<p>In the last two decades, there has been significant improvements in pile design methods. These methods include variables that have been studied using laboratory and full-scale experiments. Refined understanding of the underlying mechanisms controlling pile response to loading enables introduction of variables in the design equations that reflect observations made in high-quality experimental data.</p><p>The mechanisms involved in the mobilization of the pile resistance (both base and shaft resistance) are studied in this thesis using a large-scale model pile testing facility consisting of a half-cylindrical calibration chamber with image analysis capabilities, instrumented model piles, and data and digital image acquisition systems. The thesis focuses on the effect of the pile surface roughness on the mobilization of tensile shaft resistance, the effect of the pile base geometry on the mobilization of base resistance, the analysis of possible mechanisms responsible for time-dependent increases in pile axial capacity, and particle crushing produced by pile installation. </p><p>A set of model pile tests were performed to study the effects of three different surface roughnesses on the shaft resistance of model piles jacked in the half-cylindrical calibration chamber. Digital images of the model piles and surrounding sand captured during tensile static loading were analyzed using the digital image correlation (DIC) technique. The base and shaft resistance measured for the instrumented model piles and the displacement and strain fields obtained with the DIC technique show that an increase in the roughness of the pile shaft results in an increase in the average unit shaft resistance and in the displacements and strains in the sand next to the shaft of the pile. Guidance is provided for consideration of pile shaft surface roughness in the calculation of the tensile limit unit shaft resistance of jacked piles in sand.</p><p>Base geometry effects were studied using jacked and pre-installed model piles with flat and conical bases tested in the DIC calibration chamber. The results show that the mobilized base resistance of a model pile with a conical tip was less than that of an equal pile with a flat base, all other things being equal, by a factor ranging from 0.64 to 0.84. The displacement and strain fields obtained with DIC also show that the slip pattern below the pile with a conical base differs from that of a pile with a flat base. </p><p>Finally, the degree of crushing of silica sand particles below the base of model piles jacked in sand samples is studied. The particle size distribution curves are obtained before and after pile installation. Relationships between the load mobilized at the base of the model piles and two well-known breakage parameters are proposed. This work also provides detailed measurements of the trajectories followed by crushed and uncrushed particles during pile installation, and characterizes the typical particle crushing modes produced by piles jacked in silica sand.</p><div><br></div>

Ce travail de thèse porte sur l’étude du comportement de l’interface sol-pieu sous sollicitations cycliques dans les sols fins. Dans ce cadre, un important programme expérimental a été réalisé en chambre d’étalonnage, à partir d’une sonde-pieu et d’un piézocône testés dans des massifs d’argile saturée. Pour réaliser cette étude, on a, en particulier, développé un consolidomètre prototype pour la reconstitution des massifs d’argile. On s’est intéressé plus particulièrement, lors de chargements cycliques réalisés à déplacement contrôlé, aux évolutions du frottement local mobilisé à l’interface sol-pieu et de la résistance en pointe. L’étude expérimentale a permis de mettre en évidence l’influence des paramètres clés sur le comportement observé et, plus spécifiquement, sur l’évolution des propriétés de frottement d’interface. On s’est intéressé, en particulier, à l’influence de l’amplitude du chargement cyclique, de la fréquence des sollicitations, ainsi que du nombre de cycles appliqués. L’accent a été mis sur le cas des grands nombres de cycles, typiquement quelques centaines de milliers, encore peu étudié dans la littérature. Il faut également ajouter l’étude de l’influence du niveau des contraintes initiales appliquées au massif. Les résultats expérimentaux ont permis de mettre en évidence, après une phase de dégradation initiale, une phase de renforcement du frottement local. Ce type de comportement n’a pas encore été décrit dans la littérature. La phase de dégradation est attribuée à l’augmentation progressive de la surpression interstitielle à l’interface au cours des cycles, tandis que la phase de renforcement est attribuée à la dissipation progressive de la surpression interstitielle. Par ailleurs, un essai de chargement cyclique sur un piézocône a été réalisé afin de quantifier précisément la variation des surpressions interstitielles à l’interface sol-pieu et dans le massif, lors des différentes phases de chargement, et ainsi valider les interprétations faites concernant les phases de dégradation et de renforcement. / This Ph.D. dissertation focuses on the behavior of the pile-soil interface under cyclic loading. In this context, an experimental program was conducted on the Navier calibration chamber using an instrumented pile-probe and a piezocone installed in saturated clay samples. In order to carry out this study, we developed, In particular a slurry consolidomètre prototype to reconstitute fully saturated clay samples. Special attention was given, during displacement controlled cyclic tests, to local friction evolution mobilized at the pile-soil interface and mobilized tip résistance. The experimental study has allowed the demonstration of the influence of the key parameters on the observed behavior and more specifically, on the local friction at the interface. We examined, in particular, the influence of amplitude of cyclic displacement, the frequency of cyclic loading and the applied number of cycles. Emphasis was put on the case of large number of a few hundred thousand of cycles that is still little studied in the literature. We also study the influence of initial consolidation pressure. The experimental results allowed us to identify, after an initial phase of stress softening, a new phase of stress hardening of local friction. This behavior is not yet described in the literature. The stress softening phase is related to the progressive generation of pore water pressure while the stress hardening, for its part, is related to a gradual dissipation of the generated pore water pressure. Furthermore, a complete sequence of monotonic and cyclic tests were conducted on a piezocone in order to quantify, in a precise manner, the generated pore water pressure at the pile-soil interface during different loading phases and thus validate interpretations given for stress-softening and stress-hardening phases based on the observed results.

Ce travail de thèse porte sur l'étude du comportement de l'interface sol-structure sous sollicitations cycliques. Pour cela, un important programme expérimental à la sonde-pieu, mise en place dans des massifs de sable siliceux en chambre d'étalonnage, a été réalisé. On s'est intéressé, particulièrement, à l'évolution du frottement latéral et de la résistance enpointe à grand nombre de cycles (100 000 cycles), en faisant varier l'état initial du massif (état de densité et contrainte de consolidation) et les paramètres de chargement (amplitude du déplacement cyclique). Lors des essais à déplacement contrôlé, un renforcement important du frottement latéral à grand nombre de cycles a été observé pour les faibles valeurs de résistance en pointe initiale. Ce renforcement n'a quasiment pas été observé avant par les auteurs, car l'accent a été mis sur la phase de dégradation (autour de 1000cycles). Il est attribué à une forte dilatance partiellement empêchée. En revanche, pour les fortes valeurs de résistance en pointe, le renforcement à grand nombre de cycles diminue considérablement. Cette diminution est liée à l'effet des particules fines créées lors du fonçage. Celles-ci jouent le rôle de cimentation/scellement de la surface latérale de la sonde pieu en diminuant sa rugosité. Par ailleurs, des essais à force contrôlée ont été réalisés afin d'étudier la stabilité des pieux. Enfin, des lois empiriques d'évolution du frottement latéral et de la résistance en pointe ont été proposées afin de reproduire les évolutions observées expérimentalement. Ces lois d'évolution ont été intégrées dans un modèle de calcul de pieu sous chargement cyclique de type t-z. Les premières simulations effectuées montrent un bon accord entre le modèle et les résultats expérimentaux à petit nombre de cycles / We study in this present work the behavior of the soil-structure interface under large number of cycles (100 000 cycles). An important program with a probe-pile jacked into the beds of sand was carried out in calibration chamber. We interested particularly on the evolution of local skin friction and tip resistance at different initial state of beds (initial density and confining pressure) and the parameters of the load (cyclic displacement amplitude). We conducted an important part of the experimental work under axial cyclic displacement controlled test which allows studying the large number of cycles. After the degradation phase, already observed by several authors (about 1000 cycles), an important re-increase in the skin friction at large cyclic number was observed at low values of initial tip resistance.This re-increase is attributed to a high dilation of sand around the probe. However, for high values of initial tip resistance there is almost not re-increase in the skin friction. This is due tothe formation of a shear band around the probe made of crushed sand. These reduce theroughness of the probe by cementing/sealing. In addition to the displacement-controlled tests, we conducted the load-controlled tests in order to study the stability of the pile. The results showed a good agreement with the displacement-controlled tests. Finally, empirical model of evolution of skin friction and tip resistance have been proposed in order to reproduce the experimental results. This model was incorporated into a computational model of pile under cyclic axial loading. The first simulations showed a good agreement with the experimental results at low number of cycles

<p> The base and shaft resistance of steel pipe piles installed in silica sand is affected by several factors; these include but are not limited to: shaft resistance degradation, shaft surface roughness, installation method, pile geometry, soil density and particle size, and setup.  This thesis focuses on the first four factors, while also considering the effect of soil density within each factor. Several of the pile design formulas available do not consider the effects of shaft resistance degradation due to load cycles during installation of jacked and driven closed-ended pipe piles, plug formation and evolution during driving of open-ended pipe piles, the degree of corrosion or pitting corrosion on the shaft surface of a pile and its potential impact on setup, and the geometry of the tip of the pile. To assess the impact on pile capacity of some of these factors, a series of static compression load tests were performed in a controlled environment in a calibration chamber with a scaled down instrumented model pile. The air-pluviation technique with different combination of sieves assembled in a large-scale pluviator was used to prepare F-55 sand samples of different density in the calibration chamber. Slight changes were made to the experimental setup to study each factor: sand sample density, driving energy, mode of installation, and geometry and shaft roughness of the model pile.</p> <p><br></p> <p>The results from the experiments confirmed that each of these factors affects the pile resistance. Some of the important conclusions were:</p> <p><br></p> <p>i. The shaft resistance of the model pile is about 2.4 times greater for jacked piles than for driven piles in dense sand, due to the greater shaft resistance degradation in driven piles. </p> <p>ii. Despite the effect of degradation, the shaft resistance of the non-displacement model pile which had no loading cycles was a ratio of 0.37 to that of the driven model pile in medium dense sand and 0.60 in dense sand, due to the absence of displacement.</p> <p>iii. An increase in the surface roughness of the jacked model piles from smooth to medium-rough resulted in an increase of the shaft resistance, which had a ratio of 7.75 to the smooth pile in dense sand and 3.05 in medium dense sand. An increase from smooth to rough resulted in an increase of the shaft resistance, which had a ratio of 8.00 to the smooth pile in dense sand and 4.26 in medium dense sand.</p> <p>iv. Although rougher interfaces produce greater interface friction angles than smooth interfaces with sand, once a limiting value of surface roughness is reached, shearing occurs in a narrow band in the sand in the immediate vicinity of the model pile, with the shaft resistance depending on the critical-state friction angle of the sand. This means the shaft resistance will not increase further with changes in pile surface roughness, due to the fact that the internal critical-state friction angle of the sand has been reached in the shear band during loading.  </p> <p>v. During installation, the conical-based pile had a higher penetration per blow compared to the flat based pile from 0 to 25.6<em>B</em> in medium dense sand and 0 to 20<em>B</em> in dense sand (<em>B</em> = base diameter). After the pile was installed beyond 25.6<em>B</em> in medium dense and 20<em>B</em> in dense sand, the penetration per blow was identical. </p> <p>vi. The base resistance of a conical-based model pile was about 0.76 times that of a flat-based model pile in dense sand and 0.56 in medium dense sand. </p> <p>vii. Jacked piles had similar base resistance ratio of about 0.93 to 0.95 of driven piles in dense sand and 0.98 to 1.05 in medium dense sand. However, they had a much higher shaft resistance ratio of about 1.67 to 2.07 in dense sand and 1.44 to 1.50 in medium dense sand. </p>