CLASSICAL AND FINITE DIFFERENCE METHOD TO ESTIMATE PILE CAPACITY COMPARED WITH PILE LOAD TESTING RESULTS

摘要

Three conventional uplift tests and 12 Rapid Pile Load Tests (RLT) were performed on total of fifteen 16-inch square pre-cast concrete driven piles at a test site in Emeryville California in December of 2000. The uplift tests were performed in accordance with ASTM standards, and the RLTs were performed using a FUNDEX RLT. The RLT is carried out by dropping a mass of 25-tons on top of the pile, thereby inducing a force that is measured by a load cell. An optical lens records the vertical deflection of LED's that are mounted on the pile. Ultimate uplift and compression capacities of each of the fifteen piles were determined using the Davisson method. Soil parameters were calibrated by back-calculation (cohesion and internal angle of friction) using both the results of conventional uplift load versus deflection curve, as well as the results of the triaxial compression tests. A two-dimensional finite difference model including soil elements, and structural elements representing the test pile was setup using the computer program Fast Lagrangian Analysis of Continua (FLAC). FLAC was used to simulate both the uplift as well as the RLT. Strength increase of cohesive material was also accounted for in the analysis. Numerically simulated and theoretically computed pile capacity curves and axial deformation variation with applied load curves are in close agreement to the observed curve during the field test. Conclusions are made from this test site regarding the factors influencing results of RLT.