The paper tackles the problem of interaction between unstable slopes and pipelines. Thanks to appropriate simplifying assumptions, an engineering model, conceived for taking into consideration the most general geometrical conditions, is introduced. Generalised kinematic and static variables governing the problem are defined and an interaction domain is determined. Within it, the system response is assumed to be linear and reversible. For loading directed outwards, an elastoplastic behaviour is assumed and appropriate flow and hardening rules are proposed. It is shown that in this way it is possible to reproduce the well documented coupling between the different components of the acting loads and the irreversibility of the mechanical response of the system.

View full abstract

Multiple series of laboratory pressure chamber tests were conducted to examine the penetration resistance N_SW of Swedish weight sounding tests in sandy soils. The effects of overburden stress, relative density and grain composition were examined. Similar to the empirical formula for SPT N-value adopted previously by Cubrinovski and Ishihara (1999), the data of Swedish sounding tests were arranged and the empirical formula for the N_SW-value is proposed. This paper also addresses whether the void ratio range is a good parameter in characterizing the effects of grain composition. By combining the two empirical formulae for the N_SW-value and SPT N-value, an empirical relation between the N_SW-value and SPT N-value was also derived, which may be used to convert the penetration resistance from one to another. The empirical formula thus derived implies that the relation between the two kinds of penetration resistances is dependent upon the grain composition of soils. The proposed formulae are found to be fairly comparable with the field data with a reasonable degree of coincidence.

View full abstract

In order to establish a precise method for predicting the residual settlement, including the secondary consolidation, of clayey ground improved by vertical drains, a new consolidation test apparatus was developed. In this test apparatus, the consolidation of a hollow cylindrical body, in which the pore water flows radially toward the center, can be simulated by installing a vertical drain at the center of the specimen. The radial distributions of the radial displacement and excess pore water pressure in the specimen during the three-dimensional consolidation process also can be measured in this test apparatus. The radial displacements in the specimen were obtained by measuring the intensity of the magnetic field induced by a small magnet buried in the specimen by magnetometers set in the consolidation ring surrounding the specimen and in the vertical drain installed at the center of the specimen. From the measured results on the radial displacements in the specimen, it was found that during the three-dimensional consolidation process not only the vertical displacement but also the radial displacement occurs inside the specimen. The radial displacement developed toward the center of the specimen in the early stage of consolidation and then returned toward the outside in the later stage of consolidation but did not completely return to the original position at the end of consolidation. The mechanism of the three-dimensional consolidation by the vertical drain, which caused these behaviors, was clarified based on the test results on the radial distributions in the radial displacement and the excess pore water pressure in the specimen.

View full abstract

The response of pile groups against seismically induced lateral soil spreading is explored in centrifuge model tests. The influence of number of piles and pile spacing on the total lateral force on piles is especially examined. The silt saturated with water was used to produce geotechnical models in the laminar shear box. The model piles were installed in 1-layer and 2-layer soil models to produce various patterns of pile groups, with different number of piles and pile spacing. In one test series, the pile groups were installed in the direction perpendicular to lateral spreading, while in another test series, the pile groups were installed in the direction parallel to lateral spreading. The instrumented models were then seismically excited at elevated centrifuge gravity. The group effects of closely spaced piles on the total lateral force induced during lateral spreading are evaluated with respect to the number of piles and pile spacing, with a help of a simple theoretical consideration.

View full abstract

End bearing load transfer characteristics of bored precast piles equipped with expanded fabric bulb in toe region is investigated based on the use of hyperbolic transfer function for shaft resistance as well as for toe resistance of the enlarged bulb portion. Based on the static loading test results, a piling method specific approach to estimate the transfer function parameters from correlation with average TV-values is proposed and its reliability investigated. The reliability investigations include the vertical load resistance aspects as well as the confidence limits for vertical movement. It is confirmed that the shaft resistance of the enlarged fabric bulb portion is much larger compared to the conventional bored precast piles and the increase is a function of the pressure applied in injecting the cement slurry into the fabric sack to form the bulb. On the average, it appears that the increase in bulb shaft resistance is of the order of 7% for every 0.1 N/mm² increase in pressure.

View full abstract

This paper describes a plasticity based constitutive model for the description of cement treated clay. The model uses a bonding stress ratio m and λ'(p', e) for the simulation of cement treated clay. These parameters are used to simulate the increase in the initial stiffness and shear strength of the material and subsequent progressive reduction in stiffness and strength due to breaking of the bonding as a result of shear straining. The bonding stress ratio m takes into account the bonding effect among soil particles and the degradation of bonding with confining pressure and shear strain. The model assumed that the strength and volume change of the cement treated clay are related to the critical state parameter λ'(p', e). It is demonstrated that the model can predict stress and volume change of cement-treated soils over a wide range of confining pressures. The model shows reasonably good match with the observed response and the model parameters can easily be obtained from traditional laboratory tests.

View full abstract

Scale factors for centrifuge modelling have been traditionally defined using dimensional analysis concepts. This is the case, for example, of centrifuge modelling of unsaturated water flow. However, scale factors governing suction, discharge velocity, and time obtained using dimensional analysis have often differed from those obtained from methodologies not based on dimensionless groups. In this paper, a consistent framework is developed for analytic determination of suction profiles for steady-state unsaturated flow under both natural and increased gravitational fields. This framework allows deduction of the scale factors, which emerge from direct comparison of the analytic solutions for model and prototype without the need to use dimensionless groups. For centrifuge conditions leading to an approximately uniform acceleration field, the suction profile in the prototype is found to be the same as that in the model, while the discharge velocity is found to be properly scaled by 1 /N and time by N², where N is the average acceleration ratio between model and prototype. If acceleration field is not uniform, the scale factors should be defined as a function of the centrifuge radius and model length. In addition, evaluation of the effect of different test conditions allows identification of the suction profiles and test setup best suited for hydraulic conductivity determination using centrifuge techniques.

View full abstract

The ultimate bearing capacity of a soft ground, improved by end-bearing soil cement columns, is analyzed using the kinematic approach of the yield design theory. Presented here is an experimental analysis which is based on a three dimensional scaled model of the improved soft ground. The new results given by the kinematic approach and recorded experimental data are compared with others proposed by Broms' method and those established previously by the static approach of yield design theory. Regarding the results given by the Broms' method those predicted by the yield design theory approximate more accurately the recorded ultimate bearing capacity from experiments. Also, the influence of gravity forces on the bearing capacity is investigated.

View full abstract

The shape of the hammer used for the Standard Penetration Test is not specified in most existing standards, yet the literature suggests that it does affect the amount of energy transfer. Three cylindrical hammers and a safety hammer in different geometrical configurations are adopted in laboratory experiments to investigate the effect of hammer shape on energy delivered to a drill rod. In the experiments, true free-fall blows are employed. Energy transfer during each blow is measured using a strain-gauged drill rod and a force transducer mounted atop the drill rod. Energy transfer between the hammer and the drill rod is calculated by the stress-wave force integration of the force-time history measured in each hammer impact. Measurement data are all corrected to an infinite rod length condition. With a total of 80 controlled blows for the four hammers, the measured energy transfer ratios are all nearly 99%. The tests performed in this study are fairly repeatable and the measurement data of the two different devices are correlated very well on the basis of wave mechanics. The measurement test results indicate that the effect of hammer shape on the energy transfer of the standard penetration tests presented herein are negligible.

View full abstract

This paper describes the results from a series of large-scale triaxial tests on dense sand and gravel to investigate the effect of large number of cyclic loading on their deformation. Specimens were prismatic with dimensions of 50 cm in height and 23 cm times 23 cm in cross-section. Deformation was measured locally to avoid the effects of membrane penetration at the side surface of the specimen and of bedding error at the top and bottom ends of the specimen on the measured strains. It is observed that by applying ten thousand cycles of vertical loading at a certain amplitude, the overall stress-strain relationship at a larger stress amplitude changed largely; for example, its shape changed from concave to convex. In addition, after enough pre-straining by cyclic loading, dense granular materials showed almost non-linear elastic behaviour. The test results also suggest the existence of threshold of stress amplitude so that if a large number of cyclic loading with this amplitude is applied, dense granular materials would become stable. This amplitude would be linked with the other factors, such as the density of the specimen, the state of neutral stress and the number of cycles. On the other hand, small strain Young's moduli are affected only to a limited extent by a large number of cyclic loading.

View full abstract

Most conventional slope stability calculations are based on the linear Mohr-Coulomb failure criterion. A substantial amount of experimental evidence suggests however that failure criterion of most soils is not linear particularly in the range of small normal stresses. Particulate media like soils usually have very small tensile strength; and the present work focuses attention on zero tensile strength materials obeying a limiting form of the non-linear Hoek and Brown empirical strength criterion. Previous investigations of the effect of strength function non-linearity on results of slope stability calculations were based on limiting equilibrium procedures that include various approximations and static assumptions. The present study presents complete results (safety factors and failure modes) of a rigorous variational limiting equilibrium analysis which does not include kinematical or static assumptions. Linear and non-linear failure criteria depend on different strength parameters and significant comparison of the effect of strength function non-linearity on stability of slopes is possible only for a given state of experimental information (i.e. a given data set). By performing such a comparison it is shown that strength function non-linearity has very significant effect on the results of slope stability calculations for relatively steep and very gentle slope inclinations. In both of these inclination ranges the non-linear analysis results in conservative slope design compared with the conventional Mohr-Coulomb criterion. In the range of intermediate slope inclinations, analysis based on the Mohr-Coulomb criterion is acceptable, resulting in more conservative results than the present non-linear analysis.

View full abstract

A series of dynamic centrifuge tests (1:50 scale) of embankments founded on clay clarified the following: (i) a dynamic load of α = 250 gal, simulating an earthquake, induced large deformation and settlement in both the embankment and clay under a gravity field of 50 G; (ii) deformation and instability were minimised by installing a cement-treated layer up to the bottom of the clay foundation. Despite these findings, it is still uncertain whether an embankment would be safer than a clay foundation or which embankments or soil layers would give rise to instability during strong motion earthquakes. This paper presents results of a dynamic FE method applied to that problem. Increasing strength of cement-treated clay increases overall stability of the embankment and foundation together. A design chart has been produced to indicate which combination of embankment and cement-treated clay foundations would tend to be more stable during earthquakes. Engineers can decide whether it is necessary to strengthen the embankment or the foundation layer.

View full abstract