International Journal of Geomechanics

Separation and sliding between soil and pile have a significant effect on the complex behavior of pile groups under dynamic condition. In addition to separation between soil and pile, the behavior of soil medium surrounding the piles is nonlinear during strong excitations. Therefore, soil is modeled using a nonlinear work-hardening plastic cap model. Because of shaking, there is an increase in pore water pressure. To simulate it, the pore-pressure generation capability is added in modeling the behavior of the soil medium. A full three-dimensional finite-element soil-pile model has been developed with approximate radiation condition at infinity simulated by frequency dependent Kelvin elements. The model and the process of computation have been verified with the established literature. The effects of separation on the dynamic stiffness and seismic response of single pile and pile groups have been examined considering work-hardening soil-plasticity without and with pore-pressure generation capability for a wide range of frequencies of excitation. Real-time earthquake motions are also applied and the effects of separation on the responses of pile groups are investigated. Significant effects of separation of the soil medium on the response of the soil-pile system have been observed and approximately quantified.

Many large underground structures may be subjected to varying initial stress states, cyclic water heads, and long-term temperature variations. When the maintenance and extension of the service life of such structures are accounted for, their long-term performance under varying initial stress states, cyclic water heads, and temperature variations becomes very important. The authors were involved with the evaluation of the long-term performance of an underground powerhouse in central Japan constructed for a pumped-storage scheme. The evaluations involved both numerical studies on the long-term effects of varying initial stress states, cyclic water heads, and temperature variations on the performance of the powerhouse considering the degradation of support systems and creep phenomenon and the monitoring of ground deformations, temperatures, and vibrations induced by turbines. The outcome of these studies are presented in this article.

A system of vertical drains with combined vacuum and surcharge preloading is an effective method for promoting radial flow, which accelerates soil consolidation. This study presents the analytical solutions of vertical drains with vacuum preloading for membrane and membraneless systems under time-dependent surcharge preloading. Both vertical and horizontal drainage were considered in this analysis because they reflect realistic in situ conditions. According to the field and laboratory observations, the vacuum in the membraneless system was assumed to be decreasing along the drain; in the membrane system, it was maintained at a constant level. This model was verified by using the measured settlements and excess pore pressures obtained from large-scale laboratory testing and case histories in Australia, Korea, and China. The analytical solutions improved the accuracy of predicting the dissipation of pore water pressure and the associated settlement. The effect of the permeability of the sand blanket in a membrane system, and the possible loss of vacuum, were also discussed.

Most research has focused on using the subsidence prediction method to calculate final movement at the center line above mining operations and assess the surface structural damage without taking into account the dynamic extraction process. Nevertheless, the occurrence of subsidence caused by underground mining has three-dimensional (3D) time-dependent components of movement in which each of these components has a different effect on the types of structures subject to subsidence. In this paper, a new computational method is proposed to calculate 3D dynamic subsidence during the process of undermining by combining the stochastic theory, the Knothe model, and the Geographic Information System (GIS). A case of 3D dynamic modeling was simulated for a rapid undermining scenario to demonstrate the effect of different advancing mining faces on the development of traveling strain. In addition, an application of the GIS-based method to actual field conditions in coal-mining subsidence in China is presented in this paper. Because GIS functions are implemented greatly in the prediction model, the 3D dynamic movement during undermining can be simulated effectively and efficiently.

Knowledge of seismic active earth pressure behind a rigid retaining wall and its point of application are very important to the design of retaining walls in earthquake-prone regions. This paper presents a detailed study on the seismic active earth pressure and its point of application behind a nonvertical, rigid retaining wall supporting inclined, cohesionless backfill, using pseudo-dynamic analysis that is more realistic to representing the time and phase difference within the backfill. A planar failure surface is considered in the analysis. The effects of soil and wall friction angle, wall and backfill surface inclination, and horizontal and vertical earthquake acceleration on the active earth pressure have been explored. Results are presented in both tabular and graphical nondimensional form, including comparison with other available methods to highlight the realistic nonlinearity of seismic active earth pressure distribution.

The numerical analysis of an object penetrating deep into the seabed is a fundamentally challenging problem. This paper presents the application of a novel Eulerian-based finite-element technique to simulate the continuous penetration of a jack-up spudcan foundation into seabed of different soil profiles. The finite-element mesh is kept stationary throughout the analysis and the material is allowed to move independent of the element nodal points. Consequently, termination of computing execution from severe mesh distortion does not occur despite the material undergoing large deformation. The first part of the paper elucidates the mesh density requirement, the effect of penetration rates, and factors influencing the simulation time. The applicability of the Eulerian finite-element model is then validated through comparison with published experimental data for different soil profiles. In general, the Eulerian finite-element model is able to replicate the experimental observations well. With the Eulerian approach, classical wished-in-place approximation in spudcan penetration analysis is no longer necessary and a more accurate continuous penetration simulation can be routinely performed with minimal user-intervention.

In the design of large and tall storage containers/structures such as silos, hoppers, and mine stopes, it is often required to determine the stresses within the container, especially at the bottom where stresses are the highest. Due to arching, where a substantial fraction of the self-weight of the granular material is carried by the wall, the vertical stress at the bottom of the container is significantly less than what is given by the product of the height and unit weight. Few analytical expressions published in the literature can be used to determine the vertical stresses taking into account the arching effect and on the basis of equilibrium considerations. The objective of this paper is to propose a new analytical method for determining the vertical stresses in a long container, assuming plane strain conditions. The method is extended to containers with rectangular and circular cross sections and is used to accommodate a surcharge at the top of the granular material. The values of vertical normal stresses computed for long-strip square, and circular cross sections are in very good agreement with those computed from Marston’s theory, which was recently validated against numerical and laboratory models for providing satisfactory estimates of average vertical stresses within mine stopes. The results from the proposed model also compare well with elastoplastic numerical model results, provided $K$ and $\delta$ are assumed to be as follows: (a) $K=K_0$ and $\delta =2/3\varphi$ or (b) $K=K_a$ and $\delta =\varphi$. More than the expressions, the method itself would pave the way to future applications in similar and generalized problems related to storage structures with sloping walls.