Mobile jack-up rigs are widely used in offshore oil and gas exploration and increasingly in temporary production and maintenance work. There is a steadily increasing demand for their use in deeper water and harsher environments. A typical modem jack-up has three independent legs, each equipped with a footing known as ‘spudcan’. This thesis is concerned with the performance of spudcan foundation subjected to vertical loading correspondent to preloading during its installation into uniform clay. The chief aim of this study is to investigate the bearing behaviour with the corresponding soil failure mechanisms during spudcan penetration. Centrifuge model test and Finite Element (FE) analysis are carried out extensively. In centrifuge modelling, a half-spudcan model and a full spudcan model are used. In the half- spudcan model test, a novel system for revealing soil failure mechanisms and measuring soil deformation has been adopted, in which the half-spudcan model is placed against a transparent window and a subsequent Particle Image Velocimetry (PIV) analysis is performed. The full-spudcan model test is conducted to measure the load-penetration response. In numerical simulation, both small strain and large deformation analyses are carried out with smooth and rough soil-spudcan interfaces considered. At the initial stage of penetration, it is observed that a cavity is formed above the spudcan as it is penetrating into a uniform clay. Meanwhile, soil flows towards the surface and thus soil heave forms close to the spudcan shoulders. With further penetration, the soil underneath the spudcan starts to flow back into the cavity on the exposed top of the spudcan. This backflow causes the spudcan to be embedded while the initially formed cavity remains open. / Eventually, the spudcan becomes fully embedded and the soil flow mechanism reaches a fully localised failure mechanism with deep embedment. The lateral extent of visible distortion due to soil flow is confined well within 1.5-1.6 D (D: spudcan diameter). From both centrifuge and numerical investigations, it is found that in uniform clay, it is inevitable to form a cavity above the spudcan foundation. Thus, the stable cavity depth and soil back flow mechanisms are studied. It is clear that the back flow is caused by a Flow Failure, where it is due to the downward penetration of the spudcan. This is contrary to the Wall Failure that is the mechanism recommended by the current offshore design guidelines to estimate the stable cavity depth. In wall failure, the soil back flow is due to the cavity wall too high to stand. The stable cavity depth is estimated up to 4 times higher by the wall failure mechanism than the one by the flow failure. This explains that the wall failure is never observed in model test. Therefore, a new design chart with design formula is developed for design engineers in the stable cavity depth calculation. The spudcan bearing response is strongly correspondent with the variation of soil failure mechanisms during penetration. At the initial stage of the penetration, the spudcan bearing capacity increases with penetration, which is due to the increase of overburden pressure from cavity formation. At the second stage of the penetration, soil back flow embeds the spudcan, and the spudcan bearing capacity is increasing as the soil flow mechanism transits from its shallow failure mechanism to its deep failure mechanism. / At the final stage of the penetration, the spudcan bearing capacity reaches its ultimate value, where the deep/localised failure mechanism remains. A rough spudcan shows 14 % higher bearing capacity than a smooth spudcan. And a flat-plate shows 8 % higher capacity than a spudcan with a same surface roughness. The ultimate bearing capacity factor N, = 10.5 in uniform soil is recommended as a conservative value when the deep failure mechanism is reached. A correspondent N, = 10.1 in NC clay is suggested for a deeply embedded spudcan.

[Truncated abstract] Three-legged mobile jack-up rigs supported on spudcan foundations are used to perform most offshore drilling in shallow to moderate water depths, and are now capable of operating in water depths up to 130 m. With the gradual move towards heavier rigs in deeper water, and continuing high accident rates during preloading of the spudcan foundations, appraisal of the performance and safety of jack-up rigs has become increasingly important. A crucial aspect of this is to improve understanding of the mechanisms of soil flow around spudcan foundations undergoing continuous large penetration, and to provide accurate estimates of spudcan penetration resistance, avoiding excessive conservatism. Spudcan foundations undergo progressive penetration during preloading, contrasting with onshore practice where a footing is placed at the base of a pre-excavated hole or trench. However, spudcan penetration is generally assessed within the framework used for onshore foundations, considering the bearing resistance of spudcans pre-placed at different depths within the soil profile. The lack of accurate design approaches that take proper account of the nature of spudcan continuous penetration, which is particularly important in layered soil profiles, is an important factor in the high rate of accidents. ... It was found that when a spudcan penetrated into single layer clay, there were three distinct penetration mechanisms: during initial penetration, soil flow extended upwards to the surface leading to surface heave and formation of a cavity above the spudcan; with further penetration, soil began to flow back gradually onto the top of the spudcan; during deep penetration, soil back-flow continued to occur while the initial cavity remained unchanged. For spudcan penetration in stiff-over-soft clay, four interesting aspects of the soil flow mechanisms were identified: (a) vertically downward motion of the soil and consequent deformation of the layer interface; (b) trapping of the stronger material beneath the spudcan, with this material being carried down into the underlying soft layer; (c) delayed back-flow of soil around the spudcan into the cavity formed above the spudcan; (d) eventual localised flow around the embedded spudcan, surrounded by strong soil. At some stage during continuous spudcan penetration, the soil starts to flow back into the cavity above the spudcan. The resulting back-flow provides a seal above the penetrating spudcan and limits the cavity depth. It was shown that the current offshore design guidelines are based on the wrong criterion for when back-flow occurs. New design charts with robust expressions were developed to estimate the point of back-flow and hence the cavity depth above the installed spudcan. Load-penetration responses were presented in terms of normalised soil properties and geometry factors for both single layer and two-layer clay profiles, taking full account of the observed flow mechanisms. Further, guidelines were suggested to evaluate the likelihood and severity of spudcan punch-through failure in layered clays. Finally, the effect of strain-rate and strain-softening was examined, in an attempt to model real soil behaviour more closely. Adjustment factors were proposed to modify the design approaches developed on the basis of ideal elastic-perfectly plastic soil behaviour.