Non-parametric prediction of the long-term fatigue damage for an instrumented top-tensioned riser

Marine risers are susceptible to sustained vortex-induced vibration (VIV) because of their slenderness and light damping. Commonly used tools for analyzing VIV and the associated fatigue damage are based on the finite element method and rely on simplifying assumptions on the riser's physical model, the flow conditions, and characteristics of the response. In order to assess the influence of VIV and to ensure the integrity of the riser, field monitoring campaigns are often undertaken wherein data loggers such as strain sensors and/or accelerometers are installed on such risers. Given the recorded riser's dynamic response, empirical techniques can be used in VIV-related fatigue estimation. These empirical techniques make direct use of the measurements and are intrinsically dependent on the actual current profiles. Damage estimation can be undertaken for the different current profiles encountered and can account explicitly even for complex riser response characteristics. With a significant amount of data, “short-term” fatigue damage probability distributions, conditional on current, can be established. If the relative frequency of different current types is known from a separate metocean study, the short-term fatigue damage distributions can be combined with the current distributions to yield an integrated “long-term” fatigue damage model, which then can be used to predict the long-term cumulative fatigue damage for the instrumented riser. Non-parametric statistical techniques (that do not assume a specific function for the underlying distribution as parametric techniques do) are employed to describe the short-term fatigue damage data. In this study, data from the Norwegian Deepwater Programme (NDP) model riser experiments are used to demonstrate the effectiveness of empirical procedures and non-parametric statistics applied to field measurements to predict long-term fatigue damage, life, and probability of fatigue failure.

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01689883
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Nov 21 2018 3:15PM