Vortex Induced Vibrations Response Evaluation

Abstract

In designing oil and gas pipelines, many aspects should be considered. One of those aspects is fatigue damage due to time varying stresses in the pipe. The stresses in the pipe may be caused by dynamic loads and the most dangerous in the field of pipeline engineering is Vortex-Induced Vibrations (VIV). The VIV might happen when current is passing a free spanning pipeline. A free span pipeline may occur due to seabed unevenness as in most cases the routing of pipelines is crossing uneven seabed.The VIV may occur both in-line with the current velocity and cross flow or perpendicular to the current velocity.Excessive excitations will happen if the frequency of vortex shedding coincides with the natural frequency of the pipeline. However, Vortex-Induced Vibrations are considered as a self-limiting response, where there is a limitation of the response amplitude.During the Ormen Lange Project, Norsk Hydro and Reinertsen Engineering performed model tests (Basin Test) of long free span pipelines in the Ocean Laboratory at Marintek, Trondheim in 2000. The aim of the model tests was to study the effect of multiple mode response caused by VIV for long free spans. By going into the raw data, that was available for the longer spans, it has been shown that the stress range shows a close to a linear dependency on the current velocity.Further evaluations of the stress range as a function of current velocity has been performed. Analysis is carried outby using Reinertsen’s in-house analysis tool i.e. Pipeline Analysis System (PAS) on simulated long free spans with a ratio L/D = 350 for two different cases. Further, comparisons between the Basin Test and PAS results have alsobeen done.From the simulation results, it is found one of the Cross-flow frequencies calculated by PAS does not match with the frequency measured in Solvia (Basin Test). The discrepancy is considered to appear due to the sag and multiplemode excitation effects on a long free spanning pipeline where PAS seems unable to consider those effects. Aninteresting point is that the plots from PAS are showing similarity with the Basin Test i.e. the maximum stressranges have a dependency that is close to a linear variation with the current velocity.For further work, analysis using other programs such as Vivana, Shear7 or alike is recommended. Further analysis on multiple span pipelines is also recommended.