Assessing the potential for a punch-through failure during spudcan installation in sand overlying clay is crucial for reducing risk in the operations of offshore mobile jack-up platforms. A basic indicator of the severity of the punch-through event is the punch-through distance, which is defined as the vertical penetration from the peak punch-through load at the onset of failure until vertical force equilibrium is re-established in the underlying clay. Current methods used to calculate punch-through distance are deterministic. However, with the uncertainties in soil parameters (e.g. soil layers and soil strength), which cannot be modelled and processed explicitly in the deterministic models, there is a need to provide probabilistic methods to calculate the range of punch-through distance. This paper presents a probabilistic approach to estimating the punch-through distance in sand-over-clay soils. The uncertainties in the soil-spudcan system are characterized using a database of both experimental tests and numerical simulations through the selection of suitable probability distributions and propagated to the deterministic model of punch-through distance. Example probabilistic calculations illustrate how the proposed method can be used to estimate the possible range and distribution of the punch-through distance. Retrospective simulation of geotechnical centrifuge experiments verifies its suitability. The proposed approach provides additional information to offshore operators on the potential and severity of any punch-through event. This should be useful in deciding on what operation procedures to adopt during a jack-up spudcan installation.
展开▼
