International Journal of Geomechanics

The geotechnical design of drainage or dewatering systems requires a free surface seepage analysis to evaluate the water flux that corresponds to the sought lowering of the water table. It is observed in practice that sometimes the calculated flux is not sufficient to reach the predicted result, even when the hydraulic conductivity of the soil mass has been properly determined. This prevents the initiation of the in situ works (e.g., excavation) with a relevant economical loss. The possible causes of this problem are investigated through a series of two- and three-dimensional unconfined seepage analyses. They show that the numerical results strongly depend on the characteristics of the numerical model and that, consequently, the computed fluxes and lowering of the water table do not necessarily correspond to the field conditions. To overcome this drawback a back-analysis approach is suggested for calibrating the numerical model adopted in design. An application of this technique to an actual field problem is finally discussed.

The use of multiple-mode dispersion data in surface wave inversion to derive shear-wave velocity $\left(v_s\right)$ profiles has increased in the past decade as the inclusion of higher mode data can improve the accuracy of the inversion results. However, the error associated with nonuniqueness in the multiple-mode inversion has not been clarified and quantified. This research focuses on the attempt to improve the accuracy of multiple-mode surface wave inversion result by optimizing the use of multiple-mode dispersion data to reduce the error associated with the nonuniqueness in inversion. In this research, an alternative approach was used where inversion of surface wave dispersion data was performed using three distinct modes. Four different $v_s$ profiles, representing regular and irregular cases, were used, and multiple-mode dispersion data were synthesized from these profiles using the dynamic stiffness matrix method as the theoretical model. The dispersion data were then inverted using the Levenberg–Marquardt method. The results demonstrated that, as expected, inclusion of higher modes did not improve the accuracy of the inversion results for the regular profiles. However, inclusion of higher modes significantly improved the uniqueness of the inversions for the irregular profiles. The results also demonstrate that regardless of the nature of the profile, the accuracy of the inversion improves when the starting profile more closely matches the true profile. Of all the inversion approaches investigated, the best approach was one where three successive inversions, using one, two, and three modes, respectively, was used, where the inverted profile from one inversion was used as a starting model for a subsequent inversion that used one additional mode.

Although the availability of discrete-element method (DEM) codes has improved, the need still exists to solve simple verification problems to obtain an understanding of these codes. Different DEM codes may have subtle differences in the manner in which the method is implemented, and the significance of these differences may be problem dependent. This paper investigates a series of simple, one and two-particle contact problems. These problems, which employ various types of damping, are shown to be equivalent to classical one-dimensional vibration problems. The solutions are discussed in the context of the DEM, and results from the DEM are shown to compare very well with the classical solutions. It is demonstrated that results from a well-known commercial two-dimensional code (PFC2D) and the open source three-dimensional code (YADE) yield identical solutions to these problems provided the problem solution process is manipulated properly. A discussion of the differences in how gravity and damping are implemented may be of interest to users of PFC2D.

Closed-form and semianalytical solutions for induced poroelastic stresses and strains are extremely useful for the design of subsurface fluid storage in caverns because of their relative ease of implementation and their suitability for parameter sensitivity analyses. This paper describes the use of Eshelby’s inhomogeneity theory to derive equations that can be used to predict the induced stresses and strains for reservoirs that are elliptical in cross section, under plane strain conditions. Sensitivity analyses demonstrate that the induced stresses are relatively insensitive to the Poisson’s ratio of the surrounding rock, but they are strongly affected by the Poisson’s ratio of the reservoir, and the ratio of shear modulus of the reservoir to that of the surrounding rock. Results are presented in terms of dimensionless parameters, which facilitate their application to a broad range of reservoir dimensions and pressure-change magnitudes. These equations can also be used to predict the induced stresses around a cavity, which represents the special case of an inhomogeneity with a shear modulus of zero.

This contribution deals with the use of the computationally efficient spectral element technique as a means of analyzing the dynamic behavior of axisymmetric multilayer systems consisting of homogenous and nonhomogeneous material layers and subjected to a stationary but transient force. Focus is placed on the mathematical formulation and numerical verification of a nonhomogeneous spectral element for soil material. The general solution of the wave equations and the boundary value problem are achieved by triple summation over fast Fourier, Fourier–Bessel, and Frobenius series. Practical utilization of the model is presented via a numerical example, which simulates a pavement structure subjected to the dynamic action of a falling weight deflectometer test as typically utilized for pavement structural evaluation.

An analytical solution for the evaluation of scattering of waves by a circular cavity in infinite isotropic elastic porous media is presented. Two groups of complex functions for solid skeleton and pore fluid in a two-dimensional complex plane are introduced in order to solve the Biot equations. Stress, displacement, and pore pressure fields induced by incident and scattered waves in the medium and especially in the vicinity of the cavity are evaluated in this complex plane. The validation of the proposed solution is shown by various numerical examples. A parametric study including the effects of fluid compressibility changes, shear modulus, and permeability variations, several wave numbers, and wave types (fast, slow, and shear waves) is performed.

Recent laboratory investigations indicate that the stress–strain–strength responses of granular soils are appreciably affected by the fabric orientation of the soil relative to the frame of principal stresses. Especially, a sand specimen exhibiting a dilative response during triaxial compression may show a contractive response during triaxial extension under otherwise identical conditions. This observation is of practical importance for applications concerning essentially undrained loading conditions, because the effective mean normal stress at failure, and consequently, the shear strength, associated with an undrained contractive path are considerably lower than those following a dilative path. This raises a question about the impact of soil anisotropy on seismic performance of retaining structures subjected to active and passive earth pressures, because the directions of principal stresses in retained soils for the two cases are very different. This note presents a set of fully coupled finite-element analyses incorporating an anisotropic sand model. The analyses reveal that the impact of fabric anisotropy could be significant when the retaining structure is under passive earth pressure conditions.

A simple practical equivalent continuum numerical model for simulating the behavior of jointed rock mass has been extended to three-dimensional using FLAC3D. This model estimates the properties of jointed rock mass from the properties of intact rock and a joint factor $\left(J_f\right)$ , which is the integration of the properties of joints to take care of the effects of frequency, orientation, and strength of joint. A new FISH function has been written in FLAC3D specifically for modeling jointed rocks using the Duncan and Chang hyperbolic model. This model has been validated first with simple element tests at different confining pressures for different rocks with different joint configurations. Explicit modeling of the joints has also been done in element tests and results obtained compare well with the results of equivalent continuum model and also with experimental results. Further, this has been applied for a case study of a large underground power house cavern in the Himalayas. The analysis has been done under various stages of excavation, assigning a null model available in FLAC3D for simulating the excavation.

In order to study pore water pressure response and liquefaction characteristics of sand, which has previously experienced liquefaction, two series of cyclic triaxial tests were run on medium dense sand specimens. In the first test series the influence of the soil saturation under undrained cyclic loading has been studied. It summarizes results of cyclic triaxial tests performed on Hostun-RF sand at various values of the Skempton’s pore-pressure coefficient. Analysis of experimental results gives valuable insights on the effect of soil saturation on sand response to undrained cyclic paths. In the second series of tests, the preloading influence on the resistance to the sands liquefaction has been realized on samples at various histories of loading. It was found that a large preloading induces a reduction of the resistance of sands to liquefaction.