With ever increasing wind turbine sizes and a move towards locations further offshore and with deeper waters, a significant number of future wind turbines are going to be established on jacket type support structures. These multi-member structures are robust but can be costly. Accurate knowledge of the wave forces acting on these structures is necessary to reduce risks and develop more economic designs.

A major challenge is the complex geometry of joints and multiple members, leading to hydrodynamic shadowing and wake effects. The main objective of this project is to develop a better understanding of fatigue loads on jacket structures, based on extensive computational fluid dynamics simulations.

The secondary objective is to develop an efficient data-driven model that, after training, allows the prediction of wave forces for different jacket geometries without large computational expenses.

The fellow will develop novel load models for jacket structures.

A first result will be a computational fluid dynamics model for accurately predicting wave forces on such a complex, multi-membered structure, with particular focus on different possible joint geometries and important geometric features (e.g. boat landings and J-tubes). A second result will be a database of various structural details and their loadings, evaluated with the simulation model. Finally, ML techniques will be used to develop an efficient meta-model able to interpolate and predict loads for different geometries from non-linear wave kinematics, with an eye towards use with structural optimisation algorithms.

Academic secondment at ETH Zurich (3 months, supervised by Prof. Eleni Chatzi, M13-15) to work on the theoretical foundations and techniques for solving the optimization problems. Industrial secondment at EnBW (3 months, supervised by Dr. Lisa Ziegler, M22-24) to work with real-world wind turbine data and develop a practical approach that can be used in industry.