The extension of underground transportation systems in urban areas often involves the construction of new tunnels above/below existing tunnels at varying angles of intersection. To ensure that the existing tunnel continues to comply with serviceability requirements, it is important to simulate the influence of construction on the behavior of the existing tunnel. Intersecting tunnels are irreducible 3-D problems—their soil-structure interaction mechanisms and stress-deformation fields can only be modeled realistically using a 3-D finite element mesh. If the surrounding soil exhibits time-dependent behavior, consolidation effects have to be incorporated into the analyses as well. The cost of 3-D finite element analyses incorporating consolidation is generally prohibitive to average practitioners in terms of run time and memory requirements.

Cyclic mobility response of a dense sand stratum

Abstract The cyclic mobility response of a dense sand stratum is examined using a semi-empirical effective stress model. The model is based on two families of intersecting yield loci, one for loading and the other for unloading. The yield loci are generated from Cam Clay type energy balance equations and the associated flow rule. Comparison of model with observed behaviour in cyclic simple shearing indicates that, although the model is unable to reflect the full spectrum of sand behaviour in cyclic loading, the salient features of cyclic mobility behaviour are adequately represented. The model is then applied to the problem of an infinitely long, sloping, saturated dense sand stratum subjected to base excitation. Comparison of computed results with centrifuge data shows that the model is able to explain much of the observed behaviour in terms of the interaction between soil skeleton, pore water and propagating shear waves. Such interaction is not fully considered in existing methods of designs.