Abstract

This paper presents a combined time-frequency domain procedure for estimation of SCR fatigue damage caused by heave-induced intermittent Vortex-Induced Vibration (VIV). The procedure is implemented in a computersimulation which combines time domain motions with frequency domaincalculations of VIV damage. It has been calibrated to fatigue damage dataacquired from Phase 1 of the PMB Highly Compliant Rigid Riser JIP. Simulationshave been performed for a number of host platforms in various environments. Thefatigue calculation procedure, as well as the results for these simulations arepresented. Results show that significant damage due to this effect can beexpected for certain combinations of floating system type and environment.

Introduction

A series of large scale model tests were conducted in 1998 at Lake Pond Oreille in Idaho as part of the PMB Highly Compliant Riser JIP, as documentedin [1]. These tests were originally designed to determine the onset of bucklingin Steel Catenary (SCR), Lazy Wave (LWCR), and Catenary Vertical Access Risers(CVAR). The testing consisted of forcing a sinusoidal displacement at the topend of the riser and recording the section forces and accelerations at a numberof instrumented "pup" joints along the riser, mostly located in the sag-bendareas.

Although not previously expected, and certainly not accounted for in present SCR design methodologies, a great deal of out-of-plane Vortex-Induced Vibrationcan be seen in the displacement and bending stress time series. Although it isnot clear that these bending moments can be directly scaled to a full scaleriser, there is every indication that these out-of-plane stresses could governthe design of these risers.

The nature of the observed VIV displacements is quite chaotic and a regularsinusoidal displacement of the top of the riser induces non-repeatingout-of-plane displacements. In general, response is reasonably narrow-bandedwithin one half-cycle of top excitation, but there can be large differences inthe frequency of the oscillation from one half-cycle to the next. Analysis ofthe data did not reveal any pattern for the frequency jumps.

This content is only available via PDF.
You can access this article if you purchase or spend a download.