Abstract

In June 2003, ExxonMobil performed vortex-induced vibration (VIV) testing on a long flexible cylinder to assess the validity of key theoretical assumptions that underly VIV prediction methods presently used by industry for riser design.

A rotating test rig was used to expose a 9.63-m long test pipe with 20-mm diameter to flow conditions simulating uniform and linearly sheared current. The model was heavily instrumented to acquire records of bending strain and lateral acceleration in both cross-flow and in-line directions at a sufficient number of stations to allow accurate reconstruction of static and dynamic bending deflections along the riser as a function of time. The dense array of sensors permits an unprecedented level of detail in visualizing and interpreting response data.

The data were analyzed to examine the amplitudes and frequency content of the vibrations, extract the wavenumbers of bending variations along the riser, and compute fatigue damage rates due to combined cross-flow and in-line bending at many positions along the riser. Results show that common modeling assumptions reproduce response trends for bare risers with no VIV suppression devices and lead to a reasonable qualitative description of the vibration character.

The data from these experiments provide a benchmark for both qualitative and quantitative validation of present and future riser VIV prediction formulations and prediction codes.

Introduction

Vortex-induced vibration (VIV) is a riser response to timevarying hydrodynamic forces that arise when ocean currents cause vortices to form and shed into the riser's wake. Bending vibrations excited by vortex shedding in high-speed currents can cause significant long-term fatigue damage. Therefore, designers need reliable methods for predicting vibration levels and fatigue damage rates for the range of flow conditions to which risers may be exposed in service. Vibration amplitude, frequency, and wavelength are the critical parameters needed to estimate the fatigue damage rate in a given current condition. Vibration amplitude and wavelength determine the magnitude of cyclic bending stresses in the riser wall. Vibration frequency determines the number of stress cycles encountered over a specified time.

Riser VIV prediction is difficult and complex due to a) the strongly non-linear nature of the viscous hydrodynamic forces associated with vortex shedding, and their interaction with structural response, b) varying current velocity along the span of a riser in ocean currents, and c) the potential for structural response at a number of frequencies, either singly or in combination. Scientifically rigorous modeling of these factors is not possible due to the limitations of present experimental and numerical hydrodynamics technology. For practical riser design, semi-empirical prediction formulations have been developed by experts based on their best understanding of the riser VIV phenomenon. Of necessity, these formulations incorporate modeling assumptions and approximations that need further validation for risers in deep-water, high-speed current environments. Prediction robustness for these challenging applications is an ongoing area of investigation, and the industry continues to conduct extensive proprietary and joint research programs to improve and fully validate the methods.

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