Full-scale Pile Load Tests Research Articles

An optimised product-unit neural network with a novel PSO–BP hybrid training algorithm: Applications to load–deformation analysis of axially loaded piles

In general, neural network training is a nonlinear multivariate optimisation problem. Unlike previous studies, in the present study, particle swarm optimisation (PSO) and back-propagation (BP) algorithms were coupled to develop a robust hybrid training algorithm with both local and global search capabilities. To demonstrate the capacity of the proposed model, we applied the model to the predictions of the load–deformation behaviour of axially loaded piles. This is a soil–structure interaction problem, involving a complex mechanism of load transfer from the pile to the supporting geologic medium. A database of full scale pile loading tests is used to train and validate the product-unit network. The results show that the proposed hybrid learning algorithm simulates the load–deformation curve of axially loaded piles more accurately than other BP, PSO, and existing PSO–BP hybrid methods. The network developed using the proposed algorithm also turns out to be more accurate than hyperbolic and t−z models.

Measurement of Strain Distribution along Precast Driven Pile During Pile Load Test

One of the basic parameters required for forecasting pile deformation under loads is the characteristics and profile of the shaft load transfer from the pile to the surrounding soil. This parameter can be obtained by measuring the strain distribution along the pile during full-scale pile load test. For cast-in situ bored pile, the strain measurement can be easily done by fixing the sensors to the reinforcement cage before pouring the concrete. For precast driven / jack-in piles, the application of instrumented full-scale static load tests is far more challenging than their bored pile counterparts due to significant difference in method of pile installation. Due to practical shortcoming of conventional instrumentation method instrumented full-scale static load tests are in fact rarely used in driven pile application in this region. For hollow precast spun concrete piles attempts have been made by geotechnical engineers to measure the strain distribution by installing either an instrumented reinforcement cage or instrumented pipe into the hollow core of spun piles followed by cement grout. However in this method, known as approximate method, the measured characteristics cannot be considered to represent the working piles because the presence of reinforcement/pipe and the grout would alter the stiffness of the test pile. In this paper a method to measure the strain distribution along the installed precast spun concrete pile during full-scale pile load test is described. The method utilizes the hollow core of the spun pile without changing the physical properties e.g. stiffness, of the pile. The main advantages of this method are: the measurement can be done at any location along the pile, the sensors can be retrieved and the measured characteristics are representative of those of the working piles.

高支持力埋込み杭の支持力に必要な根固め部の強度に関する研究

The vertical bearing capacity of bored piles is examined by 4 full scale pile loading tests varying the strength of soil-cement in the enlarged base. Based on 3 failure patterns of the enlarged base, the end bearing capacity is estimated by the soil SPT values and the unconfined compressive strength of the soil-cement. The pile loading tests show that the relation between the strength of soil-cement and pile bearing capacity agrees well to the proposed failure condition by the simple analytical model.

Comprehensive Load Test on Prestressed Concrete Piles in Alluvial Clays and Marl in Savannah, Georgia

This paper introduces a comprehensive full-scale pile load test program on 457-mm (18-in.) square prestressed concrete (PSC) piles in Savannah, Georgia. The program consisted of pile driving analyzer testing during initial pile driving and restrikes, Statnamic tests, static axial compression load tests, and reciprocal lateral load tests. On the basis of the interpretation of the test data, some important findings were obtained: (1) the alluvial clays in Savannah can only provide very limited resistance; (2) the time-dependent pile capacity gain after pile driving (i.e., setup effect) was approximately proportional to the pile embedment length into the Marl formation; (3) the estimated equivalent static pile capacities from the Statnamic tests were comparable to those from the static axial load tests; (4) the Marl formation is a competent bearing stratum for piles; (5) the potential degradation of pile concrete stiffness caused by pile driving should be accounted for in pile capacity analysis; and (6) the piles exhibited stiffer response under the monotonic lateral loading condition than the cyclic lateral loading condition. Finally, predictions on both axial and lateral pile capacities, using the soil parameters derived from the instrumentation data and back-analysis of the pile load tests, were compared with the corresponding pile load test results. The comparisons demonstrate that in combination of the static-bearing capacity formulas and the LPILE program, the developed soil models can make reliable predictions on both the vertical and lateral behaviors of the PSC piles driven through the soft alluvial clays to end bearing in the Marl formation.

Numerical simulations and parametric study of SDCM and DCM piles under full scale axial and lateral loads

The new kind of reinforced Deep Cement Mixing (DCM) pile namely, Stiffened Deep Cement Mixing (SDCM) pile is introduced to mitigate the problems due to the low flexural resistance, quality control problem and unexpected failure of DCM pile. The SDCM pile consists of DCM pile reinforced with concrete core pile. Previously, the full scale pile load test and the full scale embankment loading test were successfully conducted in the field. To continue the study on the behavior of SDCM and DCM piles, the 3D finite element simulations using PLAXIS 3D Foundation Software were conducted in this study. The simulations of full scale pile load test consisted of two categories of testing which are the axial compression and the lateral loading. For DCM C-1 and C-2 piles, the clay–cement cohesion, C DCM , and clay–cement modulus, E DCM , were obtained from simulations as 300 kPa and 200 kPa as well as 60,000 kPa and 40,000 kPa, respectively. For the SDCM piles, the simulation results show that increasing length ratio, L core/L DCM , increased the bearing capacity whereas the sectional area ratio, A core/A DCM , has only small effects on the bearing capacity for the axial compression loading. The verified parameters such as the clay–cement cohesion, C DCM , and clay–cement modulus, E DCM , from simulations of axial compression tests were 200 kPa and 30,000 kPa, respectively. On the other hand, increasing the sectional area ratio, A core/A DCM , significantly influenced the ultimate lateral resistance while the length ratio, L core/L DCM , is not significant in the ultimate lateral load capacity when the length of concrete core pile is longer than 3.5 m. In addition, the tensile strength of DCM, T DCM , and concrete core pile, T core , are very important to the lateral pile resistance. The back-calculation results from simulations of tensile strength were 5000 kPa and 50 kPa for the T core and T DCM , respectively.

Field behaviour of stiffened deep cement mixing piles

Full-scale pile load tests were performed on soft Bangkok clay improved by stiffened deep cement mixing (SDCM) piles and deep cement mixing (DCM) piles installed by jet-mixing to compare their performance. The SDCM pile is a DCM pile with a precast reinforced concrete core pile inserted in the middle. A series of full-scale tests consisting of axial compression, lateral and pullout interface between the concrete core pile and surrounding DCM material were performed. The length of the concrete core pile influenced both the ultimate axial bearing capacity and the settlement of the SDCM piles more than its section area. Furthermore, the section area of the concrete core pile affected both the lateral ultimate bearing capacity and the lateral displacements of SDCM piles significantly. Moreover, the SDCM piles with area ratio (Acore/ADCM) of 0·17 and length ratio (Lcore/LDCM) of 0·85 increased the axial and lateral ultimate bearing capacities so that they were as much as 2·2 and 15 times higher than the corresponding values of DCM piles, respectively. The flexural strength of the DCM pile obtained from the laboratory was 16% of its unconfined compressive strength whereas that obtained from full-scale lateral load tests was much lower at 4–7% of its unconfined compressive strength. The strength reduction factor, Rinter, at the interface between the concrete core pile and DCM pile in the field averaged 0·40, which agreed with the data from laboratory tests of 0·38–0·46.

Effect of joints on p– y behaviour of laterally loaded piles socketed into mudstone

Secondary structures such as joints play an important role in determining the p–y behaviour and the corresponding pile head load-deflection response of laterally loaded piles socketed into jointed rock mass. In this study, a comprehensive numerical modelling was undertaken for piles socketed into mudstone rock using three-dimensional distinct element code, 3DEC. Two full-scale pile load tests results were employed to validate the numerical model and to derive the p–y curves for the mudstone. The validated numerical model was subsequently extended to study the effect of various joint sets on both the p–y and pile head load-deflection behaviour. It was found that an increase in the number of joint sets weakens the p–y behaviour of the rock mass and the corresponding pile head load-deflection response considerably. In addition, it was found that the existing p–y criteria were inadequate to capture the variation in the p–y response of the rock when physical joint sets were incorporated.

Performance of helical piles in oil sand

The results of a comprehensive pile load-test program and observations from field monitoring of helical piles with either a single helix or double helixes installed in oil sand are presented in this paper. Eleven full-scale pile load tests were carried out including axial compression, uplift, and lateral load tests. The results of the full-scale load tests are used to develop a theoretical design model for helical piles installed in oil sand. Test results confirm that the helical pile is a viable deep foundation option for support of heavily loaded structures. The test results also demonstrated that circular-shaft helical piles can resist considerable lateral loads.

Effect of Cracking on the Response of Pile Test under Horizontal Loading

Capacity-based design of structures limits the soil-structure interaction mechanism to the determination of the bearing capacity of a pile group. However, in many cases the criterion for the design of piles to resist lateral loads is not the ultimate lateral capacity but the deflection of the piles. Many procedures exist for estimating the response of single piles and pile groups under lateral loading, ranging from application of empirical relationships and simple closed-form solutions to sophisticated nonlinear numerical procedures. With the aim of investigating the effect of cracking, disregarded by most of the above-mentioned methods, a three-dimensional (3D) nonlinear analysis that accounts for cracking is presented. Response prediction correlates well with the experimental data from a full-scale pile load test. Interesting conclusions have also been drawn regarding the discretization of the computational domain and the combination of 3D numerical nonlinear analysis and the structural beam theory.

Evaluation of High-Capacity Composite Spun Piles

Traits of two broad categories of deep foundations–-replacement and displacement piles–-are best combined by a newly introduced composite spun pile. This composite spun pile is constructed by deep mixing first and then inserting a precast, prestressed, spun concrete pile into the deep mixed column. The spun pile is then driven into an end-bearing layer to meet the criterion of end resistance. The pile uses a larger peripheral area of soil–cement mix to transfer load to its surrounding ground at the same time it uses higher stiffness of the spun pile to minimize axial compression. The piling operation is relatively quiet, minimizes spoil problems, and facilitates better quality control. A pile design principle for the composite pile is presented. The performance of this new pile system was verified by five instrumented field full-scale pile load tests. Improvement on shaft resistance was observed, as depicted by the correlation between standard penetration test N values of the surrounding soil and the measured skin friction along the pile. The contribution of the measured toe resistance over the total applied load varied from 2.3% to 36.7%, depending on the geotechnical properties of the foundation layers. The field tests clearly showed that composite spun piles had high capacities and could be used as a reliable alternative to traditional driven as well as bored piles in current practice.

Single piles under horizontal loads in sand: determination of P–Y curves from the prebored pressuremeter test

Lateral load-deflection behaviour of single piles is often analysed in practice on the basis of methods of load-transfer P–Y curves. The paper is aimed at presenting the results of the interpretation of five full-scale horizontal loading tests of single instrumented piles in two sandy soils, in order to define the parameters of P–Y curves, namely the initial lateral reaction modulus and the lateral soil resistance, in correlation with the pressuremeter test parameters. P–Y curve parameters were found varying as a power of lateral pile/soil stiffness, on the basis of which hyperbolic P–Y curves in sand were proposed. The predictive capabilities of the proposed P–Y curves were assessed by predicting the soil/pile response in full-scale tests as well as in centrifuge tests and a very good agreement was found between the computed deflections and bending moments, and the measured ones. Small-sized database of full-scale pile loading tests in sand was built and a comparative study of some commonly used P–Y curve methods was undertaken. Moreover, it was shown that the load-deflection curves of these test piles may be normalised in a practical form for an approximate evaluation of pile deflection in a preliminary stage of pile design. At last, a parametric study undertaken on the basis of the proposed P–Y curves showed the significant influence of the lateral pile/soil stiffness on the non-linear load-deflection response.

Skin Friction of Non-Displacement Piles Related to Simple Shear Mode with Large Strain State Friction Angle

A rational method for evaluating the skin friction of non-displacement piles in sandy soils is presented by reconsidering the model which has already been proposed by the authors. The revision of the model is mainly made by assuming the soil-pile interaction mechanism as a simple direct shear mode in relation to the large strain state of soils. The model is characterized by a coefficient of effective horizontal stress averaged through the whole length of the pile derived as a function of a large strain state friction angle. The characteristics and ability of the model are shown by the parametric studies and the applicability of the model is verified through a database of full scale pile load tests.

Dynamic response of laterally loaded piles in clay

The behaviour of single piles under lateral dynamic loading is critical, and has been an important field of research since the 1950s. Many analytical or semi-analytical linear and non-linear models are available to estimate the dynamic lateral stiffness, but it is essential to determine the dynamic characteristics of the soil–pile system through full-scale lateral dynamic pile load tests for important structures and for validation of existing models. This paper presents the results of field lateral dynamic load tests conducted on 33 piles of varying types–driven precast concrete, driven cast-in-situ concrete and bored cast-in-situ concrete–at different petrochemical complexes in India. The results indicate that driven precast concrete piles have stiffnesses that are four to five times higher than those of driven cast in situ piles. The lateral stiffness was also estimated using the computer program PILAY for all piles and compared with the stiffness determined from the field tests. The estimated stiffness shows good agreement with the field values for stiff clay sites, but greatly overestimates the values for soft clay sites.

Nonlinear Response of Single Piles in Sand Subjected to Lateral Loads using khmax Approach

This paper presents results of analysis of full-scale pile load test data of 14 piles embedded in either loose or medium dense sands. The analysis was performed using two methods, p–y curve approach and a more recently developed khmax approach. Comparison of the results obtained using both the methods is also presented. A step-by-step analysis procedure is presented for predicting lateral load deflection response of single piles in sand using the khmax approach. The results presented show that the khmax approach has promise over the p–y curve approach because of its simplicity and the fact that it provides upper- and lower-bound curves, which are valuable guides to making engineering decisions. For loose sands, a new range of khmax values is recommended to better predict the lateral load–deflection response of single piles.

Determination of Bearing Capacity of Open-Ended Piles in Sand

The bearing capacity of open-ended piles is affected by the degree of soil plugging, which is quantified by the incremental filling ratio (IFR). There is not at present a design criterion for open-ended piles that explicitly considers the effect of IFR on pile load capacity. In order to investigate this effect, model pile load tests were conducted on instrumented open-ended piles using a calibration chamber. The results of these tests show that the IFR increases with increasing relative density and increasing horizontal stress. It can also be seen that the IFR increases linearly with the plug length ratio (PLR) and can be estimated from the PLR. The unit base and shaft resistances increase with decreasing IFR. Based on the results of the model pile tests, new empirical relations for plug load capacity, annulus load capacity, and shaft load capacity of open-ended piles are proposed. The proposed relations are applied to a full-scale pile load test performed by the authors. In this load test, the pile was fully instrumented, and the IFR was continuously measured during pile driving. A comparison between predicted and measured load capacities shows that the recommended relations produce satisfactory predictions.

Pore Pressure Response of Liquefied Sand in Full-Scale Lateral Pile Load Tests

Several full-scale lateral pile load tests were performed at Treasure Island in the San Francisco Bay in sand liquefied by controlled blasting. Cyclic lateral loads were applied to the foundations with displacements up to 225 mm. Piezometers were installed adjacent to, and at a distance up to 4.2 m in front of, the deep foundations. Excess pore pressure ratios observed during testing are presented for the purpose of providing insight into the behavior of liquefied soil in response to laterally loaded piles. Observed excess pore pressure ratios show that soil in the loading path of the foundations experienced a phase transformation at distances as great as 4.2 m. The soil behavior in response to lateral foundation movement is presented based on interpretation of the measured excess pore pressure response.

Permanent Strains of Piles in Sand due to Cyclic Lateral Loads

The strain superposition concept, proposed for ballast study, is applied here to evaluate strain accumulation for laterally loaded piles in sand. It is shown that the soil properties, types of pile installation, cyclic loading types, pile embedded length, and pile/soil relative stiffness ratio are important factors that influence the pile behavior under mixed lateral loads. These factors are quantified by means of a degradation factor, t, which is derived from the results of 20 full-scale pile load tests and then verified using 6 additional full-scale pile load tests.

SALLOP: Simple Approach for Lateral Loads on Piles

The problem of a pile subjected to lateral loading is often solved by assuming that the pile is an elastic member and that the soil can be represented by a series of nonlinear horizontal springs. The P-y curves describe the nonlinear behavior of the soil springs. Previously, a method was developed at Texas AM it makes use of the pressuremeter limit pressure and the pressuremeter modulus. It is a semitheoretical or semiempirical method in that the framework is theoretical but the factors in the theoretical equations are adjusted by comparison to 20 full-scale load tests. The accuracy of SALLOP is very reasonable as shown by comparing measured and predicted behavior for the 20 full-scale pile load tests used to develop the method.

Lateral pile response to monotonie pile head loading

This paper presents results from a series of model tests of single vertical piles subjected to lateral monotonic pile head loading. Model tests were carried out in sand under a simulated field stress condition using the hydraulic gradient similitude technique. Studies were focused on examining various factors that affect the soil-pile interaction in terms of P–y curves. It was found that the P–y curves are highly nonlinear and stress-level dependent but are insensitive to the pile head loading conditions. The P–y curves at depths below one pile diameter were found to be normalized by the maximum soil Young's modulus Emax and the pile diameter. Comparison was made between the experimental P–y curves and those recommended by the American Petroleum Institute (API). It was found that the experimental P–y curves were significantly different from the API P–y curves. New procedures for constructing P–y curves are proposed and verified by numerical analyses of the observed model pile response. The results indicate that the API P–y curves based on a hyperbolic tangent function tend to overestimate the pile head stiffness, especially for fixed-head piles due to their stiffer shape in the small to medium pile deflection range. The proposed parabolic P–y curves can better resemble experimental P–y curves and consequently give a better prediction of pile response for both free- and fixed-head conditions. Finally, the proposed procedures were applied to analyze a full-scale pile load test, and good agreement was found between the predictions using the proposed parabolic P–y curves and the field test data. Key words : lateral loads, piles, model tests, sands.

Drilled pile behaviour in granular deposits

The results of full-scale pile load tests, field monitoring, and laboratory tests on undisturbed soil samples have been used to compare soil parameters determined from in situ tests and in the laboratory for pile design in granular deposits. These results indicate a close agreement between the field- and laboratory-determined rebound or recompression index of oil sand. Also, a good agreement is indicated between field-measured and calculated pile head settlements. However, the measured average skin friction along the pile shaft in sandy till is about 20% more than the calculated values obtained from empirical and finite element analytical work using laboratory- and field-determined soil parameters. Theoretical t–z curves provided a good approximation of the field behaviour for the pile shaft resistance. Key words: drilled piles, belled piles, oil sand, skin friction, rebound, settlement, load test.