Discrete element method analysis of non‐coaxial flow under rotational shear

This study focuses on non‐coaxial flow behavior of cohesionless soil undergoing cyclic rotational shear, with a special interest in the effects of particle‐scale characteristics. To this end, we perform a series of 2D discrete element simulations with various particle shapes, inter‐particle coefficient of friction, initial density, and stress ratios. The validity and efficacy of the numerical model is established by systematically comparing numerical simulation results with existing laboratory testing results. Such comparison shows that the numerical simulations are capable of capturing mechanical behavior observed in laboratory testing under rotational shear. We further demonstrate and quantify a strong yet simple relationship between the deviatoric part of the normalized strain increment and the non‐coaxial angle, denoted by and ψ, respectively. This quantitative correlation between ψ and is independent of applied stress ratio, initial and current void ratio, and the number of cycles applied, but dependent on the principal stress orientation and particle‐scale characteristics. At the same , specimens with higher inter‐particle friction angle or smaller particle aspect ratio show greater non‐coaxial angles. A simple model is able to fit this ψ‐ relationship well, which provides a useful relationship that can be exploited in developing constitutive models for rotational shearing. Copyright © 2014 John Wiley & Sons, Ltd.     

Improved rotational hardening rule for cohesive soils and definition of inherent anisotropy

Improved, microfabric‐inspired rotational hardening rules for the plastic potential and bounding surfaces associated with the generalized bounding surface model for cohesive soils are presented. These hardening rules include 2 new functions, f_η and , that improve the simulation of anisotropically consolidated cohesive soils. Three model parameters are associated with the improved hardening rules. A detailed procedure for obtaining suitable values for these parameters is presented. The first 2 parameters affect the simulation of constant stress ratio loading where, because of the presence of f_η, the third parameter is inactive. The second new function, , accelerates the rotation of the plastic potential and bounding surfaces during shearing, which is particularly important for overconsolidated soils tested in extension. This paper also describes the proper manner in which to define the inherent anisotropy. This seemingly straightforward test has rarely been discussed in sufficient detail.     

Discrete element modeling of tool‐rock interaction II: rock indentation

The failure mechanisms induced by a wedge‐shaped tool indenting normally against a rock surface are investigated using the discrete element method (DEM). The main focus of this study is to explore the conditions controlling the transition from a ductile to a brittle mode of failure. The development of a damage zone and the initiation and propagation of a brittle fracture is well captured by the DEM simulations. The numerical results support the conjecture that initiation of brittle fractures is governed by a scaled flaw length Λ, a ratio between the flaw size λ and the characteristic length (where K_Ic is the toughness and σ_c the uniaxial compressive strength). The size of the damage zone agrees well with analytical predictions based on the cavity expansion model. The effects of a far‐field confining stress and the existence of a relief surface near the indenter are also examined.Copyright © 2012 John Wiley & Sons, Ltd.     

Estimation of diaphragm wall deflections for deep braced excavation in anisotropic clays using ensemble learning

This paper adopts the NGI-ADP soil model to carry out finite element analysis,based on which the effects of soft clay anisotropy on the diaphragm wall deflections in the braced excavation were evaluated.More than one thousand finite element cases were numerically analyzed,followed by extensive parametric studies.Surrogate models were developed via ensemble learning methods(ELMs),including the e Xtreme Gradient Boosting(XGBoost),and Random Forest Regression(RFR)to predict the maximum lateral wall deformation(δhmax).Then the results of ELMs were compared with conventional soft computing methods such as Decision Tree Regression(DTR),Multilayer Perceptron Regression(MLPR),and Multivariate Adaptive Regression Splines(MARS).This study presents a cutting-edge application of ensemble learning in geotechnical engineering and a reasonable methodology that allows engineers to determine the wall deflection in a fast,alternative way.     

Estimation of Pullout Capacity of Vertical Plate Anchors in Cohesionless Soil Using MARS

Vertical plate anchors provide an economical solution to safely resist the large horizontal forces experienced by the foundation of different structures such as bulkheads, sheet piles, retaining walls and so forth. This paper develops a multivariate adaptive regression spline (MARS) model-based approach for the determination of horizontal pullout capacity (P _u ) of vertical plate anchors buried in cohesionless soil by utilizing experimental results reported by different researchers. Based on the collection of forty different pullout experimental test results reported in the literature for anchors buried in loose to dense cohesionless soil with an embedment ratio ranges from 1 to 5, a predictive approach for P _u of vertical plate anchors has been developed in terms of non-dimensional pullout coefficient (M _γq ). The capability of the proposed MARS model for estimating the values of M _γq is examined by comparing the results obtained in the present study with those methods available in the literature. Using different statistical error measure criteria, this study indicates that the present approach is efficient in estimating the horizontal pullout capacity of vertical plate anchors as compared to other methods. The sensitivity analysis indicates that the embedment ratio (H/h, where H = embedment depth of anchor, and h = height of anchor) and internal friction angle (?) of soil mass are the two most important parameters for the evaluation of non-dimensional pullout coefficient (M _γq ) using the proposed MARS model.     

A novel insight into vertical ground motion modelling in earthquake engineering

Recent observations of failure and damage of buildings and structures under seismic action has led to an increasing interest for an in-depth analysis of the vertical component of site ground motion. In particular, when dealing with saturated soils, the current engineering practice does not usually go beyond the simplified $u$–$p$ formulation of the Biot's equations describing the coupled hydro-mechanical behaviour, thus neglecting some terms of fluid inertial forces, despite the presence of more refined formulations, for example, the $u$–$U$ formulation. Therefore, a theoretical and numerical validation of the $u$–$p$ formulation as compared with the $u$–$U$ formulation is proposed in this work, where the numerical simulations are compared with the analytical solution for the $u$–$p$ formulation, which is also derived and illustrated in this text. The comparison between the two formulations and the analytical solution is provided for different levels of permeability and dynamic actions, which are representative of a wide scenario of site ground properties and seismic hazard in the vertical direction. In particular, the soil response is analysed in terms of acceleration and pore pressure time history, frequency content, acceleration response spectrum, and amplification ratio of acceleration. This study extends the discussion of the limits of applicability of the $u$–$p$ formulation with respect to the rigorous solution of Biot's equations (obtained here with $u$–$U$ formulation) to the context of a complex dynamic regime provided by the vertical components of real earthquake records, and paves the way for further investigations.     

Application of statistical learning algorithms to ultimate bearing capacity of shallow foundation on cohesionless soil

This study employs two statistical learning algorithms (Support Vector Machine (SVM) and Relevance Vector Machine (RVM)) for the determination of ultimate bearing capacity (q_u) of shallow foundation on cohesionless soil. SVM is firmly based on the theory of statistical learning, uses regression technique by introducing varepsilon‐insensitive loss function. RVM is based on a Bayesian formulation of a linear model with an appropriate prior that results in a sparse representation. It also gives variance of predicted data. The inputs of models are width of footing (B), depth of footing (D), footing geometry (L/B), unit weight of sand (γ) and angle of shearing resistance (?). Equations have been developed for the determination of q_u of shallow foundation on cohesionless soil based on the SVM and RVM models. Sensitivity analysis has also been carried out to determine the effect of each input parameter. This study shows that the developed SVM and RVM are robust models for the prediction of q_u of shallow foundation on cohesionless soil. Copyright © 2010 John Wiley & Sons, Ltd.     

On the Study of Pile Driving Formula Based on Blow Counts

Accuracy of predicting pile capacities by pile driving formulas have been investigated. Five test piles were driven up to a depth of about 9 m of clay deposit and the penetrations due to final blows were recorded. The pile bearing capacity of each pile was predicted using 6 different pile driving formulas and the predicted pile capacity was compared with measured pile capacity from the pull up tests. Hiley formula, Modified Engineering News Record (ENR) formula, Janbu formula, Dutch formula, Danish formula, and Gates formula were used. The performance and accuracy of each formula was evaluated and the correlation coefficient of each pile driving formula was determined for a more accurate pile capacity prediction. Methods used to evaluate the performance of each formula were; (1) the best fit line for Q _p versus Q _m (2) cumulative probability for Q _p/Q _m and (3) the arithmetic mean and standard deviation for Q _p/Q _m. From the study, it was found that using Dutch formula provided the most accurate pile capacity estimate compared to the other formulas with an average of 7% deviation from value obtained from the field pull up test. It was followed by the Danish formula, Janbu formula, Hiley formula, Modified ENR formula, and Gates formula. The ability to predict the accuracy of estimating pile capacity using an appropriate method is very important and valuable to contractors, developers, geotechnical engineers, and manufacturers.     

Development of analytical models to estimate the increase in pile capacity with time (pile setup) from soil properties

This paper presents the analyses of twelve prestressed concrete (PSC) instrumented test piles that were driven in different bridge construction projects of Louisiana in order to develop analytical models to estimate the increase in pile capacity with time or pile setup. The twelve test piles were driven mainly in cohesive soils. Detailed soil characterizations including laboratory and in situ tests were conducted to determine the different soil properties. The test piles were instrumented with vibrating wire strain gauges, piezometers, pressure cells that were monitored during the whole testing period. Several static load tests (SLTs) and dynamic load tests were conducted on each test pile at different times after end of driving (EOD) to quantify the magnitude and rate of setup. Measurements of load tests confirmed that pile capacity increases almost linearly with the logarithm of time elapsed after EOD. Case pile wave analysis program was performed on the restrikes data and was used along with the load distribution plots from the SLTs to evaluate the increase in skin friction capacity of individual soil layers along the length of the piles. The logarithmic linear setup parameter “A” for unit skin friction was calculated of the 70 individual clayey soil layers and was correlated with different soil properties such as undrained shear strength (S_u), plasticity index, vertical coefficient of consolidation (c_v), over consolidation ratio and sensitivity (S_t). Nonlinear multivariable regression analyses were performed, and three different empirical models are proposed to predict the pile setup parameter “A” as a function of soil properties. For verification, the subsurface soil conditions and setup information for additional 18 PSC piles collected from local database were used to compare the measured versus predicted “A” parameters from the proposed models, which showed good agreement.

     

Assessment of direct CPT and CPTU methods for predicting the ultimate bearing capacity of single piles

Twelve methods to determine axial pile capacity directly based on cone penetration test (CPT) and piezocone penetration test (CPTU) data are presented, compared and evaluated. Analyses and evaluation were conducted on three types of piles of different size and length. All the tested piles have failed at the end of static load test. Both the CPT methods and the CPTU methods were used to estimate the load bearing capacities of the investigated piles (Q_p). The static load test was performed to determine the measured load bearing capacities (Q_m). The pile capacities determined through different methods were compared with the measured values obtained from the static load tests. Two criteria were selected as bases of evaluation: the best fit line for Q_p versus Q_m and the arithmetic mean and standard deviation for the ratio Q_p/Q_m. Results of the analyses showed that the best methods for determining pile capacity are the two CPTU methods. Furthermore, the CPTU method is simple, easy to apply, and not influenced by the subjective judgements of operating staff. Therefore, it is quite suitable for the application in pile engineering practice.     

Effect of soil variability on the bearing capacity of clay and in slope stability problems

Site-specific geotechnical data are always random and variable in space. In the present study, a procedure for quantifying the variability in geotechnical characterization and design parameters is discussed using the site-specific cone tip resistance data (q_c) obtained from static cone penetration test (SCPT). The parameters for the spatial variability modeling of geotechnical parameters i.e. (i) existing trend function in the in situ q_c data; (ii) second moment statistics i.e. analysis of mean, variance, and auto-correlation structure of the soil strength and stiffness parameters; and (iii) inputs from the spatial correlation analysis, are utilized in the numerical modeling procedures using the finite difference numerical code FLAC 5.0. The influence of consideration of spatially variable soil parameters on the reliability-based geotechnical deign is studied for the two cases i.e. (a) bearing capacity analysis of a shallow foundation resting on a clayey soil, and (b) analysis of stability and deformation pattern of a cohesive-frictional soil slope. The study highlights the procedure for conducting a site-specific study using field test data such as SCPT in geotechnical analysis and demonstrates that a few additional computations involving soil variability provide a better insight into the role of variability in designs.     

Application of the Dynamic Cone Penetrometer (DCP) for determination of the engineering parameters of sandy soils

Determination of the in situ engineering properties of foundation materials has always been a challenge for geotechnical engineers and, thus, several methods have been developed so far. Dynamic Cone Penetration (DCP) test is one of the most versatile amongst them. In the present research, a light weight simple DCP device was developed and used for evaluation of the engineering properties of sandy soils in laboratory conditions. The device consisted of an 8-kg hammer that drops over a height of 575 mm, and drives a 60° cone tip with 20 mm base diameter into the ground. To control the validation of the results, laboratory direct shear and plate load tests were used as reference tests. The soil sample was a poorly graded sandy soil (SP) taken from alluvial deposits of the Tehran plain. All DCP tests and PLTs were undertaken on compacted soil in a mould with 700 mm diameter and 700 mm height. Based on the results of the experiments, the relationships between Dynamic Penetration Index (DPI), relative density (D_r), modulus of elasticity (E), shear modulus (G), modulus of subgrade reaction (K_S), and the friction angle of the soil were obtained with a high coefficient of determination (> 90%). The repeatability of the test results was also evaluated by calculating the coefficient of variations (C_v), which was less than 30% for all tests.     

Influence of random heterogeneity of the friction angle on bearing capacity factor Nγ

ABSTRACT

Probabilistic methods in geotechnical engineering have received a lot of attention during the last decade and different methodologies are used to capture the inherent variability of soil in different geotechnical engineering problems. In this paper, numerical simulations are conducted to obtain the bearing capacity factor, N_γ, for a purely frictional heterogenous soil where the friction angle is modelled as randomly distributed throughout the domain and the effect of its spatial variability on N_γ is investigated. A finite element method, based on the upper bound limit analysis was combined with random field theory and linear programming to develop a probabilistic analysis. Monte Carlo simulations were performed and the effect of the variability of the friction angle defined by statistical parameters on the bearing capacity factor was investigated. Results show that the mean bearing capacity factor N_γ of a footing on a spatially variable cohesionless soil is generally higher than the deterministic N_γ obtained from a constant mean value. Increasing the heterogeneity of the friction angle by an increase in the coefficient of variation generally increases this deviation. This can be explained by the nonlinearity of the relationship between N_γ and the friction angle.     

Stratigraphic delineation by three-dimensional clustering of piezocone data

Piezocone soundings are a fast and economical approach for geotechnical site characterization, providing three separate and continuous channels of data with depth, including: tip resistance q _T, porewater pressure u ₂ and sleeve friction f _s. Literally hundreds to thousands of data points are collected by a single sounding. Since these readings are functions of both soil type and soil behaviour, they can be used for the delineation of soil stratigraphy.

One way to process large amounts of data involves clustering. Cluster analysis is an efficient statistical way to analyse the stratigraphic vertical profiling of geomaterials and means to detect the inherent similarity between data sets and group them together. Clustering in previous geotechnical research was based on only two channels of piezocone data (q _T and u₂). The method works well for soils that are under the groundwater table and was applied to soundings in clay deposits.

In the present paper, a new cluster analysis approach is developed based on all three channels of data, thus extending the method to soils above the water table and applicable to sands, silts, and clays. Example soil profiles derived by three-channel cluster analysis are presented herein and compared with conventional soil boring and sampling data.     

Large-scale pile tests in Mercia mudstone: Data analysis and evaluation of current design methods

Full-scale load tests were carried out on six instrumented large diameter bored, cast in-situ piles formed in Mercia mudstone, as part of the design of a new Viaduct in Cardiff, UK. In this paper, the results from six test piles and extensive data from 218 ground investigation boreholes are systematically processed in order to study the load transfer and resistance mechanisms in Mercia mudstone. Data from strain gauges embedded in each pile are first analysed to calibrate the load-deformation relationship of each pile as-built, taking into account (i) the non-linearity of concrete and (ii) the effect of partial steel encasement on pile stiffness at various levels. The shaft and base capacity of the piles are each predicted using 10 calculation methods belonging to the four basic categories: (i) Undrained analysis, (ii) Drained analysis, (iii) Mixed approach and (iv) Empirical correlation. It is found that the shaft capacity prediction methods are moderately consistent. The standard deviations of the ratio Q _sp/Q _sm of predicted to observed shaft capacity lies in the range 0.06–0.24. However, 8 of these methods are over-conservative, giving Q _sp/Q _sm values in the range 0.29–0.67. The remaining two methods yield Q _sp/Q _sm = 1.01 and 1.49. In contrast, the prediction methods for base capacity are found to be much less consistent. The ratio Q _bp/Q _bm of predicted to measured base capacity falls in the interval 0.52–1.93, with corresponding standard deviations of 0.16–0.82. This revised version was published online in July 2006 with corrections to the Cover Date.     

Least‐square support vector machine applied to settlement of shallow foundations on cohesionless soils

This paper examines the potential of least‐square support vector machine (LSVVM) in the prediction of settlement of shallow foundation on cohesionless soil. In LSSVM, Vapnik's ε‐insensitive loss function has been replaced by a cost function that corresponds to a form of ridge regression. The LSSVM involves equality instead of inequality constraints and works with a least‐squares cost function. The five input variables used for the LSSVM for the prediction of settlement are footing width (B), footing length (L), footing net applied pressure (P), average standard penetration test value (N) and footing embedment depth (d). Comparison between LSSVM and some of the traditional interpretation methods are also presented. LSSVM has been used to compute error bar. The results presented in this paper clearly highlight that the LSSVM is a robust tool for prediction of settlement of shallow foundation on cohesionless soil. Copyright © 2008 John Wiley & Sons, Ltd.     

Interpretation of load tests on bored piles in the city of Matera

This paper briefly reviews several calculation methods to evaluate the bearing capacity expressed in terms of undrained strength (c_u) of piles bored in clay for their entire length and of piles whose tip is embedded into weak rock. The scope of the paper is to compare the results obtained with those from full-scale pile tests. These tests were carried out within the city of Matera which is well studied from a geotechnical point of view and for which there are statistically significant data on the geomechanical properties of the Subapennine Blue clays and the underlying Gravina Calcarenites. This represents the first attempt to show, on the basis of laboratory and field data, the influence of variability of the above mentioned soils and rocks on the real behaviour of deep foundations.For piles completely bored into Matera clay, the calculation in terms of total stress are able to interpret sufficiently well the bearing capacity of the piles. For piles having their toe embedded in calcarenite, the variability of the strength of the weak rock presents greater uncertainties in the calculation of base and soft resistance.     

Predicting standard penetration test N-value from cone penetration test data using artificial neural networks

Standard Penetration Test(SPT) and Cone Penetration Test(CPT) are the most frequently used field tests to estimate soil parameters for geotechnical analysis and design.Numerous soil parameters are related to the SPT N-value.In contrast,CPT is becoming more popular for site investigation and geotechnical design.Correlation of CPT data with SPT N-value is very beneficial since most of the field parameters are related to SPT N-values.A back-propagation artificial neural network(ANN) model was developed to predict the N6o-value from CPT data.Data used in this study consisted of 109 CPT-SPT pairs for sand,sandy silt,and silty sand soils.The ANN model input variables are:CPT tip resistance(q_c),effective vertical stress(σ'_v),and CPT sleeve friction(f_s).A different set of SPT-CPT data was used to check the reliability of the developed ANN model.It was shown that ANN model either under-predicted the N_(60)-value by 7-16%or over-predicted it by 7-20%.It is concluded that back-propagation neural networks is a good tool to predict N_(60)-value from CPT data with acceptable accuracy.     

Origin of abundant moonmilk deposits in a subsurface granitic environment

Subsurface granitic environments are scarce and poorly investigated. A multi‐disciplinary approach was used to characterize the abundant moonmilk deposits and associated microbial communities coating the granite walls of the 16th Century Paranhos spring water tunnel in Porto city (north‐west Portugal). It is possible that this study is the first record of moonmilk in an urban subsurface granitic environment. The morphology and texture, mineralogical composition, stable isotope composition and microbial diversity of moonmilk deposits have been studied to infer the processes of moonmilk formation. These whitish secondary mineral deposits are composed of very fine needle‐fibre calcite crystals with different morphologies and density. Calcified filaments of fungal hyphae or bacteria were distinguished by field emission scanning electron microscopy. Stable isotope analysis revealed a meteoric origin of the needle‐fibre calcite, with an important contribution of atmospheric CO ₂, soil respiration and from weathering of Ca‐bearing minerals. The DNA ‐based analyses revealed the presence of micro‐organisms related to urban contamination, including Actinobacteria, mainly represented by Pseudonocardia hispaniensis , Thaumarchaeota and Ascomycota, dominated by Cladosporium . This microbial composition is consistent with groundwater pollution and contamination sources of the overlying urban area, including garages, petrol stations and wastewater pipeline leakage, showing that the Paranhos tunnel is greatly perturbed by anthropogenic activities. Whether the identified micro‐organisms are involved in the formation of the needle‐fibre calcite or not is difficult to demonstrate, but this study evidenced both abiotic and biogenic genesis for the calcite moonmilk in this subsurface granitic environment.     

Growth-rate-dependent prediction of pile set-up and its application in driven pile foundation construction

In this paper research was presented on the development of a growth-rate-dependent model for pile set-up prediction using the restrike and static/statnamic load testing data collected from different projects. The data included: a) restrike records from ninety-five production piles and restrike and load test results of nine instrumented piles driven in soft clays from the relocation project of Highway No. 1 in Louisiana (LA-1); and b) restrike and static load testing data of five fully instrumented square PPC piles driven at four different bridge sites in various soil layers from sands to clays in Florida. Research effort was focused on the prediction of the ultimate shaft resistances with pile set-up formulated using the pile resistance growth rate-dependent model. The timeframe of interest was studied for a practical set-up magnitude such as 90% of the ultimate shaft resistance (Q₉₀). As an application of the rate-dependent model, it was found that piles at the LA-1 relocation project, in general, reached about 95% of the ultimate shaft resistances at the time of 2 weeks after pile installation. The strategy of incorporation of pile set-up in adjusting pile driving criteria or/and design during pile construction, such as the experience-based plan of a two-week waiting period adopted by Louisiana DOTD, was investigated and justified.