Approximation of non-normal responses for drag dominated offshore structures

Abstract

For deep water offshore structures, the modeling of random waves loads by the Morison equation yields an equation of motion which admits no analytical solution except in few limit cases. This also holds for response moments which have a great influence on the results of reliability analyses. If polynomial approximations of the Morison drag loads are introduced, some procedures are available to obtain the stationary moments of the approximate response. These procedures result in large computer codes and time consuming computations if large structures are considered. Therefore, only first order approximations of the drag nonlinearity usually can be handled. In the paper, the ability of two famous linearization schemes to predict response variance and response fourth moment is numerically investigated. Improvements of these schemes are considered. For quasi-static structures, it is shown that except in very simple cases it is practically impossible to optimize linearization schemes to fit response variance or fourth moment.

Sensitivity approach for modelling the environmental loading of marine structures through a matrix response surface

“…RSM has been largely developed during the last decade, and response function models, including non-linear ones, are now available, especially for the specific needs of structural reliability analysis [25][26][27][28] and stochastic problems such as crack propagation [29]. This methodology is basically a formal representation based on geometrical ideas: surface building in the probabilistic space.…”

“…For reliability analyses, the investigated response is either strength, load variables or limit state function. The surrogates to original models are classically analytical functions (mainly polynomial expression) whose parameters are estimated from experimental results [26,27,30]. The building criteria must be developed according to the specific reliability analysis.…”

“…It is a time-consuming option when computing, and which does not give necessarily more realistic results. [26]. Table 3 gives the results for both approximation models: Caughey's procedure and the proposed response surface.…”

Surface response meta-models for the assessment of embankment frictional angle stochastic properties from monitoring data: An application to harbour structures

“…This analysis focuses on an analytical response function defined by a polynomial of order less than 3, with or without interaction terms, known as a quadratic response surface (QRS). QRS has been developed in several fields [19][20][21][22]. This form has specific properties, e.g., asymptotic behaviour for the transfer of distribution tails, which fit the physical problem in this study.…”