Please login to view abstract download link
Offshore wind turbines have grown steadily in size over the past years to optimize their energy output and are expected to continue to do so. These future large wind turbines will consist of slender structures more prone to large deformations than current generation wind turbines. Nonlinear dynamic phenomena in addition with not yet well-known loads will become more dominant and the interactions between the different components of the wind turbine and with the environment will have to be considered more detailed. Therefore, currently used methods such as Blade Element Momentum Theory and linear beams might no longer sufficient to accurately predict the behaviour of such large wind turbines. Since high-fidelity methods like CFD are to computationally expensive, mid-fidelity methods seem to present an attractive compromise between accuracy and computational effort. Consequently, we are developing a mid-fidelity nonlinear aeroelastic approach, which combines the unsteady vortex-lattice method (UVLM) and a flexible multi-body system/finite elements approach. The structural model consists of rigid bodies, geometrically exact beams and holonomic and non-holonomic constraints. The nonlinear structural and aerodynamic equations are coupled strongly using an approach based on the principle of virtual work. The hydrodynamic forces acting on the sub-structure are computed using the well-known approach based on the Morison equation. In this work, we verify our tool by comparing the results of aero-hydro-servo-elastic simulations using our tool with results obtained by OpenFAST, one of the current standard tools. We compare the displacements and stress resultants at different locations in the wind turbine for different load cases, both regarding the aero- and hydrodynamic loads. By investigating wind turbines of different sizes, e.g., the NREL 5 MW reference wind turbine with rotor blades of length 61.5 m as well as the NREL 15 MW turbine with large slender blades of length 117 m, it can be shown that for larger wind turbines geometrical nonlinearities become more relevant for the dynamic behaviour of the complete wind turbine and thus a nonlinear model is required.