There have been a number recent improvements for the analysis of self-elevating units during the transition phase commonly referred to as the "going on location" or "emplacement" phase (i.e., the transition from the afloat condition to the pinning of the legs). Unlike the well-established guidelines from Class or ISO for units in the elevated condition, the guidelines for going on location analyses have yet to be fully developed, let alone become standard. State-of-the-art analysis methods to establish adequate limits for going on location include nonlinear time-domain response analysis of the units, accounting for jacking speed and spud can shape as well as wave directionality, water depth and soil conditions. The majority of analyses represent the seabed as a nonlinear spring, attempting to match the penetration curves which account for spud can shape and soil properties.
There are two significant effects that have not been addressed by these methodologies: the fact the seabed profile changes after each leg impact, and the energy dissipation properties of the soil. In this paper, the Combined Eulerian-Lagrangian (CEL) capabilities of Abaqus are used to determine both vertical and lateral soil resistance to leg motions of a representative jack-up while going on location. The CEL results are compared with the forces from idealized representations of the soil-structure interaction in the nonlinear response simulation.
A single wave period is considered. Two general soil conditions are considered: a stiff soil and an intermediate soil. The forces from the CEL analysis are then used to replace the elastic seabed on the nonlinear response analysis to see if the jack-up motions are affected.
The results of this study show that the impact loads for jack-ups going on location based on elastic representations of the seabed are over-predicted, and as such they can be considered to be on the conservative side when used in the determination of permissible wave heights.