International Journal of Geomechanics

A significant number of petroleum reservoirs and almost all geothermal reservoirs are characterized by high in situ stress and fractures, and fractures act as major flow paths for fluids. An integrated tensor model is proposed to solve three tasks: characterization of a heterogeneous fracture network, simulation of fluid flow through a complex system for estimation of the grid-based permeability tensor, and unsteady-state fluid flow simulation for estimation of production and pressure losses. Deformation of the matrix and fractures are solved separately and used to compute their dynamic porosity and permeability. Finite-element methods and boundary element methods are used for numerical modeling. The results of this study show that the proposed model can overcome problems requiring excessive computational resources, flow interactions between the matrix and fracture, and the effect of matrix deformation on fluid flow. Results also show that the integrated tensor model serves as an efficient tool for predicting the effect of stress on fracture deformation and consequent productivity and/or injectivity of naturally fractured reservoirs.

Field vibration tests were carried out at a proposed site for the vibration testing room, and 2D numerical analysis using finite difference tool FLAC 5.0 was performed to suggest effective vibration isolation systems. In the analysis, the numerical model is first calibrated with respect to material properties, damping value, and boundary conditions to obtain the output comparable to the field test results. The calibrated model was further used to perform a parametric study by (1) providing vibrating input motions from vibrating machines to be operated; (2) using two depths of cutoff trench; and (3) providing gravel bed, gravel bed with rubber pad, and gravel bed with rubber pad and cutoff trench to study the isolation effects. Comparing the results from the parametric studies with the human perception level of vibration, a decision on the isolation system was determined.

This paper presents a state-dependent constitutive model for sand formulated within the critical-state framework and its implementation into a numerical analysis (FLAC3D) program. The implemented model was verified by using drained triaxial results on sands. The proposed model is shown to capture the stress path dependent behavior of sand over a wide range of densities and confining pressures well based on a unique set of parameters. Numerical simulations of the behavior of a micropile under vertical loading shows that the side and tip resistance, and thus the total resistance of the pile, are functions of the “in situ state” of soil as defined by the state parameter $\psi =e-e_c$ in which $e$ is the void ratio and $e_c$ the void ratio at the critical state.

Solutions for the ultimate bearing capacity of footings on purely cohesive slopes are obtained by applying finite element upper and lower bound methods. In a footing-on-slope system, the ultimate bearing capacity of the footing may be governed by either foundation failure or global slope failure. The combination of these two factors makes the problem difficult to solve using traditional methods. The importance of a dimensionless strength ratio in determining the footing capacity is broadly discussed, and design charts are presented for a wide range of parameters. In addition, the effect of footing roughness and surface surcharge are briefly quantified.

This paper deals with the dynamic steady-state force-displacement relationships (impedance) for rigid foundations of arbitrary shape resting on a semi-infinite medium consisting of homogeneous, isotropic, linear elastic materials using the boundary element method (BEM). An elaborate numerical technique is used to calculate Green’s function, yielding a highly accurate numeric approximation in an explicit form. By using a BEM formulation for triangular elements (linear and infinite shape function) compliance of rigid massless surface foundations was obtained. The use of triangular elements causes stress continuity between elements to be satisfied. Meanwhile, the tendency for foundation edge stresses toward infinity can be modeled by using these triangular infinite stress elements. As a result of this, computations are more accurate but less time-consuming. This procedure is used to evaluate the vertical rocking compliance functions for rigid rectangular, circular, and square foundations with internal holes. Finally, comparisons are made between the results obtained from the proposed approach and those from other methods.

The effects of boundary conditions on the stability of a river levee built on a low-permeability soil layer overlying a coarse-grained deposit were studied by using the finite-element method (FEM). The FEM analyses could predict stable or unstable levee conditions depending on the assumed distance between the levee and the external boundary of the mesh where the water table was assumed undisturbed. Possible causes of this notable drawback are discussed. The calibration of the numerical seepage model, through a back analysis of piezometer measurements, that could limit the observed boundary effects is suggested.

This study describes a technique for reducing the computational effort for performing linear optimization while solving any geotechnical stability problem with the use of the lower bound finite-element limit analysis. In the proposed method, a lower order polygon is initially used to model the Mohr-Coulomb yield function; the order of the polygon refers to its total number of sides. The initial solution is used to identify the governing side of the yield polygon that lies nearest to the point defining the existing stress state. Subsequently, this governing side of the linearized yield polygon is replaced with a number of the relevant sides of the higher order polygon. Because all the sides of the higher order polygon for imposing the linearized yield constraints do not enter the formulation, the associated computational effort becomes much smaller. With the proposed algorithm, the collapse loads were determined for smooth and rough strip footing, and the computational results were found to be quite convincing.