Modeling of breaking wave loads on jacket structure in FEM software USFOS

Abstract

This thesis will in the first part present introduce and state the current development of offshore wind energy, and some of the motivations behind it. It will then present the current configurations and justify the need for better an understanding of the conditions, in designing and modeling the new foundations types like jackets are needed as the industry is expected to move into deeper waters. As a step in bettering the understanding of the conditions that are expected in deeper water, the WaveSlam project was lunched. This project had a goal to better understand the magnitude of the breaking wave forces one could expect to see on a jacket in deep water conditions. The second part will go through some of the central used for theory for calculating breaking waves. With most of the theory being later applied directly in the report or indirectly through the use of the computer software tools. In the third part the theory is applied to a model in USFOS in an effort to simulate the conditions expected to be observed. As the model is geometrically equal to that used in the WaveSlam, and the model was adjusted so its response in the simulation was comparable. In the final parts of the report, the comparison of the experiment structure and the model allowed for an investigation into the problems that may arise from the use of an FEM software such as USFOS. The central structural parts have to be fitted with the correct parameters individually, as the earlier adjustments was not sufficient for new load cases. Once responses of the model was in accord with the responses of the structure, it was possible to establish that there are inconsistencies that is introduced by USFOS wave generator. It caused the Eigen frequency of the model to shift to a lower value that it originally was fitted for. The cause of this Eigen frequency shift, and to how large an extent it effects the results was not established. However the thesis was able to compare the model best fitted response to that of the experiment. The results from then introducing a load to the model based on a load case from the current monopile theories it is possible to see the extent of inaccuracy in the monopile theories when used on jackets. The thesis have not looked at accuracy of the load distributions on the jacket, and there might be some effects that are caused by an error in it. Therefore there are still some problems that have to be addressed before a certain conclusion can be drawn. Concluding remarks - USFOS is a FEM tool that can recreate slam events the associated dynamic response, there is still some inaccuracy that has to be addressed when the modeling the waves used in combination with the slam impulse. - Monopile theory offers little coherence between the load and response calculated, and the actual observed loads on a jacket. Further investigations should focus on finding a correct way to represent the load distribution and duration accurately.