Enhanced vortex-induced vibration prediction of marine risers using excitation coefficient databases at high Reynolds numbers

Abstract

Current methods for predicting vortex-induced vibration (VIV) of marine risers mainly depend on excitation coefficient databases obtained from forced vibration experiments on a rigid cylinder at a low Reynolds number (Re = 1.0E4). However, investigations at higher Reynolds numbers have revealed notable effects on VIV behaviors, resulting in discrepancies when predictions are based on previous experimental databases. To resolve this, forced vibration experiments were carried out on a rigid cylinder model over a Reynolds number range of 5.0E4 to 3.5E5. The acquired data enabled the development of comprehensive excitation coefficient databases covering both subcritical and critical Re regimes. A non-iterative frequency-domain prediction method integrating the updated excitation coefficient databases was proposed and validated through experiments on a flexible riser model. Subsequently, predictive calculations were conducted to investigate the effects of the Reynolds numbers in the excitation coefficient databases on the prediction outcomes. Comparative analysis shows that method based on a higher subcritical Reynolds number (2.0E5) tend to predict larger vibration amplitudes and amplified fatigue damage, whereas those corresponding to critical Reynolds numbers (3.0E5 and 3.5E5) indicate lower amplitudes and reduced fatigue damage. These findings underscore the importance of using excitation coefficient databases that match the relevant Reynolds number conditions to mitigate the risks of overly optimistic or conservative marine riser designs.