Offshore tethered platform springing response statistics

Abstract

This paper demonstrates the validity of the Naess–Gadai method for extrapolating extreme value statistics of second-order Volterra series processes through application on a representative model of a deep water small size tension leg platform (TLP), with specific focus on wave sum frequency effects affecting restrained modes: heave, roll and pitch. The wave loading was estimated from a second order diffraction code WAMIT, and the stochastic TLP structural response in a random sea state was calculated exactly using Volterra series representation of the TLP corner vertical displacement, chosen as a response process. Although the wave loading was assumed to be a second order (non-linear) process, the dynamic system was modelled as a linear damped mass-spring system. Next, the mean up-crossing rate based extrapolation method (Naess–Gaidai method) was applied to calculate response levels at low probability levels. Since exact solution was available via Volterra series representation, both predictions were compared in this study, namely the exact Volterra and the approximate one. The latter gave a consistent way to estimate efficiency and accuracy of Naess–Gaidai extrapolation method. Therefore the main goal of this study was to validate Naess–Gaidai extrapolation method by available analytical-based exact solution. Moreover, this paper highlights limitations of mean up-crossing rate based extrapolation methods for the case of narrow band effects, such as clustering, typically included in the springing type of response.

Offloading operation bivariate extreme response statistics for FPSO vessel

“…In this study, authors employed a general Monte Carlo (MC) based technique, able to tackle inherent nonlinear effects without significant simplifications except those inherent in the numerical model itself. The target bivariate extreme response distribution was obtained by combining MC simulations with the ACER2D extrapolation method [12][13][14][15][16] . Unlike univariate statistical techniques, the ACER2D method accounts for bivariate response correlation.…”

Improving extreme offshore wind speed prediction by using deconvolution

“…The countless strategies endeavoured included the method presented by Mohande [ 26 ], which predicts wind speeds using neural networks or the different approaches to analyse wind speed estimates by Yan et al, as in Refs. [ [1] , [2] , [3] , [4] , [5] , [6] , [7] , [8] , [9] , [10] , [11] , [12] , [13] , [14] , [15] , [16] , [17] , [18] , [19] , [20] , [21] , [22] , [23] , [24] , [25] , [26] ]. Similarly, the authors of this paper have also previously used different statistical approaches to estimate better extreme statistical values such as, i.e.…”

“…Similarly, the authors of this paper have also previously used different statistical approaches to estimate better extreme statistical values such as, i.e. wind speed, wave height, response and load [ [1] , [2] , [3] , [4] , [5] , [6] , [7] , [8] , [9] , [10] , [11] , [12] , [13] , [14] , [15] , [16] , [17] , [18] , [19] , [20] , [21] , [22] , [23] , [24] , [25] , [26] ]. Furthermore, the authors developed a novel reliability approach to forecast COVID-19 epidemic levels, indicating the approach's versatility and reliability in various fields [ 14 , 16 , 26 ].…”

“…[ [1] , [2] , [3] , [4] , [5] , [6] , [7] , [8] , [9] , [10] , [11] , [12] , [13] , [14] , [15] , [16] , [17] , [18] , [19] , [20] , [21] , [22] , [23] , [24] , [25] , [26] ], have attempted to use novel statistical approaches to estimate wind speed or pattern more accurately. Meanwhile, two recent and pertinent scientific articles [ 8 , 26 ] illustrated the importance of wind forecasting and strategies adopted to maximise efficiency. Like many of the papers mentioned above, the deconvolution methods implemented in this paper hope to accomplish better extreme offshore wind speed prediction.…”

“…This is in contrast to many traditional extrapolation methods widely used in engineering practice where an assumed probability function is used [ [1] , [2] , [3] , [4] , [5] , [6] , 30 ]. Some other examples also include the Pareto-based distribution peak over the threshold (POT) [ 1 ], The Naess-Gaidai (NG) method fitting procedure was used in averaged conditional exceedance rate (ACER) method [ [7] , [8] , [9] , [10] , [11] ], bivariate ACER fitting method [ [12] , [13] , [14] , [15] , [16] , 47 ], Weibull distribution-based fitting method [ 38 ] and Gumbel distribution based fitting method [ 17 ].…”