Computer Applications In Geotechnical Engineering

Consolidation of accreting soft soils is a topic of significant interest for engineers dealing with mine tailings deposition, hydraulic fills, dredging deposits, and wetland construction. In these cases, the classical Terzaghi's theory often fails to produce satisfactory solutions due to variations in compressibility and permeability under large deformations. The finite-strain theory introduced by Gibson allows for the non-linearity of material properties and provides an accurate description of soft soil deposits undergoing large displacements. Numerical accretion models based on Gibson's theory need to account for both the non-linearity of the governing equation and the continuous domain change. This paper investigates the numerical performance of several one-dimensional finite-difference schemes with different time-stepping algorithms and mesh discretization procedures. An optimal mass-conservative scheme is selected and implemented into a numerical model. The presented field example demonstrates that the employed mapping technique ensures both mass and water conservation, essential for water balance and storage capacity prediction in slurry disposal projects.

This paper deals with the application of Artificial Neural Networks (ANN) to estimate maximum steady state kinematic pile bending moments in floating piles at the interface of two soil layers of vastly different shear modului. The data generated for training and testing the ANN model is based on Beam on Dynamic Winkler Formulation. The inputs to ANN model are the ratio of the elastic modulus of pile to that of the soil, the length to diameter ratio of pile, shear wave velocity ratio of the two soil layers, and the ratio of the thickness of the top soil to the length of the pile. The output of the ANN model is the normalized maximum steady state pile bending moment at interface of the two layers. Feed forward Levenberg-Marquardt backpropagation algorithm is used to train the ANN model and its performance model is evaluated using test data that forms 50% of the total data. The results show that ANN can be used to estimate the maximum steady state pile bending moment and the corresponding maximum pile moment in time domain using reduction factors.

This paper presents results obtained from finite element simulations of dense and loose samples of sand loaded under globally undrained plane strain conditions. The effective behavior of the sands is captured using a newly developed elastoplastic model for granular materials based on critical state theory (CST). The novelty here is the ability of the constitutive model to account for inhomogeneities in the density field at the meso-scale while still maintaining a macroscopic approach. The granular medium is assumed to be fully saturated and the interplay between the grains and fluid is fully coupled using mixture theory. Bifurcation analyses to study the different failure mechanisms for dense and loose saturated sands are performed on samples considered to be macroscopically `homogeneous'. The finite element simulations clearly show that the relative density and small defects present in the samples have a profound impact on the mechanical behavior of saturated granular media.

Because of their rapid construction and cost-effectiveness, mechanically stabilized earth (MSE) walls are one of the most preferred wall types in many civil engineering projects. As part of an attempt to reduce wall deformations, researchers have been studying the effect of properties of wall components, such as backfill soil stiffness and facing unit rigidity. The effect of concrete key size on wall deformations was investigated in this study using numerical analysis. Currently the concrete key is used for an alignment and leveling of the facing units. The effect of concrete key size for varying reinforcement lengths and foundation soil friction angles was studied. Numerical analyses were performed using Plaxis, commercially available finite element software. The results indicated that longer key lengths result in decreased wall deformations. The effect of the longer key lengths in reducing wall deformations was more significant for the walls with shorter reinforcement lengths and for the walls placed over weaker soils.

The Taum Sauk Dam Upper Reservoir located in southeast Missouri failed on December 14, 2005. Two days after the catastrophic failure event an aerial survey were conducted to collect Light Detection and Ranging (LiDAR) data. The LiDAR data were interpreted and evaluated to a sub-meter resolution. The data were used to quantify the extent of the damage produced by the water released into a Missouri State Park. Damage consisted of washing out trees, soil, and the reservoir embankment, and scouring the ground to bedrock in places. Additionally, the data also were used in the evaluation of the reservoir damage, which consisted of a 207.3m (680ft) wide breach and scoured slopes. This paper focuses on the multiple uses of LiDAR data to assess damage to natural resources, and hydraulic and geotechnical evaluation of the failure.

This paper reflects on the failure of the New Orleans levee system during hurricane Katrina. Experience of upgrading the levee system in the southern province of Guangdong, China is introduced. Based on this experience, it is recommended that sluices to be constructed on the estuaries of major canals to reduce the length of high standard levee. Superdikes can be constructed for the external levee system by expanding the existing ones. The concept of green hydraulics can be embraced in the design of superdikes, which comprehensively considers the functional, environmental, and aesthetic factors. Selected vegetations can be used to provide wave energy barrier as well as aesthetic attractions. Strict wet land preservation plan can be implemented.

A major concern involving any steep embankment is its stability. This is the case whether it is the side wall of an open pit mine, a steep slope next to a highway, or a man made structure like a dam or dike used to contain tailings or generate hydroelectric power. In general, for each case there is a slow movement of the formation before a failure actually occurs. With the continuous monitoring of a phenomenon's creep, particularly its acceleration, critical displacements can be detected before the object reaches its failure point. Among new technological developments, robotic total stations (RTSs) give one possible cost effective solution to creating a near real time monitoring system. Repeated surveys at predetermined time intervals can automatically generate warnings if slope movements exceed a preset tolerance level. In addition, the system can be augmented with GPS or geotechnical sensors to make it a more reliable system. The strengths of each sensor type can be exploited to optimize the cost and efficiency of the monitoring system to ensure pre-failure detection. The latest monitoring techniques apply the principles of Synthetic Aperture Radar (SAR), using both satellite and portable ground based systems, to determine complete surface deformations rather than the few discrete points that are obtained with the current systems. A methodology has been developed at the Canadian Centre for Geodetic Engineering (CCGE), in cooperation with ACA Ltd, for optimizing the design of monitoring systems by combining geometrical and deterministic modeling methods to ensure proper sensors are installed at proper locations and utilized correctly. The methodology has been realized in the creation of a software package that has been successfully implemented at large earthen dams and open pit mining operations around the world.

During earthquakes retaining walls move due to the increased earth pressure exerted on them. A simple method to reduce this pressure is to use a leaning back slope for such walls, but doing this for the whole rear face is not only uneconomical due to the large volume of materials used, but also less efficient since the earth pressure is considerably smaller at upper parts of the wall. So a broken back wall can be an efficient and economical structure to mitigate the pressures. In order to study this behavior of broken back walls, a series of shaking table tests were carried out on such retaining walls made of concrete blocks. The results showed that after shaking the earth pressure increases in the upper portions of the wall and decreases by the backward leaning slope at lower elevations. Depending on the shape of back-face of the wall the total thrust and overturning moment would be increased or decreased during and after an earthquake. In general the broken back shape of the wall tends to raise the point of application of the total thrust exerted on the wall.

The evaluation of the safety of natural or man-made slopes involves uncertainties. With a deterministic safety factor, the geotechnical engineer tries to deal with the uncertainties by choosing reasonably conservative parameters for stability evaluation. The deterministic approach fails to deal consistently with uncertainties. As a complement to the deterministic approach, a probabilistic approach that allows for hazard and risk assessment of slopes is recommended. The paper proposes a framework integrating existing methods to exploit the benefits of the two approaches. Societal aspects, including allowable and tolerable risk, are briefly considered. The need for input on the social dimension of landslides, and the role of risk assessment in easing communication among the different experts are also discussed.

The Vaiont Slide is frequently cited as illustrating one of the hazards caused by dam construction even when the dam itself is shown to be safe. Here, we briefly review the chronology of events, monitoring performed prior to failure, likely causes of the failure, and suggest instrumentation that could be used in similar situations to potentially avoid failure. Specifically, time domain reflectometry techniques could be used in similar situations to provide substantive water pressure measurements and reliable movement records. Clearly, such consequential projects need thorough site investigations and carefully planned and maintained monitoring systems to avoid future disasters.

This paper presents a unified thermo-mechanical experimental study on a remoulded unsaturated sandy silt and brings a contribution to the understanding of the fundamental mechanics of unsaturated soils in non-isothermal conditions. The experimental program was carried out at four temperatures and four suction levels using two thermo-hydro-mechanical (THM) cells, one isotropic and the other oedometric. The effect of suction and temperature on the compressibility and on the apparent preconsolidation pressure of the soil is addressed. Finally, an analytical expression of the evolution of the apparent preconsolidation pressure with respect to temperature and suction is proposed.

The production of nuclear energy is facing at the present time to the issue of the management of radioactive waste. A great deal of attention is focused on the feasibility of deep geological underground disposal. Consequently, deep knowledge of thermal, hydraulic and mechanical processes is necessary. Indeed, excavation of the underground works involves development of cracks and fractures in the rock. That will modify the hydraulic properties of the soil creating preferred zones for water flows. It appears essential to evaluate accurately the area of the excavation damaged zones (EDZ) and their evolution in order to forecast the migration pathway for radio nucleides in porous media. In this goal and to describe the coupling and the transfer phenomena in these geological repositories, thermo-hydro-mechanical models are developed. In a first part we will present the disposal concept proposed at the moment and we will describe phenomena to take into account in the simulations : desaturation and resaturation of the geological medium, thermal evolution owing by the packages, gas pressure induced by steel containers corrosion. Then we will talk about numerical difficulties linked to thermo-hydraulic unsaturated problem and brittle mechanical laws.

Compacted swelling clays are often envisaged as the main component of engineered barriers for radioactive waste disposal. These barriers are subjected to thermal loading due to the heat emitted by the waste and to hydration from water coming from the adjacent rock. As a consequence of these thermo-hydraulic phenomena, mechanical and chemical changes arise that, in turn, may affect all other aspects of behaviour. A correct understanding and prediction of these barriers require, therefore, the performance of fully coupled thermo-hydro-mechanical and chemical (THMC) numerical analyses. In this paper a fully coupled THMC formulation for unsaturated expansive clays is described and used to analyze the THMC behaviour of the bentonite selected as raw material to construct engineered barriers according to the Spanish concept for high level radioactive waste isolation. The aim of this modelling is to check the capabilities and limitations of the formulation to explain and reproduce the observed THMC behaviour of the bentonite.

One of the main issues in soil modelling is the dynamic coupling between the porous solid skeleton and the water-gas mixture contained in the pores. This article describes some conceptual aspects of a dynamically coupled system of this kind. This article introduces a simple solid-water-gas interaction model that takes into account the variation of compressibility of pore water-gas mixtures and the compressibility of the solid mass. The more complex relationships have been schematized in accordance with a system dynamic approach. One of the major factors is how the quantity of free gas influences the compressibility of the water-gas mixture. It is therefore impossible to ignore the exchange process between free and dissolved gas. The current result seems to be satisfactory. On the one hand the result warns against applying excessive back pressures and on the other hand addresses the relationship between B_w and S_r, like Vaughan.

Viscosity is in fact a relaxation of the shear stresses over time, which results for constant stress states in a corresponding elastic strain relaxation. In other words, viscosity is a sort of slow plasticity before the Coulomb criterion is reached. Creep settlement can be regarded as one of the viscous soil phenomena, but also the slow movement (a few mm or cm every year) of many natural slopes in mountainous area's. Therefore a viscous constitutive model is derived from the standard principles that the reduction rate of the relative shear stress (the viscosity) is a function of this relative shear stress and that the reduction rate approaches zero while the stress-state approaches the K₀-state. As an example a User-Defined (Soft) Soil Model is developed in order to model the continuous moving slopes. Although the model works fine for viscous behavior and creep problems, it seems as if the slope movements do not depend on viscosity but on dilatancy.

Internal erosion processes in geomaterials like fluid-saturated sand-gravel layers play a significant role in geotechnical and petroleum engineering. We suggest a general modeling framework within such a solid-liquid phase transition phenomenon can be described. The phenomenological continuum-based model is formulated within the thermodynamically-consistent Theory of Porous Media. The mixture consists of four constituents. Furthermore, mass exchange between a solid and a liquid phase is taken into account. The hydromechanically-driven erosion process is mainly described by the constitutive equation of mass production. The degradation of the stiffness of the solid skeleton is taken into account by a damage-like behavior.

Mechanisms of free shrinkage strains of desiccating deformable porous media are studied. The roles of surface evaporation rate, surface tension, and viscosity of pore fluid, soil compressibility and permeability are investigated. For drying tests on two geomaterials with three different pore fluids the evolution of shrinkage strain and fluid content is reported. Most of the strains occur in the saturated phase of drying prior to cracking. Simulations of this phase include evaporative fluxes at the external surfaces, a consequent water transport across the sample toward the surface producing pore pressure and stress gradients, local water content change, and deformation. Biot theory is used.

A model for frost-susceptible soils, effective in simulating frost heave, is presented. This model is based on a porosity rate function that describes the global increase in the soil volume due to growth of ice. Individual ice lenses, however, are not simulated. The model was implemented in the finite element analysis, and the displacement of a retaining wall was simulated due to freezing of the frost-susceptible backfill. Because heaving requires the freezing front to propagate into the frost susceptible soil, it is demonstrated that thermal insulation along the wall inhibits the frost heave to some extent, and prevents excessive tilting of the structure.

This article presents an experimental study of the water retention curve involving the measurement of void ratio on a drying path. Tests were conducted to determine the variations of water content w and void ratio e induced by variations of the suction s. This study concerns the whole hydric domain of the soil from saturation to dry state. To determine the water retention curve, we used a pressure cell for low suctions and a vapor equilibrium technique (salt solutions) for high suctions. This water retention curve is based on the constitutive relation between water content and suction, for a given void ratio. From saturation, this curve was investigated for different initial states due to different conditions of soil compaction. The results lead to the plotting of a characteristic surface w=w(s,e) which expresses water content as a function of suction and void ratio. It can be considered as a generalization of the classical retention curve. Moreover, this curve can be represented using the degree of saturation as function of suction and void ratio.

In the context of using engineered clay barrier in deep nuclear waste disposal, an experimental work was carried out to observe the behavior of compacted swelling clay under thermo-hydro-mechanical (THM) loading. Firstly, the measurement of thermal conductivity has allowed the elaboration of a theoretical model to estimate the thermal conductivity of unsaturated soils. After an evaluation study on the proposed model and three existing models found in the literature, it was concluded that the thermal conductivity of unsaturated soil could be correlated with only one parameter: the air volume fraction. Secondly, the water retention curve, that influences the water transfer in unsaturated soil, was determined at different temperatures. The results showed a decrease of water retention capacity due to heating. A simple model based on the interfacial tension air-water was formulated to simulate this effect. Finally, the soil volume change under complex THM loading was studied. Thermo-mechanical phenomena, which are often observed on saturated clay, were equally observed on the compacted unsaturated clay.