Research Challenges in the Vortex-Induced Vibration Prediction of Marine Risers

Abstract

Abstract Current research in the prediction of vortex-induced vibration of marine risers is driven by problems in several areas. These include understanding the physics of the fluid-structure interaction, developing adequate structural dynamic modeling tools, acquiring high quality full-scale response data, finding appropriate techniques for analysis of response data, calibrating response prediction programs, and inventing cost effective suppression and protection methodologies. Problems in each of these areas are defined and key limiting issues are described. Introduction Exploration drilling and production in many new locations around the world are being confronted by significant current hazards, Challenges posed by greater water depths, more hostile conditions and greater field development costs are resulting in many new design concepts, such as catenary risers, slim risers and novel vibration protection strategies. In some cases our ability to predict response amplitudes, stress levels and fatigue damage rates of risers due to vortex-induced vibration (VIV) is not adequate. This paper addresses five of the problem areas confronting us. They are (l) the understanding of the fluid mechanics of the fluid-structure interaction, (2) the development of better structural dynamicmodeling techniques, (3) the acquisition of high quality, fullscale, riser response data for the purpose of calibration of response prediction models, (4) the analysis of multi-channel field data for comparison to response prediction models, and (5) the development of VIV suppression and protection technologies. No doubt there are many other important issues, but these five are highlighted here. Fluid-Structure Interaction There are many issues involving the hydrodynamics of VIV that we do not fully understand. Those that are the most important to the fatigue life prediction of risers have to do with our ability to model the lift force and damping on the cylinder as a function of the local fluid velocity and the cylinders own motion. It is a non-linear interaction that is sensitive to Reynolds number, roughness, velocity, and turbulence, as well as the interaction between the amplitude and frequency content of the cylinder motion. There are many aspects that are not well understood. One is examined here by means of an hypothetical example. A uniform straight riser as shown in Figure 1. It is exposed to a current profile consisting of two regions, each occupying one-half of the length of the riser. The upper region has velocity U1 and the lower region has a lesser velocity U2. The hypothetical problem is to predict the response of this riser to the lift forces that result from the vortex shedding from these two regions. To further simplify the problem, assume that the Strouhal number is the same in both regions and is estimated to be 0.2. This allows us to say that the expected frequency of the lift force in each region is given by the simple relationship,(mathematical equation)(available in full paper) Further assume that this riser has natural frequencies that are coincident with the expected lift force frequency in each of the two zones of the riser. A completely linear solution would say that the total response of the riser would be the sum of the responses to the excitation in each of the two different flow regions.