Influence of Local Effects on Fatigue Assessments during Hold Periods of Offshore Pipeline Installation by Reeling

Abstract

There are many situations during offshore installation of pipelines in which the pipe is held clamped for extended periods on the lay vessel. In the case of installation by reeling this is done at the hang off clamp (HOC) below the workstation of the reel-lay vessel. The increased fatigue damage in the pipe over the duration of the hold period, especially at or just below the HOC, then has to be evaluated and included in the overall fatigue assessment. The behaviour of pipelines during offshore installation under various anticipated sea states have long been routinely predicted using established and mostly purpose-written numerical methods. In these assessments the time histories of vessel movement and wave and current action on the immersed pipe are used as input and the dynamic response of the pipeline is evaluated. The higher local fatigue damage at the girth weld nearest the clamp is then evaluated by using a stress concentration factor (SCF) applied to the predicted stress ranges at that location to take into account the local fixity. In most simulations, however, the SCF is assigned an unnecessarily high value based on full fixity at the clamp which usually results in the predicted fatigue damage being significantly higher than that at the sagbend. The allowable hold period for a given sea state could then be governed by the SCF. In reality, many different factors would be at play at the HOC which reduce the magnitude of an SCF based on full fixity. The more important of these factors are the local frictional slip between pipe outer surface and the pads, the thickness dependent higher compliance of outer thermal insulation coating material that may be on the pipe, and the local flexibilities of the HOC housing assembly on the vessel. This paper provides a discussion and semi-numerical evaluation of the effects of some of these factors on the SCF and consequent assessment of fatigue damage.