Forces on simplified offshore structures according to different wave models.
MetadataVis full innførsel
The reason for creating this thesis was because of the new and revised version of NORSOK N-003 standard. Therefore a comparison between the old and the revised version of NORSOK N-003 standard has been performed. This thesis have been divided in to three main parts. The first part describe how to estimate the 10−2 annual probability crest height, c0.01 and wave height ℎ0.01. (100-years wave) with the metocean contour lines method. Which resulted in: ℎ0.01. = 28.61m and c0.01 = 17.87m by using ℎs = 14.9m and tp = 15.8s The second part revolves around regular waves. A comparison between the old and new method of calculating the ULS design wave have been discussed. The old method uses a Stokes wave profile defined by the 10−2 annual probability wave height, ℎ0.01 with an unfavorable period. Where the new method uses the 10−2 annual probability crest height, c0.01 with a mean wave period to define the ULS design wave. With the same defined wave profiles as the new and old recommendation, one have also compared the Stokes wave with a first order approach. By obtaining the kinematics from all approaches and compared them, one may see that a linear approach has the ability to obtain very close kinematics as the Stokes wave. This depends on the amplitude used and which approximation above mean surface level used. After words, the base shear and overturning moment where calculated by Morison equation. Those results shows that the new method using Stokes wave with c0.01 as the amplitude, results in a larger base share and overturning moment for drag and non-dominated forces but a lower overturning moment for a mass dominated case comparing to the old method using a wave height equal to, ℎ0.01. The conclusion for this part is that the new N-003 standard is more efficient with time and describe the waves in a more accurate manner. For the third and last part one have chosen to discuss irregular wave, where the old N-003 standard suggests a first order process to obtain the corresponding kinematics of a time simulation. Where the revised N-003 standard in other hand require a second order process to describe the surface process and a second order theory to obtain the kinematics of the time history. Matlab has been used to create those simulations and calculated all the data for this thesis. By comparing the two different processes, one found out that the: - Formula used to create the first order irregular surface process follows a Rayleigh distribution for crest heights and the second order surface process follows a Weibull distribution for crest heights. - Wheeler stretching for a first order process underestimates the kinematics, but a constant value above mean surface level is a very good approximation to a second order process using Standsbergs approached. - First order process underestimates the crest heights but overestimates the kinematics, which achieves almost the same result as second order process.
Master's thesis in Offshore structural engineering