Design loads and long term distribution of mooring line response of a large weathervaning vessel in a tropical cyclone environment. (September 2018)

• Record Type: Journal Article
• Authors: Stanisic, Dunja; Efthymiou, Mike; Kimiaei, Mehrdad; Zhao, Wenhua
• Is Part Of: Marine structures. Volume 61(2018)
• Page Start: 361
• Page End: 380
• Publication Date: 2018-09
• Subjects: Extreme mooring response -- Long term distribution -- Turret moored vessel -- Extreme value analysis -- Tropical cyclone environment; Naval architecture -- Periodicals; Offshore structures -- Periodicals; Architecture navale -- Périodiques; Structures offshore -- Périodiques; Naval architecture; Offshore structures; Periodicals; 620.4162
• DOI: 10.1016/j.marstruc.2018.06.004
• Languages: English
• ISSNs: 0951-8339

Abstract

Abstract: The mooring design of a floating offshore structure requires the estimation of mooring responses corresponding to annual exceedance probabilities of 10 −2 (extreme event) and sometimes 10 −4 (survival event). The most straightforward method to determine the extreme design response under a specified design sea state, is to carry out N time domain simulations, so as to capture the inherent randomness of this sea state and use the N maximum values to estimate the most probable maximum response for design. However, this requires typically 30–40 time-domain analyses of the same design sea state, which is computationally extensive. In this paper it is shown that the required number of time domain simulations can be reduced significantly by utilising the peaks of the mooring tension time series, obtained from time domain simulations, to derive a distribution for the maxima. Different variations of using these peaks are explored and a "best practice" for this technique is proposed. In order to establish a robust benchmark for evaluating and validating this "best practice", extensive time domain simulations have been carried out for a large permanently connected, weathervaning vessel, with catenary mooring system, in a tropical cyclone environment. Both extreme and survival conditions are explored, by running 170 3-h simulations for each condition, thereby representing in detail the random nature of each sea state. It is shown that a reliable distribution for the maxima, (within ± 4 % from the benchmark) can be obtained in a manner which is simple and computationally efficient, based on just 4–7 time domain analyses. Thus by using more peaks from the time domain analyses, there is a significant gain in terms of accuracy and efficiency. The above "best practice" is used to calculate the most probable maximum mooring line response and the variability of this maximum (short term variability) within a 3-h sea state for environmental conditions with annual exceedance probabilities of 10 −2 and 10 −4 . It is shown that the short term variability is not invariant but may be described in terms of a Gumbel distribution whose parameters depend on the magnitude of the response. These expressions provide a means of calculating the long term distribution of mooring line load, accounting for the short term variability, which can be used to address the reliability of a mooring system. Highlights: A "best practice" procedure is presented to determine the extreme design response obtained from time domain simulation. Peaks, defined as maximum values between two mean up crossings, should be fitted, using MoM, to a two parameter Weibull distribution. The number of peaks needed for accurate fit corresponds to four 3-h simulations under 100yr conditions and seven 3-h simulations for 10, 000yr conditions. Long term distribution of the mooring line load was developed, accounting for short term variability. The contribution of short term variability in a tropical cyclone environment is lower than in winter storm environments.