Prototype Reynolds Number VIV Tests on a Full-scale Rigid Riser

Slender offshore structures in deep water subjected to currents may experience vortex-induced vibrations (VIV), which can cause significant fatigue damage. Extensive experimental researches have been conducted to study the VIV in the past several decades. However, most of the experimental works have small-scale models and relatively low Reynolds number (Re) – ‘subcritical’ or even lower Reynolds number regime. There is a lack of full understanding the VIV in prototype Re flow regime. Applying the results with low Re to a full scale riser with prototype Re might have uncertainties due to the scaling effects. In addition, the surface roughness of the riser is also an important parameter, especially in prototype Re regime. In present study, two full-scale rigid riser models with different surface roughness ratios were tested in the towing tank of MARINTEK in 2014. Stationary tests, pure cross-flow (CF) free oscillation tests and forced/controlled motion tests were carried out. Several conclusions could be made: • The drag coefficient is dependent on the Re number and surface roughness ratio. • At critical and supercritical flow regimes, the displacement amplitude ratio is less sensitive to Re than that at lower Re. The displacement amplitude ratio in subcritical flow regime is significantly larger than that in critical and supercritical flow regimes. • Two excitation regions for the ‘smooth riser’ and one excitation region for the ’rough riser’ are identified. INTRODUCTION A literature review on the effects of Reynolds number (RRRR = UUUU/νν) and surface roughness ratio on the VIV responses were done by [1]. The drag coefficient, maximum response amplitude, Strouhal number and excitation coefficients are strongly depending on Re and surface roughness in the critical and post-critical flow regime, indicating that these effects should be accounted for in future VIV analysis. A 'scaling' method on the excitation coefficients was introduced to account for various Reynolds number and surface roughness. Several studies investigated the Reynolds number effects on the peak CF amplitude ratio of a freely oscillating rigid circular cylinder [2] [3]. By studying experimental results, both studies demonstrated that the peak CF amplitude ratio depend on the Reynolds number and damping. However, mainly due to the limit of experimental setup, the Reynolds number ranges are 525 − 2600 in [2] and 500 − 33000 in [3] respectively, both are in the subcritical flow regime. In addition, the surface roughness effect on the response amplitude was not studied. Response of flexible pipes were reviewed by [4], the Reynolds number ranges from 103 to 2×105, it was found that the response amplitude increases with increased Reynolds number. [5] studied Shell flexible pipe VIV model tests, which was carried out at MARINTEK's Ocean Basin. The Reynolds number range is roughly 5×10 − 2.2×105. The Reynolds number effect on the response found by [4] was confirmed, the influence of surface roughness ratio is also mentioned in [5]. The influential parameters on the responses have similarities between a flexible pipe and a freely oscillating rigid cylinder. There are few experimental studies on VIV at prototype Reynolds number (>105), mainly due to the limitation of the test facilities. ExxonMobil performed full scale Re number VIV model tests on rigid bare riser and riser with helical strakes, the Reynolds number ranges from 8×104 to more than 106 [6]. Various surface roughnesses were modelled by using sandpaper. It was found that in critical Reynolds number regime, the VIV response amplitude and the excitation coefficient of a bare riser are sensitive to Reynolds number and surface roughness. In Deepstar high Reynolds number experiments, combined in-line (IL) and CF VIV experiments were carried out at a Reynolds number range from 3.1×105 to 7.1×105 [7]. A rough cylinder was tested, desired roughness was achieved by fit a fiberglass sleeve outside the smooth cylinder, and covered in sand particles, the surface roughness ratio kk/UU = 2.3×10-3 [7]. 'Dual resonance' was observed for both subcritical smooth cylinder