Abstract
This thesis discusses the development of a Numerical Wave Tank (NWT) capable of describing the coupled behaviour of Wave Energy Converters (WECs) and their moorings under extreme wave loading. The NWT utilises the open-source Computational Fluid Dynamics (CFD) software OpenFOAM R to solve the fully nonlinear, incompressible, ReynoldsAveraged Navier-Stokes (RANS) equations for air and water using the Finite Volume Method (FVM) and a Volume of Fluid (VOF) treatment of the interface. A method for numerically generating extreme waves is devised, based on the dispersively-focused NewWave theory and using the additional toolbox waves2Foam. A parametric study of the required mesh resolution shows that steeper waves require finer grids for mesh independence.
Surface elevation results for wave-only cases closely match those from experiments, although an improved definition of the flow properties is required to generate very steep focused waves. Predictions of extreme wave run-up and pressure on the front of a fixed truncated cylinder compare well with physical measurements; the numerical solution successfully predicts the secondary loading cycle associated with the nonlinear ringing effect and shows a nonlinear relationship between incident crest height and horizontal load. With near perfect agreement during an extreme wave event, the reproduction of the six degree of freedom (6DOF) motion and load in the linearly-elastic mooring of a hemispherical-bottomed buoy significantly improves on similar studies from the literature.
Uniquely, this study compares simulations of two existing WEC designs with scale-model tank tests. For the Wavestar machine, a point-absorber constrained to pitch motion only, results show good agreement with physical measurements of pressure, force and float motion in regular waves, although the solution in the wake region requires improvement. Adding bespoke functionality, a point-absorber designed by Seabased AB, consisting of a moored float and Power Take-Off (PTO) with limited stroke length, translator and endstop, is modelled in large regular waves. This represents a level of complexity not previously attempted in CFD and the 6DOF float motion and load in the mooring compare well with experiments.
In conclusion, the computational tool developed here is capable of reliably predicting the behaviour of WEC systems during extreme wave events and, with some additional parameterisation, could be used to assess the survivability of WEC systems at full-scale before going to the expense of deployment at sea.