Validation of seakeeping calculations of a wind farm installation vessel

Abstract

The title of this master thesis is "Validation of Seakeeping Calculations of a Wind Farm Installation Vessel". The work has been conducted for Fred. Olsen Windcarrier, by Stud. Tech. Sven Ole Mollestad Nicolaysen at the Norwegian University of Science and Technology (NTNU) in Trondheim, during the spring semester 2016.

Fred. Olsen Windcarrier s specialized offshore wind installation vessel, the Bold Tern , has been subject for seakeeping calculations. The focus has been on the roll motion, which is challenging to predict, since it is highly affected by nonlinear damping. In various computer software, semiempirical theory can, to a certain extent, capture nonlinear effects, but the accuracy is limited by the model tests which the theory is based upon. In practical engineering, when the vessel is analyzed in computer software based on simplified theory, it is common to compensate for the uncaptured damping effects, by adding artificial bilge keels to the model. A bilge keel is an effective antirolling device which increases a vessel s roll damping. The size of the bilge keels is tuned so that the motion characteristics match other more reliable measurements. The reference for tuning should preferable be a model test, but this is unfortunately not available for the Bold Tern. The specific aim of the thesis has been to establish a new and better roll motion tuning reference based on full scale measurements. The results should include the correct level of damping in roll, along with suggested bilge keel dimensions.

Six months of full scale measurements of roll motion and sea state were available for this thesis. By analyzing the data, it quickly became clear that the vessel had experienced very little roll motion. This is due to the vessel s strict operational limits which prevents it from sailing in harsh weather. However, two time windows with similar properties, little variation in sea state and a maximum average significant roll amplitude between 0.9 and 1.1 degrees were selected for further spectral analysis. In the first spectral analysis it was assumed a unidirectional sea state with the total wave energy approaching the vessel from one single direction. Unidirectional response amplitude operators (RAOs) were estimated from the two time windows with similar properties, but they turned out to be quite different. The RAOs were used as a tuning reference for estimating the necessary damping level and corresponding bilge keel dimensions in a the computer program ShipX Veres. The results were inconsistent, leading to the conclusion that the unidirectional assumption limited the quality of the results too much in order to draw any further conclusions.

In the second spectral analysis, directional distribution of the wave energy was taken into account. In practice, this was done by using the measured directional roll response spectra as a tuning reference, instead of the unidirectional RAOs. This gave more consistent results. The necessary total damping level was found to be in the range of 7.1 % and 8.3 %. As a result of the calm and linear sea state, the nonlinear roll damping from the bilge keels was small. The corresponding bilge keel dimensions hence became unphysically large, with a length of 45 % of the total ship length, and a height of 57 cm and 114 cm.

It was concluded that the uncertainty related to the sea state measurements seems to be significant. Further, and most importantly, it had to be concluded that the available measurements were far too linear to give reasonable estimates of nonlinear effects. Gathering and analyzing data with more roll motion is recommended for further work. The directional spectral method seemed to give promising results which potentially can reduce the Bold Terns strict operational limits.