Search

Soil–vegetation–atmosphere interaction has long been known to induce significant pore pressure variations at shallow depths and associated superficial slope movements. Recent findings suggest that the effect of this interaction may also extend to large depths in natural clay slopes. Multiple examples of weather-induced deep landslide mechanisms can be found in the Southern Apennines (Italy), where slopes are often formed of fissured clays. The relationship between the activity of these landslides and the hydro-mechanical processes due to soil–vegetation–atmosphere interaction was investigated herein by means of a two-dimensional coupled hydro-mechanical finite element analysis. A constitutive model capable of simulating the behaviour of highly overconsolidated clays, in both saturated and unsaturated states, was adopted in the analysis, in conjunction with a boundary condition capable of reproducing the combined effects of rainfall infiltration, evapo-transpiration, and run-off. The results of the analysis corroborate the connection between weather conditions, pore pressure variations, and slope movements in natural clay slopes. The importance of adequately reproducing the geological history of a natural slope to define its current state is also demonstrated. [ABSTRACT FROM AUTHOR]

Seawalls are constructed by covering and protecting the sloping seashore ground with concrete plates or blocks. Their purpose is to sustain high waves induced by strong winds and prevent ground erosion, but they often collapse, mobilising different modes of failure, including that of the ground. Nevertheless, limited research has been conducted on ground failure caused by high waves. In this study, a series of novel centrifuge model tests was first conducted to investigate the failure mechanisms of seawalls due to wave propagation, focusing on the failure of the ground. Finite-element analyses were subsequently conducted to explore the failures observed in the model tests and to provide further insight as to the state of the ground leading to failure. Two failure modes were demonstrated to prevail: floating of the covering panel and sliding failure of the ground. In addition, of the possible causes of failure, the following three were identified in the current study: increased unit weight and reduced suction from wetting; enhanced seepage force under the panel and around the toe block during backwash; water pressure on the back of the panel and the landward side of the toe block during backwash. [ABSTRACT FROM AUTHOR]

The present-day condition of a grey cast iron (GCI) tunnel was investigated numerically by means of a series of plane-strain geotechnical analyses that focused on establishing the influence of the modelling approach adopted to simulate the behaviour of longitudinal tunnel joints. The GCI tunnel lining was simulated in three different ways: (i) as a continuous lining; (ii) as a jointed lining adopting a simple joint model that neglects the presence of the joint bolts and (iii) as a jointed lining adopting a more sophisticated model that considers both the nonlinear response of the joint after opening and the contribution of the bolts to the ultimate capacity. The numerical results demonstrate that accounting for the tunnel joints has a considerable impact on the tunnel response. Significant differences are also observed in the results obtained with the two joint models, hence demonstrating the practical significance of employing an advanced model able to reproduce accurately the rotational behaviour of the joint. The drainage behaviour of the joints was also investigated and the numerical results indicate that drainage even through only the knee joints causes significantly more soil consolidation than that from a watertight tunnel and consequently, larger tunnel deformations and forces. [ABSTRACT FROM AUTHOR]

The structural assessment of segmental grey cast iron (GCI) tunnel linings to nearby construction is challenging due to the presence of the joints affecting the stiffness of the tunnel lining. This paper presents an extensive investigation, using 3D finite element (FE) analyses, into the bending moment-rotation (M-θ) behaviour of two GCI tunnel joint geometries. These two geometries correspond to standard running and station tunnels of the London Underground network. The contribution of this study is two-fold. i) The novel characterisation of the M-θ response enables the development of new models for simulating the mechanical response of GCI tunnel joints with structural elements which can be used in simplified, 2D geotechnical analysis for tunnel safety assessments. ii) The analyses provide insight into the behaviour of GCI tunnel linings that would be difficult to achieve through experimental and field observations alone. More specifically, the analyses show that when the bolts are removed from the joints the possibility of tensile failure can be disregarded; that the initial bolt preload influences the rotational stiffness only after some rotation has taken place and does not alter the bending moment of opening; and that the out-of-plane displacement restraint has little influence on the joint response. [ABSTRACT FROM AUTHOR]

The structural testing and finite element (FE) analysis described in this paper were part of a major research project undertaken at Imperial College London to investigate the deformation of bolted segmental grey cast iron (GCI) tunnel linings. A key aim was to quantify how joints influence the behaviour of the lining, through a three-path approach comprising physical experiments, finite element modelling, and field instrumentation. The laboratory results have been used to assess the validity of the tunnel assessment methods used by industry. This study examined joint articulation under the serviceability limit state in the absence of hoop force focussing on factors such as applied bolt preload, the loading direction and the freedom of the circumferential flange to deflect. Two half-scale GCI lining segments were bolted together at the longitudinal flanges to form a bolted arch in a similar fashion to the tests performed by Thomas (1977). Modern instrumentation was implemented to gain detailed measurements quantifying changes in global displacements of the two segments, bolt forces and joint opening under applied loading. For the first time, the physical experiments were conducted contemporaneously with the development of a three-dimensional FE model of the joint. The experimental data and the results from the FE analysis indicate a reduction in joint stiffness as the joint articulates under applied load. It is shown that the presence of a joint has far greater influence on the behaviour of the ‘arch’ than the level of preload applied to the bolts in the joint. The FE analysis allowed the deformation behaviour of the joint under positive and negative bending to be investigated: its response under the two modes differs significantly. [ABSTRACT FROM AUTHOR]

The paper presents the Finite Element formulation of the equations proposed by Tsiampousi et al. (2016) for coupled consolidation in unsaturated soils. Their coupling is discussed in relation to a conceptual model which divides soil behaviour into zones ranging from fully saturated to dry states. The numerical simulation of a laboratory experiment involving drainage of water from a vertical column of sand is used to validate the equations. Finally, the example of rainfall infiltration into a cut slope highlights how aspects of the conceptual model are reflected in the numerical analysis of boundary value problems involving unsaturated soils. [ABSTRACT FROM AUTHOR]

The equations governing coupled consolidation in unsaturated soils are known to contain additional parameters when compared to the equations for saturated soils. Nonetheless, the variation of these parameters with suction or degree of saturation is not generally agreed upon. The paper introduces a novel approach to deriving general equations for each of these parameters and their variation, and explains that, for consistency with the constitutive and soil-water retention curve models adopted, these general equations need to be transformed into case-specific expressions. Finally, a conceptual model is presented highlighting how the behaviour of unsaturated soil reflects aspects of its water content. [ABSTRACT FROM AUTHOR]

The structural response of a segmental grey cast iron (GCI) tunnel lining ring under distortion was investigated by means of finite element (FE) analysis. Building on previous experimental investigations, a 3D numerical model, capable of reproducing accurately the behaviour observed in the laboratory, was developed with the aim of providing guidelines for the structural assessment of GCI linings in engineering practice. A comprehensive validation of the segmental ring model with the laboratory data was first completed. Subsequently, a parametric study was conducted using a set-up that replicated the widely adopted elastic continuum method, so that differences between the numerical and the analytical solution could be attributed to the presence of the longitudinal joints. In this manner, the influence of the joints on the ring response was quantitatively established and recommendations for routine engineering calculations developed. A set of bending stiffness reduction factors are proposed as a function of the tunnel ovalisation, providing upper and lower limits of the bending stiffness, as well as a global reduction factor which is an average measure of the bending stiffness reduction. These factors can be integrated into the calculation procedure of closed-form solutions in order to account for the segmental nature of GCI linings. [ABSTRACT FROM AUTHOR]

Abstract: While the evolution with time of the factor of safety against slope failure is well understood for excavations performed under fully saturated conditions, this evolution has not been extensively studied for excavations carried out in unsaturated soils. The objective of this paper is to investigate the relationship between the factor of safety and time for an excavation performed in an unsaturated silty soil, employing the finite element method. A hypothetical boundary value problem is considered and two types of analyses performed; in the first one, unsaturated soil behaviour is modelled through appropriate constitutive and soil–water retention curve models, whereas in the second one full saturation is assumed. The effect that the saturated soil permeability, its variation with suction, the increase of apparent cohesion due to suction, the depth of the groundwater table and the hydraulic hysteresis have on the results of the unsaturated analysis is examined in a parametric study. The analyses results demonstrate that for unsaturated soils the factor of safety may increase with time, in contrast to what is commonly accepted to be the case in fully saturated soils. Furthermore, it is not possible to know in advance which one of the two types of analysis performed (i.e. the saturated or the unsaturated) will produce conservative results. It is, therefore, advisable to perform unsaturated analysis in geotechnical practice, when dealing with excavations in such soils. [Copyright &y& Elsevier]

Abstract: Constitutive models for unsaturated soils are most commonly based on conventional critical state type models for saturated soils. The ellipse of the modified Cam-clay model is usually adopted for the formulation of the yield and the plastic potential surfaces in the mean stress – deviatoric stress plane. Despite the wide use of this ellipse it has long been shown that the peak deviatoric stress observed for highly overconsolidated states may be largely overestimated. For fully saturated conditions, a planar surface, termed the Hvorslev surface, has been employed in the literature in order to overcome this shortcoming on the dry side of the critical state. In addition, by employing a no-tension cut-off, stress states are restricted within the compressive region. In unsaturated conditions and due to the effect of suction the soil may be subjected to tensile total stresses which should not, however, exceed a model dependent limit. Nonetheless, adopting a planar surface and in the absence of any provision, illegal tensile stresses may be obtainable. For this reason, a new expression for the Hvorslev surface, which restricts the stress state within acceptable boundaries, is proposed, based on a small number of model parameters. The new surface is curved rather than planar and can also be used in combination with critical state type constitutive models developed for saturated conditions. The new surface is calibrated based on fully saturated undrained triaxial compression and extension tests and on unsaturated drained triaxial tests. The superiority of the new surface in comparison with commonly adopted shapes for the yield and the plastic potential functions – such as the ellipse and the bullet shape of the modified and the original Cam clay models – is demonstrated through numerical analyses of the unsaturated drained triaxial laboratory tests. [Copyright &y& Elsevier]

To enable the use of beam elements in the modelling of coupled thermo-hydro-mechanical (THM) geotechnical problems, a fully coupled and robust THM formulation is required. This paper presents such a formulation which allows both fluid flow and heat transfer along a 2D curved beam, while ensuring compatibility with coupled THM solid elements commonly used to discretise soils. Verification exercises and application with the proposed coupled beam element are carried out to demonstrate its satisfactory behaviour. The results of these analyses are compared against closed form solutions, solutions obtained using solid elements, and field measurements, showing an excellent agreement. [ABSTRACT FROM AUTHOR]

Interface elements are frequently employed in finite element (FE) analyses to represent soil-structure interfaces or rock joints. The modelling of coupled thermo-hydro-mechanical (THM) problems in geotechnical engineering requires equally a coupled and robust THM formulation for interface elements. This paper presents such a formulation which is capable of reproducing the coupled THM behaviour of discontinuities and soil-structure boundaries, and is compatible with other types of finite elements used to discretise the soil and structural domains (e.g. solid and shell elements). The coupled THM three-dimensional (3D) zero-thickness interface element is implemented into the bespoke FE code employed in this research and its features are verified using a number of numerical exercises. To demonstrate their performance, the proposed interface elements are employed in the simulation of the coupled THM behaviour of a fissured triaxial sample subjected to a thermal load and the influence of the presence of fissures on soil behaviour is presented. [ABSTRACT FROM AUTHOR]