ABSTRACT

This paper presents an approach to the prediction of long term responses of offshore vessels for design and operability assessments. The evaluation of the vessel responses takes account of the random variations of spectral shapes within the same seastate due to sea/swell directionality and other oceanographic phenomena. The resulting response statistics which use this approach will not only simulate the average condition, but will also indicate the probability of the extreme conditions that may be encountered by the vessel during its mission time.

The methodology is applied to offshore construction planning and transportation studies to illustrate the importance of the long term effects of ocean wave spectra in an overall design and operability assessment. A detailed mathematical development of the approach is also included in the Appendices.

INTRODUCTION

The offshore industry has long used the concepts of design wave height/period and sea spectra to describe environmental extremes and operational conditions for structure response evaluation. One way to estimate the extreme wave height and period is to utilize the statistical concept of asymptotic extreme values and the joint probability distribution of height and period. This is called long term prediction. Typical examples of the long term prediction method for extreme wave height are shown in Reference (1,2). Another approach is to extrapolate the extreme wave height from a steadystate sea condition, taking into account the exposure time during the structure's life time. This is called short term prediction approach (3,4).

Both of these procedures produce a single wave height which describes the extreme environmental condition that is likely to be encountered by the structure. However, since a structure's responses are frequency dependent, unless the random nature of the wave energy distribution in the frequency domain is properly reflected in the prediction technique, there exists some reservations with regard to the confidence of the predicted response which are significantly effected by the wave periods.

One approach to account for such variations in wave spectral energy contents, and hence logically interpretate the extreme values obtained by the long term procedure, is the use of the wave spectral family concept (5,6). Instead o£ using a single theoretical spectrum to represent a given sea state, several spectra of the same level of sea severity as characterized by its significant wave height are selected. The spectra selected represent the variation of spectral energy contents due to long term variation of sea/swell directionality and other oceanographic phenomena.

The methods for selecting spectra may vary; some will utilize Monte Carlo techniques in generating random samples within a wave height group to match the spectral properties of the group (5), while others will use different parameters of theoretical spectral formulation based on statistical analysis of recorded data (6). Then, the procedure to evaluate the extreme wave height through the short term approach is essentially the same as that for the long term, with a restriction that only one sea condition be considered (3).

This content is only available via PDF.
You can access this article if you purchase or spend a download.