Investigating the Relevance of Strip-Theory for Pipelines Subjected to Vortex Induced Vibration

Empirical models for prediction of vortex induced vibrations (VIV) apply hydrodynamic coefficients to represent the fluid forces on the structure. The coefficients are found by measuring forces on a rigid cylinder under harmonic pure inline (IL) and pure cross flow (CF) forced motions, and presented as functions of non-dimensional motion amplitude and frequency. In the response analysis the forces at a specific cross-section are assumed to be defined by the motion of this cross-section, which implies that possible three-dimensional (3D) hydrodynamic effects are neglected. The approach is often referred to as ‘strip-theory’, which is a term originally used in ship hydrodynamics. Here, a classical method for calculation of motions and beam forces is based on the same type of simplification. The strip-theory has been used for VIV analysis both in combination with empirical coefficients but also combined with 2D numerical solutions of the Navier-Stokes equation for force calculation. The approach as such has never been verified and the loss of accuracy from neglecting 3D hydrodynamic effects has never been quantified. The purpose of the present work is to contribute to such verification. The investigation reported herein consists of three steps. – Experiments with a flexible beam subjected to VIV. Response amplitudes on CF and IL directions were measured so that the trajectories for several cross-sections along the beam could be found; – Measurement for hydrodynamic forces on a rigid cylinder that was forced to follow the same trajectories as found from the beam experiments; – Use of a finite element program to calculate the dynamic response of a flexible beam with the same properties as for the first test and subjected to forces from the second test. If the calculated response is found to be identified to the measured the verification exercise could be accepted as successful. Discrepancies between measured and calculated response might be caused by 3D hydrodynamic effects or poor quality of the experiments. The present study is a first attempt and the reported results do not lead to a firm conclusion.