OCR Prediction Using Support Vector Machine Based on Piezocone Data

The determination of the overconsolidation ratio (OCR) of clay deposits is an important task in geotechnical engineering practice. This paper examines the potential of a support vector machine (SVM) for predicting the OCR of clays from piezocone penetration test data. SVM is a statistical learning theory based on a structural risk minimization principle that minimizes both error and weight terms. The five input variables used for the SVM model for prediction of OCR are the corrected cone resistance (qt), vertical total stress (σv), hydrostatic pore pressure (u0), pore pressure at the cone tip (u1), and the pore pressure just above the cone base (u2). Sensitivity analysis has been performed to investigate the relative importance of each of the input parameters. From the sensitivity analysis, it is clear that qt=primary in situ data influenced by OCR followed by σv, u0, u2, and u1. Comparison between SVM and some of the traditional interpretation methods is also presented. The results of this study have shown that the SVM approach has the potential to be a practical tool for determination of OCR.     

Determination of Hydraulic Conductivity Using Piezocone Penetration Test

An elastoplastic, finite-strain, coupled theory of mixtures in an updated Lagrangian reference frame is applied to the piezocone penetration test to estimate the hydraulic conductivity of the soil via analysis of the steady-state excess pore pressure generated during piezocone penetration. The results of this approach were compared with piezocone penetration test data. It showed that reliable hydraulic conductivities can be estimated conveniently without performing pore pressure dissipation tests. This study also shows that the change in the dimensionless excess pore pressure (excess pore pressure is normalized by the effective overburden pressure) at the cone tip is almost constant when the dimensionless hydraulic conductivity (hydraulic conductivity is normalized by the penetration speed and cone radius, hereafter called DLHC) is less than 10?7 or greater than 10?4. It is also shown that the drainage condition around the cone tip is close to a fully undrained condition when the DLHC of the soil is less than 10?7, while it is close to a fully drained condition when the DLHC of the soil is greater than 10?4.     

High Overburden Stress Effects in Liquefaction Analyses

A reevaluation is presented of two factors that can strongly affect the estimation of liquefaction resistance for clean sands under high effective overburden stresses (σv′): the relation used to normalize penetration resistances to a σv′ of 1 atm (i.e., CN), and the adjustment factor for the effects of σv′ on cyclic resistance ratio (i.e., Kσ). These two factors have been investigated in a number of ways and several relations exist for each of them. An improved CN relation is developed based on cone penetration theory and validation against calibration chamber test data for both cone penetration and standard penetration tests. A relative state parameter index (ξR) is shown to provide a consistent theoretical framework for interrelating the penetration and cyclic loading resistances. It is subsequently shown that the CN and Kσ relations are interrelated through the sand properties and relative density (DR) in ways that have compensating effects on the predicted cyclic resistance. The derived relations provide an improved representation of the effects of high σv′ levels, and reduce the conservatism that results when some established relations are extended to σv′ levels higher than they were calibrated for.     

Spatial Distribution of Excess Pore-Water Pressure due to Piezocone Penetration in Overconsolidated Clay

This paper presents the results of an analysis of the spatial distribution of the excess pore-water pressure induced by piezocone penetration into overconsolidated clays. From the experimental results obtained for moderately and heavily overconsolidated clays, it was observed that the excess pore-water pressure increases monotonically from the piezocone surface to the outer boundary of the shear zone and then decreases logarithmically, approaching zero at the outer boundary of the plastic zone. It was also found that the size of the shear zone decreases from approximately 2.2 to 1.5 times the cone radius with increasing overconsolidation ratio (OCR), whereas the plastic radius is about 11 times the piezocone radius, regardless of the OCR. The expressions developed in this study based on the modified Cam clay model and the cylindrical cavity expansion theory, which take into consideration the effects of the strain rate and stress anisotropy, provide a good prediction of the initial pore-water pressure at the piezocone location. The method of predicting the spatial distribution of excess pore-water pressure proposed in this study is based on a linearly increasing Δushear in the shear zone and a logarithmically decreasing Δuoct, and was verified by comparing the pore-water pressure measured in overconsolidated specimens in the calibration chamber.     

Neural Networks for Profiling Stress History of Clays from PCPT Data

This paper evaluates the feasibility of using artificial neural network (ANN) models for estimating the overconsolidation ratio (OCR) of clays from piezocone penetration tests (PCPT). Three feed-forward, back-propagation ANN models are developed, and trained using actual PCPT records from test sites around the world. The soil deposits range from soft, normally consolidated intact clays to very stiff, heavily overconsolidated fissured clays. ANN model 1 is a general model applicable for both intact and fissured clays. ANN model 2 is suited for intact clays, and ANN model 3 is applicable to fissured clays only. The models are validated using new PCPT data (not used for training), and by comparing model predictions with reference OCR values obtained from oedometer tests. For intact clays, ANN model 2 gives better OCR estimates compared to ANN model 1. For fissured clays, ANN model 3 gives better estimates compared to ANN model 1. Some of the existing interpretation methods are reviewed. Compared to the existing methods, ANN models 2 and 3 give very good estimates of OCR.     

Numerical Parametric Study of Piezocone Penetration Test in Clays

This paper presents a numerical model for the simulation of the piezocone penetration test that is used to carry out a parametric study of the piezocone penetration test in cohesive soils. The piezocone penetration is numerically simulated using an axisymmetric elasto-plastic large deformation finite-element analysis code. The numerical simulation is accomplished in two stages. First, the piezocone is expanded radially from an initial small radius (0.1ro) to the piezocone radius, ro, at the specified depth. Second, the continuous penetration of the piezocone is simulated by applying incremental vertical displacements of the nodes representing the piezocone boundary. The constraint approach is used to model the soil-piezocone interface friction. The Mohr-Coulomb frictional criterion is used to define the sliding potential of the nodes. The main objective of this paper is to present the numerical model and to investigate the effect of the lateral and vertical stresses and the overconsolidation ratio on the cone tip resistance and the developed excess pore pressure around the piezocone. The variation of the horizontal and vertical hydraulic conductivity coefficients on the developed spatial excess pore pressure and its dissipation are also investigated. The results of the numerical study are also compared with the miniature piezocone penetration tests in cohesive soil specimens conducted at the Louisiana State University Calibration Chamber. Results of this study are in good agreement with the measured values.     

Undrained Shear Strength of Pleistocene Clay in Osaka Bay

This study presents the undrained shear characteristics of Holocene and Pleistocene clay samples from depths of 20–200 m under the seabed in Osaka Bay. Automated triaxial K0 consolidation tests and anisotropically consolidated-undrained triaxial compression and extension tests are conducted using the recompression method. The average undrained strength ratio (su/σv0′) is 0.33 (SD = 0.03) when the extension strength is defined as the peak strength or the strength at an axial strain of 15%, while su/σv0′ is 0.29 (SD = 0.04) when the extension strength is defined as the shear stress at the axial strain corresponding to the peak compression strength. Circular arc stability analyses are carried out with the modified Fellenius and Bishop methods for the design cross section of the seawall structure of the Kansai International Airport to study the effects of different definitions of shear strength. The seawall is founded on 19 m of soft Holocene clay and 10 m of Pleistocene sand overlying the Pleistocene clay. The stability analyses show that the factor of safety and depth of the critical circle (i.e., above versus below the sand layer) are sharply affected by both the value of su/σv0′ (0.33 versus 0.29) and the method of slices (Fellenius versus Bishop). The marginal stability calls for careful monitoring of construction with field instrumentation.     

Finite-Element Analysis of Inclined Piezocone Penetration Test in Clays

A three-dimensional finite-element analysis was performed to analyze the effect of soil anisotropy on the inclined piezocone penetration test in normally consolidated clay. The piezocone penetration was numerically simulated based on a large strain formulation using the commercial finite-element code ABAQUS, and the anisotropic modified cam clay model (AMCCM) was chosen and implemented into ABAQUS through the user subroutine UMAT. For verification purposes, numerical simulations were first performed on previously conducted calibration chamber tests, and the predicted results were compared with the measured values. For different initial stress conditions and different penetration angles, the cone tip resistance profile; excess pore pressure profile at the cone tip; typical stress, strain and excess pore pressure distributions around the cone; and excess pore pressure dissipation at the cone tip are provided. This study shows that when the initial stress state is anisotropic, the soil behavior is different under different angles of penetration.     

Numerical Simulation of Vertical Pullout of Plate Anchors in Clay

The behavior of strip and circular plate anchors during vertical pullout in uniform and normally consolidated clays was studied in this paper by means of small strain and large deformation finite-element analyses. Both fully bonded (attached), and “vented” (no suction on rear face), anchors were considered. The current numerical results were compared with existing laboratory test data, finite-element results, and analytical solutions. This study showed that, in small strain analysis, the scatter of existing data was mainly due to the effect of soil stiffness. In large deformation analysis, when soil and anchor base were attached with suction, the pullout capacity factor formed a unique curve independent of the soil strength (su), soil effective unit weight (γ′) and anchor size (B=width of strip anchor and D=diameter of circular anchor). The transitional embedment depth ratio, HSD/B or HSD/D, (where HSD=transition depth between shallow and deep embedment) was 1.4 for a strip anchor and 0.75 for a circular anchor. The ultimate pullout capacity factors (Nc) for deep embedment were 11.6 and 11.7 for smooth and rough strip anchors and 13.1 and 13.7 for smooth and rough circular anchors, respectively. However, when the anchor base was vented, the soil stayed attached to the anchor base for deep embedment, and the pullout capacity was therefore the same as for the attached anchor. The separation depth ratio, Hs/B or Hs/D, (where Hs=embedment depth at which the soil and anchor base separated) was found to increase linearly with the normalized strength ratio, su/γ′B or su/γ′D.     

Analyzing Liquefaction-Induced Lateral Spreads Using Strength Ratios

The writers backanalyzed 39 well-documented liquefaction-induced lateral spreads in terms of a mobilized strength ratio, su(mob)/σvo′ using the Newmark sliding block method. Based on the inverse analyses results, we found that the backcalculated strength ratios mobilized during lateral spreads can be directly correlated to normalized cone penetration test tip resistance and standard penetration test blow count. Remarkably, Newmark analysis-based strength ratios mobilized during these lateral spreads essentially coincide with liquefied strength ratios backcalculated from liquefaction flow failures. The mobilized strength ratios appear to be independent of the magnitude of lateral displacement (at least for displacements greater than 15?cm) and the strength of shaking (in terms of peak ground acceleration). Furthermore, the mobilized strength ratios backcalculated from these cases appear to be consistent for a given depositional environment and do not appear to be severely impacted by potential water layer formation.     

Undrained Anisotropic Monotonic Behavior of Sand from In Situ Tests

A procedure for estimating the undrained stress-strain behavior of sand from drained self-boring pressuremeter and seismic piezocone penetration tests is proposed in this paper. The procedure offers an inexpensive alternative to laboratory testing and avoids the uncertainty of the empirical methods based on index measurements such as the Standard Penetration Test blow count and the tip resistance in a Piezocone Penetration Test (CPTU). To check its validity, the proposed procedure was used to infer the undrained triaxial stress-strain curves and the results were compared with laboratory triaxial tests on undisturbed samples. The undrained limit equilibrium stability of a dike was also assessed using the inferred stress-strain behavior to illustrate the usefulness of the procedure. The result of the stability analysis was found to be in qualitative agreement with the observed performance of the dike during a recent field experiment attempting to trigger static liquefaction.     

New Structure-Based Model for Estimating Undrained Shear Strength

This paper presents an experimental study of the strength in anisotropic clays by means of centrifuge model, cone penetration, and vane shear tests. To understand the effects of void ratio, overconsolidation ratio, and testing rate on the undrained shear strength (Su) of anisotropic Speswhite clay, a new centrifugal testing technique is designed to obtain constant overconsolidation ratio (OCR) profiles with varying void ratios (e), called the “descending gravity test.” The parameters controlling the generation of peak shear strength are quantified. As a result of this function, a new material and rate-dependent surface is defined in the e-OCR-Su space, which is identified as a “structural state capacity surface” since it relates the anisotropic structure to structure inherent capacity and properties. A new function for the estimation of excess pore pressure (uex) generated by cone penetration is found. By combining the strength and pore pressure functions a new model is proposed, called the “CU model.” The CU model is a structure-based model that provides reliable estimates of shear strength for in situ saturated clays using the knowledge of void and overconsolidation ratios. Finally, by combining Su-e-OCR and uex-e-OCR relationships, it estimates the void ratio and OCR profiles of anisotropic clays from piezocone penetration test results.     

Coupled Use of Cone Tip Resistance and Small Strain Shear Modulus to Assess Liquefaction Potential

Resistance against earthquake-related liquefaction is usually assessed using relationships between an index of soil strength such as normalized cone tip resistance and the cyclic resistance ratio (CRR) developed from observed field performance. The alternative approach based on laboratory testing is rarely used, mainly because of the apprehension that laboratory results may not reflect field behavior since the quality of laboratory data is often compromised by sampling disturbance. In this study, a database of laboratory data obtained mainly from cyclic testing of frozen (undisturbed) samples and in situ index measurements from near sampling locations comprised of cone tip resistance, qc, and shear wave velocity, Vs, have been assembled. These data indicate that neither normalized cone tip resistance nor normalized shear wave velocity individually correlate well with laboratory-measured CRR. However, the ratio of qc to the small strain shear modulus, G0, relates reasonably with CRR via separate correlations depending on geologic age. The derived qc/G0-CRR relationships were also found to be consistent with earthquake field-performance case histories.     

Determining Intrinsic Compressibility of Fine-Grained Soils

Results of laboratory oedometer tests on reconstituted specimens of four clays prepared at different initial water contents, ranging from the liquid limit to 1.75 times the liquid limit, show that the intrinsic compression line may not be “unique” for a given soil. This suggests that the “intrinsic” parameter Iv, which is based on the constants of intrinsic compressibility, e100?, (void ratio corresponding to σv′ = 100?kPa), and Cc?, (e100??e1000?), may in fact not be a truly intrinsic parameter of the soil, but is dependent on sample preparation. The positioning of the normalized compression curve in e–log–σv′ space is significantly influenced by the initial remolding water content, therefore resulting in differing values of e100? for a given soil depending on the initial water content. The influence of initial water content was greater for kaolinitic and illitic clay than for montmorillonitic clay. It is hypothesized that the difference in behavior may be attributed to differences in mineralogy. The results illustrate that caution should be used when comparing tests results from widespread sources and suggest that a standard level of initial water content be used to evaluate the intrinsic compressibility.     

Consolidation and Permeability Properties of Singapore Marine Clay

The consolidation and permeability characteristics of the Singapore marine clay were investigated by laboratory and in situ tests. The coefficient of consolidation and the coefficient of permeability of the soil in both the vertical and horizontal directions, cv and ch and kv and kh, respectively, were determined by oedometer, constant rate of strain, Rowe cell, piezocone (CPTU), flat dilatometer, self-boring pressuremeter (SBPT), and BAT permeameter tests. The ch value of the clay was also back calculated from field monitoring data. Comparisons of the results obtained by different methods indicate that the ch value determined by CPTU is generally in good agreement with that by Rowe cell tests, and the ch value determined by SBPT is often the highest among all the measurements. The back calculated ch value is lower than that measured by other tests, which could be possibly due to the smear effect incurred during the installation of vertical drains.     

Evaluation of Undrained Shear Strength Using Full-Flow Penetrometers

Full-flow penetrometers (the T-bar and ball) are increasingly used on sites with thick deposits of soft clays, particularly prevalent offshore. Full-flow penetration tests were performed at five international well-characterized soft clay test sites to assess the use of full-flow penetrometers to estimate undrained shear strength. Field vane shear data were used as the reference undrained strength. Statistical analyses of strength factors indicates that full-flow penetrometers provide an estimate of undrained shear strength at a similar level of reliability compared to the piezocone. Relationships for estimating the strength factor and soil sensitivity using only full-flow penetrometer data obtained during initial penetration and extraction are developed. A strong dependence of the strength factor on sensitivity was identified and can be used for the estimation of undrained strength. The effectiveness and use of the developed correlations are demonstrated through their application at an additional test site.     

CPT-Based Probabilistic Soil Characterization and Classification

Due to lack of soil sampling during conventional cone penetration testing, it is necessary to characterize and classify soils based on tip and sleeve friction values as well as pore pressure induced during and after penetration. Currently available semiempirical methods exhibit a significant variability in the estimation of soil type. Within the confines of this paper it is attempted to present a new probabilistic cone penetration test (CPT)-based soil characterization and classification methodology, which addresses the uncertainties intrinsic to the problem. For this purpose, a database composed of normalized corrected cone tip resistance (qt,1,net), normalized friction ratio (FR), fines content (FC), liquid limit (LL), plasticity index (PI), and soil type based on the unified soil classification system was complied. Soil classification was performed by laboratory testing of the standard penetration test disturbed samples retrieved from the boreholes within mostly 2?m of each CPT hole. The resulting database was probabilistically assessed through Bayesian updating methodology allowing full and consistent representation of relevant uncertainties, including (1) model imperfection; (2) statistical uncertainty; and (3) inherent variability. As a conclusion, different sets of FC, LL, PI, and A-line boundary curves along with a new CPT-based, simplified soil classification scheme are proposed in the qt,1,net and FR domain. Probabilistic uses of the proposed models are illustrated through a set of illustrative examples.     

Piezocone Penetration Testing in Atlantic Piedmont Residuum

Piezocone soundings were performed in residual silts and fine sands of the U.S. Atlantic Piedmont geologic province to investigate penetration pore-water pressures and subsequent dissipation characteristics in these “nontextbook” geomaterials. Piedmont residuum has been formed by the in-place weathering of Paleozoic schist, gneiss, and granite. Pore-water pressures were measured at two positions: midface of the cone tip u1 and at the shoulder or behind the tip u2. At the standard rate of 2 cm∕s, penetration pore-water pressures on the midface element are fairly high positive values (u1 = +600 kPa ± 200 kPa), yet readings at the shoulder element were negative (u2 ≈ ?90+ kPa). At an increased penetration rate of 20 cm∕s, midface and shoulder readings increased in magnitude and were positive (u1 ≈ +800 kPa and u2 ≈ +200 kPa); however, measured tip resistance qc exhibited an unusual decrease, per comparisons with data available in the literature. Upon halt of penetration, both u1 and u2 decayed quickly and reached hydrostatic values u0 within 1–2 min.     

Nonlinear Dynamic Properties of a Fibrous Organic Soil

The nonlinear dynamic properties of a fibrous peaty organic soil beneath a levee in the Sacramento–San Joaquin Delta in California are described herein. Thin-walled tube samples were obtained from four locations between the levee crest and the free field such that the in situ vertical effective stresses (σvo′) ranged from about 12 kPa in the free field to about 135 kPa beneath the levee crest. The peaty organic soil was very soft and highly compressible in the free field with initial water contents (wo) of 236–588% and shear wave velocities (Vs) of typically 22–27 m/s, and moderately firm beneath the levee crest with wo of 152–240% and Vs of typically 88–129 m/s. Stress–strain measurements in a cyclic triaxial device showed that the normalized secant shear modulus (G/Gmax) and equivalent damping ratio (ξ) versus cyclic shear strain amplitude (γc) relations were dependent on the consolidation stress (σvc′). Tests involving prior overstraining followed by reconsolidation showed that the effects of sample disturbance were likely small. Stress history, creep, and loading frequency effects were also examined. Tests on reconstituted specimens provided supplementary data on the functional relation between maximum shear modulus (Gmax) and consolidation stress conditions. Summary relations are provided for G/Gmax and ξ versus γc and for Gmax versus σvc′.     

Sampling Disturbance Effects in Normally Consolidated Clays

The effects of sampling disturbance are investigated by performing single element triaxial tests in which specimens of normally consolidated resedimented Boston blue clay are disturbed according to the “perfect sampling approach” (PSA) and the “ideal sampling approach” (ISA). The effects of PSA and ISA disturbance on the compression and undrained shear behavior of the soil are quantified by comparison with the intact behavior. The results indicate that the release of shear stress associated with PSA disturbance causes a modest change in the engineering properties of the soil. The effects of ISA disturbance are, on the other hand, very significant and increase systematically with the amplitude of the strain imposed. An increase in disturbance causes a decrease in the compression ratio, a decrease in the undrained strength, and an increase in the strain at failure and the recompression ratio, but has a minor effect on the preconsolidation pressure. These effects derive from the decrease in effective stress and from the damage to the soil fabric that occur as a result of sampling. The loss in undrained strength is primarily controlled by the decrease in effective stress and the post-disturbance strength ratio (cu/σs′) may be related to the “induced” overconsolidation ratio (IOCR = σvc′/σs′) through a SHANSEP equation.