Abstract
[Truncated] Mobile offshore jack-up drilling rigs are self-elevating mobile units deployed to drill oil and gas fields in water depths up to ~ 120 m. A typical jack-up unit consists of a buoyant platform and three independent retractable truss-work legs each resting on a footing. These footings are commonly called spudcans and are generally circular or polygonal in shape with diameters between 10 to 20 m and a conical underside. Jack-ups are installed by using their own weight and that of seawater placed in their ballast to push the spudcan footings into the seabed. Installation in seabeds of sand overlying clay can be difficult if the applied load exceeds the capacity of the top sand layer: the spudcan may plunge into the underlying clay layer with sudden and large uncontrolled displacement. This may lead to buckling of the leg or even toppling of the unit. Such incident of jack-up units is referred to as punch-through. This has significant impact on the safety of structures and personnel as well as cost. This study aims to enhance the understanding of the spudcan and conical footing penetration behaviour in sand overlying clay through centrifuge tests and numerical analyses, and to develop an analytical model that predicts the full load-penetration response.
An extensive series of tests have been performed in the drum centrifuge at the University of Western Australia (UWA) to provide evidence for generalisation of the full profile prediction method. These tests are composed of 15 full model tests and 11 half model tests using the Particle Image Velocimetry technique. The full model tests provided the evidence to adjust and improve an existing failure-stress-dependent model to predict the peak penetration resistance, qpeak, before a spudcan pushes a frustrum of sand through into an underlying clay layer. The effect of embedment depth was incorporated in the modified failure-stress-dependent model developed. Further, the model was extended for use in medium dense and very dense sands by a more comprehensive recalibration. The half model tests highlighted the effect of footing shape on the peak resistance and resistance in the underlying clay and were used to observe the failure mechanisms at different penetration depths.
An extensive series of tests have been performed in the drum centrifuge at the University of Western Australia (UWA) to provide evidence for generalisation of the full profile prediction method. These tests are composed of 15 full model tests and 11 half model tests using the Particle Image Velocimetry technique. The full model tests provided the evidence to adjust and improve an existing failure-stress-dependent model to predict the peak penetration resistance, qpeak, before a spudcan pushes a frustrum of sand through into an underlying clay layer. The effect of embedment depth was incorporated in the modified failure-stress-dependent model developed. Further, the model was extended for use in medium dense and very dense sands by a more comprehensive recalibration. The half model tests highlighted the effect of footing shape on the peak resistance and resistance in the underlying clay and were used to observe the failure mechanisms at different penetration depths.
| Original language | English |
|---|---|
| Qualification | Doctor of Philosophy |
| Supervisors/Advisors |
|
| Publication status | Unpublished - Mar 2015 |