Fabrication of the Hywind Substructure
MetadataVis full innførsel
- Institutt for marin teknikk 
As the technological feasibility of the Hywind has been proven through the Hywind demo project, the way forward is to check the technological and economic feasibility of upscaling the fabrication. In the last years, several concepts have been presented to try to tap into the vast amount of wind resources available offshore. For floating wind, one of the main challenges has been to make the concepts cost efficient enough to compete with other sources of energy. One of the strengths of the floating wind turbines are their soil-independent design creating possibilities for mass production of standardised units. This thesis investigates the current technological and economic constraints in the fabrication phase of the Hywind substructure and looks at which risks exist and how these can be mitigated to ensure a safe fabrication. Furthermore, it breaks down the costs of the fabrication into its main components to look at how much the costs can be reduced to in a mass production perspective. The method of doing this is a work breakdown structure (WBS) of the activities done at the yard, their inherent risks, costs, technological limitations and effect on fabrication schedule. The schedule is simulated using Matlab to look at the main bottlenecks in the fabrication. Uncertainties in the different steps are introduced by the use of Monte Carlo Simulation to give an overall picture of the variation in costs, schedules and their risks. Through the WBS and structural analyses, the feasibility of the fabrication methods likely to be applied for the fabrication of Hywind is validated and the main risks identified. The investigations show that the main risks connected to the fabrication are well known and mitigation of these risks are connected to third party verification, dimensional control and non-destructive testing. Furthermore, the schedules and costs risks are shown to be reduced with increasing volume of output and the cumulative cost reductions of the fabrication is from the current cost of about 5.7 MUSD down to an average cost of 3.5 MUSD corresponding to a marginal cost of 3.4 MUSD per unit. The schedule of the fabrication is simulated using SimEvents and this showed improvements in resistance to both variations in time consumption and bottlenecks for the upscaling of the fabrication. Additionally, a tool for optimising the resource investments for upscaling of the production is demonstrated and it is found that the time valuation of the fabrication is of high importance to the yard s optimal investment strategies. From a theoretical point of view, it is clear that the fabrication of Hywind can exploit economies of scale and learning to reduce the current costs and that the fabrication methodology is scalable from the current five batch production to mass production. The main challenges going forward is handling the size of the structures efficiently without compromising on the structural integrity.