1 September 2005 in International Journal of Geomechanics
Finite element (FE) analysis has become an important tool for predicting building response to tunnel-induced ground movement. Because tunnel construction is a three-dimensional (3D) process, the trend is to apply 3D FE analysis to tunnel-soil-building interaction problems instead of applying the plane-strain models that are commonly used in engineering practice. Since 3D FE analyses require large amounts of computational resources, the geometric dimensions of the 3D models are often kept to a minimum to reduce calculation time. There is, however, a lack of published information concerning appropriate mesh dimensions. This paper investigates the influence of the geometry and the dimension of a 3D FE model on tunnel-induced surface settlement predictions. The paper shows how the vertical boundaries can influence the results. It demonstrates that reasonable results can be obtained by increasing the length of incremental tunnel excavation and by scaling back the settlement values to give a required tunnel volume loss. This study therefore not only highlights the limitations of 3D modeling but also shows its potential for engineering practice.
1 July 2017 in International Journal of Geomechanics
When modeling the development of slip surfaces in strain-softening soils, the FEM can suffer from mesh dependency and a lack of convergence of the analysis. The calculation of strains using a nonlocal method avoids the independent softening of a solitary point that can cause these symptoms. Nonlocal methods calculate the strain at a point with reference to the strains at calculation points surrounding it. In this study, a biaxial compression analysis is used to compare the original and two modified nonlocal methods with local strain-softening analysis. Three nonlocal parameters are introduced, and their influence on the nonlocal strain calculation is investigated. Meshes with different discretizations permit a comparison of the mesh dependency of these methods and a sensitivity analysis of nonlocal parameters. All the nonlocal methods are found to have significantly less mesh dependency than the local method. Both modified nonlocal methods exhibit less mesh dependency than the original approach, but one of the two modified methods, the overnonlocal method, is shown to be unstable because the mesh is refined for certain values of the nonlocal parameters.
1 September 2006 in International Journal of Geomechanics
Two new constitutive models formulated within the framework of kinematic hardening plasticity are presented and their implementation into a finite-element program is described. The models are extensions of two existing constitutive models for reconstituted clays and introduce a number of kinematic surfaces within the modified CAM clay bounding surface. The new key feature of the models is a hardening modulus which results in a smooth variation of stiffness with strain, from the high elastic value, within the first kinematic surface, to the value on the bounding surface. Other features include a mathematical formulation of the models in general stress space to facilitate their implementation into a finite-element program, a variety of shapes of the yield and plastic potential surfaces in the deviatoric plane, and the novel concept of changing the active yield surface, which is necessary for the consistent formulation and implementation of the models into a finite-element code. The models are shown to have ...
1 August 2021 in Computers and Geotechnics
2 March 2023 in Computers and Geotechnics
1 June 2020 in Géotechnique
The complexity of advanced constitutive models often dictates that their capabilities are only demonstrated in the context of model testing under controlled conditions. In the case of earthquake engineering and liquefaction in particular, this restriction is magnified by the difficulties in measuring field behaviour under seismic loading. In this paper, the well-documented case of the Canterbury earthquake sequence in New Zealand, for which extensive field and laboratory data are available, is utilised to demonstrate the accuracy of a bounding surface plasticity model in fully coupled finite-element analyses. A strong motion station with manifestation of liquefaction and the second highest peak vertical ground acceleration during the 22 February 2011 Mw6·2 seismic event is modelled. An empirical assessment predicted no liquefaction for this station, making this an interesting case for rigorous numerical modelling. The calibration of the model aims at capturing both the laboratory tests and the field measurements in a consistent manner. The characterisation of the ground conditions is presented, while, to specify the bedrock motion, the records of two stations without liquefaction are de-convolved and scaled to account for wave attenuation with distance. The numerical predictions are compared to both the horizontal and vertical acceleration records and other field observations, showing a remarkable agreement, also demonstrating that the high vertical accelerations can be attributed to compressional resonance. The results provide further insights into the underperformance of the simplified procedure.