No abstract

No abstract

The overall bearing capacity of a jack-up rig under horizontal load is conducted using finite element models that consider the deck–foundation–soil interaction. In these models, the simplified horizontal load acts on the deck and increases until the platform loses its stability. The effects of the self-weight of the platform W and load direction α on the ultimate horizontal bearing capacity Hult are investigated, and W- Hult failure envelopes under different α conditions are obtained. Two typical seabed types, including the double-layer seabed of sand overlying soft clay and the single-layer seabed of sand, are considered. The results show that a critical self-weight Wcritical exists in the double-layer seabed. Based on Wcritical, the failure of the platform presents two different modes. When W <  Wcritical, the windward leg is pulled up, and Hult increases with the increase in W. When W >  Wcritical, the failure mode is the leeward leg or legs puncturing the bearing sand layer, and Hult decreases with the increase in W. In the single-layer seabed, the failure mode is the windward leg being pulled up, and Hult increases with the increase in W throughout the whole range. The W- Hult envelopes in these two types of seabeds are basically the same when W <  Wcritical.

No abstract

Aiming at simulating the preloading process of jack-up rig at drilling locations with typical stratum combination near the Yellow River estuary, FE models of interaction between spudcan and sub-soil are set up based on a large deformation finite element method known as Remeshing and Interpolation Technique with Small Strain (RITSS). The ABAQUS finite element package is used to calculate small strain solutions following each step. The seabed can be simplified as a three-layer deposits with soft clay layer, sand layer and soft clay layer from surface to bottom according to geological data. The backfill phenomenon during preloading is described. The results show that the surface soft clay falls onto the upper surface of spudcan very soon after the preloading process begins, afterwards the backfill tendency occurs in the middle sand layer. SNAME recommends two sets of formulas for calculating penetration resistance respectively for backfill penetration and penetration with no backfill. In this study the model with no backfill is realized by condition control in program codes and the penetration resistance is compared with the results of practical backfill model. The variance proves the shortcomings of the method given by SNAME.