Uncertainty Quantification of Riser Fatigue Damage due to VIV Using a Distributed Wake Oscillator Model

Abstract

This study explores the use of polynomial chaos expansion (PCE) to quantify the uncertainty in accumulated fatigue damage in a top-tensioned riser (TTR) due to vortex-induced vibration (VIV). Time-domain simulations of the response of the selected riser are carried out using a distributed wake oscillator model and fatigue damage is computed using rainflow cycle-counting. The uncertainty in damage prediction that results from variability in parameters involved in the wake oscillator model is the focus of this study. The efficiency and accuracy resulting from use of the PCE model is demonstrated by comparison against Monte Carlo simulations (MCS). Specifically, starting with the use of two random variables (the cylinder maximum amplitude and a ratio of the vortex-shedding frequency to the natural frequency) and then by introducing one other random variable (the current velocity), the propagation of uncertainty from the wake oscillator model inputs to fatigue damage is studied. Numerical studies demonstrate the versatility of the PCE-based approach in such uncertainty quantification applications.