Application of Environmental Contour Lines on a Flexible Riser

Author(s):  
Gro Sagli Baarholm ◽  
Sverre Haver

This paper is concerned with estimating the response value corresponding to an annual exceedance probability. In principle, this response needs to be obtained by combining the response statistics in all sea states and wave headings. The probability distribution for a given sea state and wave direction, specified by the significant wave height, Hs, the peak period, Tp, and wave direction, β, can be obtained by frequency domain analysis for linear response. Time domain simulations are, in general, required to obtain the stochastic structure of a non-linear response quantity. To limit the statistical uncertainty, the time domain simulations must be sufficiently long. Therefore, a simplified method is necessary to improve the efficiency of the direct calculation of the long-term response value in the non-linear case. A method to determine the long-term extremes by considering only a few short term sea states is outlined. The sea states have a certain probability of occurrence and are identified by a contour line in the (Hs,Tp) plane. This will make possible practical estimation of the extreme loads the structure is exposed to. The contour line approach is merely suggested as a method for predicting load- and response maxima corresponding to a given annual exceedance probability without having to carry out a full long term analysis. The advantage with this concept is that the environmental and response analysis is decoupled. This is very convenient if the problem under consideration is of a very non-linear nature — in particular if characteristic values for design are to be established directly from model tests. The method is an approximate method, but seems to give results of reasonable accuracy for most problems. The purpose of the present paper is to present the contour line method applied to estimate responses on a flexible riser, lazy wave configuration, located in the North Sea.

Author(s):  
Gro Sagli Baarholm ◽  
Sverre Haver ◽  
Carl M. Larsen

This paper is concerned with estimating the response value corresponding to given annual exceedance probability. In principle, this requires that a full long term analysis is executed. For a linear response this can easily be done. For a non-linear response quantity however, where time domain simulations are required in order to obtain the short term stochastic structure a full long term analysis will be time consuming. An approximate method to determine the long-term extremes by considering only a few short term sea states is outlined. All sea states corresponding to a certain probability of occurrence and are given by a contour line of Hs, Tp for each wave direction. The advantage of the method is that a proper estimate of the long term extreme can be obtained by considering the most unfavourable sea state along the contour line. This will make possible practical estimation of the extreme loads the structure is exposed to. The purpose of the present paper is to illustrate how to apply directional contour lines in order to obtain a characteristic design value according to requirements regarding the marginal exceedance probability.


Author(s):  
Roque Corral ◽  
Juan Manuel Gallardo ◽  
Rahul Ivaturi

The response of aerodynamically unstable tuned bladed-discs with non-linear friction dissipation at blade-root attachments due to harmonic external excitation is studied. The bladed-disc is modeled using a simple mass-spring system and the effect of friction is modeled using a micro-slip friction model. The response is computed in time domain using a Runge-Kutta scheme. The time domain response is decomposed to obtain the evolution of traveling waves in the bladed-disc. Parametric studies have been conducted to study the non-linear response at different vibration amplitudes at high and low engine orders of excitation. It is seen that the non-linearity due to friction gives rise to a complicated interaction between the synchronous response of the system due to harmonic excitation and the non-synchronous response of the system due to aerodynamic instability. For low excitation levels the system behaves as in the pure flutter regime where a single, or at most a few, aerodynamically unstable modes may be found in the final state when a limit cycle is reached. When the forcing is large enough the aerodynamic instability is suppressed and only the non-linear response of the excited mode may be seen. It is concluded that the superimposition of the flutter and forced response analysis in terms of vibration amplitude is not valid and leads to prediction of vibration amplitudes significantly larger than that obtained when both phenomena are simulated together.


2001 ◽  
Vol 45 (03) ◽  
pp. 228-240
Author(s):  
Gro Sagli Baarholm ◽  
Torgeir Moan

This paper is concerned with estimating the response value corresponding to a return period, say 100 years. In principle, this response needs to be obtained by combining the response in all the sea states. The response for a given sea state, specified by the significant wave height, Hs, and the peak period, Tp, can be obtained in the frequency domain for the linear response. Time domain simulation is required to obtain the nonlinear response, and long time series are required to limit the statistical uncertainty in the simulations. Especially in the latter case, it is crucial to introduce ways to improve the efficiency in the calculation. A method to determine the long-term extremes by considering only a few short-term sea states is outlined. The sea states have a certain probability of occurrence and are identified by a contour line in the (Hs TJ-plane. This will make possible practical estimation of the extreme loads the vessels are exposed to.


Author(s):  
Sverre Haver ◽  
Gudmund Kleiven

Methods of prediction of structural loads corresponding to a required target annual exceedance probability are reviewed. Particular attention is given to utilization of environmental contour lines for such a purpose. This approach is based on using short term methods for predicting adequate estimates of the q-probability response. The environmental contour line approach is a very convenient approach if complicated structural problems are considered. For such problems one will often have to involve numerical time domain analyses or model tests to reveal the short term probabilistic structure of the response maxima, making a full long term response analysis impossible for most practical problems.


Author(s):  
Finn-Idar Grøtta Giske ◽  
Arnt Fredriksen

Abstract In this paper, long-term extreme response analysis is performed for a straight floating bridge across the Bjørnafjord, using a recently developed inverse first-order reliability method (IFORM) approach. Full integration of the long-term extreme response formulation is also performed for comparison. Two different environmental models are estimated based on a scatter diagram of significant wave height and peak period for the given location. The IFORM method is seen to provide reasonable estimates of the long-term extreme response, at a significantly reduced computational effort.


Sign in / Sign up

Export Citation Format

Share Document