Etude numerique de l'essai au cone effile instrumente dans les argiles Champlain.

摘要

A new truncated and slightly tapered probe called instrumented scarp cone (ISC) is studied in this thesis. The instrumented sharp cone test is performed by pushing continually this probe at a constant rate of 2 cm/s in a pre-bored pilot hole. When the cone is inserted, several measures such as pore pressure, total lateral stress, and driving force can be registered. An appropriate method allows determination of undrained shear strength ( Su).;Axisymmetric simulations were performed using the finite element method in order to model the process of deep penetration (PPP) of the ISC. Two prototypes of the instrumented sharp cone have been studied in this thesis: the sharp cone number 1 (ISC-1) and the sharp cone number 2 (ISC-2). The studied site is that of Mascouche, which is a clay deposit representing the stiff natural clays of the Champlain Sea.;It was assumed that the test has the advantage of combining the pressuremeter test (i.e., performed with a self-boring pressuremeter and noted later SBPM) to that of piezocone as the continued penetration of this cone in a cohesive soil almost produces a pressuremetric expansion. The relevance of this hypothesis has been the subject of this study.;In axisymmetric conditions, parametric analyses were conducted in order to determine the influence of the adhesion on the expected results of the tests performed with the sharp cone number 1 (ISCT-1) and the sharp cone number 2 (ISCT-2). The undrained soil is assumed to behave as an elastic-perfectly plastic material. For a weighty and weightless soil, the contact pressures (CPRESS) versus depth of penetration (D) curves of the cones (1 and 2) are predicted. The quasi-pressurmetric curves (i.e. ln(radial pressure) versus volumetric strain) and the driving force versus D curves were determined for several values of adhesion.;On the basis of the ISCT-1’s simulations results reported in this thesis, the following main conclusions are drawn: The fourth sensor Cap-4 is poorly implemented in the instrumented sharp cone number 1 because: i) it is sensitive to change in adhesion and ii) from a certain depth D, the CPRESS predicted by Cap-4 could become lower than the CPRESS predicted by its nearby sensor Cap-3 that has a volumetric strain lower than Cap-4’s volumetric strain, which is contradictory to the theory of expansion of the cylindrical cavity.;The effect of the variation of Young’s modulus (E) on predicted contact pressures was identified: increasing the value of E will cause an increase in the confinement of the cone and in the contact pressures directly applied to the cone.;Simulations of ISCT-2 permitted to predict contact pressures between the cone number 2 and the surrounding soil in terms of adhesion. It was found that: i) The contact pressure is independent of adhesion. ii) The sensors 1, 2, and 3 are located within an equal pressure area, and thus must be replaced by a single sensor, and iii) The overlapping of curves CPRESS-D disappears. This anomaly that existed with the cone-1, has been corrected here because of increasing the cone slenderness and the distance between sensor 5 and the downstream end of the shaft of the cone.;The study of the effect of the rate of penetration of the ISC-2 was carried out by adopting a rate-dependent material which was used to simulate the process of penetration of the ISC-2 in an undrained soil. This effect was investigated by imposing to the cone different values of constant penetration speed. It was found that the penetration speed of 2 cm/s at with the ISCT-2 test is performed appears to be adequate. As a consequence, penetration speeds in excess of 2 cm/s may overestimate the contact pressures, that’s why penetration speeds greater than 2 cm/s are not recommended. (Abstract shortened by UMI.).