For a steel catenary riser (SCR) in the ultra deep water of the Gulf of Mexico (GOM), the areas of major concern from a dynamic stress response point of view are the sagbend region near the touchdown and the region below the hang-off at the SCR top where local bending must be accommodated. Usually global strength (stress/strain/buckling) analysis focuses primarily on the sagbend region, while more detailed analysis associated with design of the riser hang-off assembly concentrates on the SCR top region. The global strength design of the SCR is controlled by dynamic response in the sagbend region, which is primarily driven by the host vessel motions. Vessel motions are in turn induced by metocean conditions such as hurricanes and winter storms. Given the random nature of the ocean waves, to obtain statistically sound results, random wave simulations involving multiple (usually ten (10)) three (3) hour realizations have been a well accepted practice by the offshore industry. Just as waves in an extreme or survival storm event are randomly distributed, stress (and strain) response events are randomly distributed as well. For a comprehensive design, there will be far more than one sea state to be analyzed with each sea state undergoing multiple three (3) hour simulations. In addition, the design often progresses iteratively, i.e., there will be several cycles of analyses to be performed before the final design can be concluded. Therefore, the overall computational resources in terms of time and data storage are quite significant. This paper presents the methodology that significantly reduces the computer simulation time without compromising the analysis accuracy for the strength analysis of the SCR. The paper uses the example of a SCR in the ultra deepwater of the GOM attached to a DeepDraft Semi™ designed by SBM Atlantia Inc. The methodology builds on time traces of the host vessel motions, and the correlation between the vessel/porch motion and the SCR sagbend response. Generally the maximum riser sagbend stress occurs when the wave pushes the vessel, then the riser porch toward its touchdown point (slack position). One vessel/porch motion characteristic — the downward speed at the riser porch dominates the SCR sagbend response. By screening the downward speeds at the riser porch under slack condition, the time at which the sagbend response (stress/strain/buckling) peaks is identified. A time trace window containing the peak time and with band width of about 200 seconds is located and the SCR global dynamic analysis is performed based on this time trace window. In some scenarios, up to five (5) windows associated with the top five (5) downward speeds at the riser porch for one realization are needed to capture the peak stress response in the SCR sagbend.

This content is only available via PDF.
You do not currently have access to this content.