Modelling of the settlement interaction of neighbouring buildings on soft ground.

The research described in this thesis investigated the soil deformation due to the interaction of two neighbouring buildings founded on soft clay. The study involved small-scale Ig physical modelling in which mitigation of the interaction was studied by inserting a model sheet pile wall between two footings. The results were used to validate finite element analyses that were then extended to simulate a prototype of two buildings with raft foundations constructed on Singapore clay. In the physical models, two 100 mm wide strip footings were placed on a 200 mm thick bed of preconsolidated kaolin in a rigid chamber with a transparent front wall. Tests were conducted without a wall between the footings, with a 100 mm long floating wall, and with a vertically restrained (fixed) 100-mm long wall. The footings were loaded in tum and the second footing was only loaded after the consolidation, due to loading on the first footing or subsequent wall insertion, was completed. The soil and footing displacements were photographed through the transparent chamber wall and measured using a combination of Particle Image Velocimetry (PIV) and close range photogrammetry. A stiffer soil response was observed beneath the second footing than beneath the first. Some additional settlement of the first footing was induced by the loading on the second footing and this was not significantly mitigated by a floating wall. In contrast, with a fixed wall, the settlement of the footing was substantially reduced. Overall, the tilting in the tests with a floating wall was comparable to that in the tests without a wall. The tilting of the first footing, though not the second, was effectively mitigated when a fixed wall was used. Plane strain finite element analyses with the BRICK constitutive soil model were perfonned to simulate the physical models. In general, the observed behaviour patterns were well reproduced, although the finite element analyses tended to exaggerate the stiffness of the soil response under the second footing relative to that under the first when a wall was present. The simulated prototype comprised two 10m wide, 800 mm thick concrete rafts separated by 2 m resting on 40 m of clay. The analyses were similar to those conducted for the physical model simulations. In addition to modelling the effects of installing a floating or a fixed wall, parametric studies were undertaken to examine the influence of wall length and stiffness. A longer floating wall reduced the settlement of both footings due to loading on the second footing but even greater settlement mitigation was obtained with a fixed wall. The stiffness of the wall did not have a significant effect on the settlement mitigation except in the case of an unrealistically stiff fixed sheet pile. Without any wall, both footings tilted towards each other. The presence of a wall generally reduced tilting and the reduction increased with wall stiffness Keyword: closely-spaced footings, settlement, tilting, physical modelling, numerical modelling, finite element analysis, particle image velocimetry.