Abstract
This thesis investigates the numerical modelling of the Oscillating Wave Surge Converter (OWSC). The OWSC is a bottom hinged, flap-type, pitching wave energy converter. The device is installed in shallow water (10 - 15 metres} and covers the entire water column from the sea floor to the water surface. Its motion is driven by the horizontal fluid velocity due to incoming waves.Numerical models of the OWSC were constructed with the aim of providing a tool for power production predictions, setting up a testing platform for various control strategies and to feed back information into the device design process. In particular, two independent modelling techniques are investigated which rely on different kinds of input data.
First a hydrodynamic model is presented. For the construction of this model it is necessary to know all the coefficients that influence the motion of the device. The hydrodynamic coefficients are calculated with the linear wave – structure interaction analysis software WAMIT and the remaining coefficients are determined from data acquired from tests in a wave tank with a physical scale model of the device.
The second numerical model is based on Volterra theory. This method models the device as an ordered expansion of convolution integrals. The convolution kernels are extracted from tank test data using a system identification technique.
The accuracy of the hydrodynamic and the 1st order Volterra model is found to be comparable. The accurate prediction of the motion of the OWSC makes the models suitable for the testing of control strategies for the full scale version of the device (Oyster, developed by Aquamarine Power Ltd.}. The models can also be used to predict the power Oyster will produce in a wide range of sea states.
Non-linear effects are believed to have a significant influence on the motion of Oyster in typical operating conditions. While a non-linear damping term can be implemented in the hydrodynamic model the coefficients that are determined with WAMIT remain linear. System identification groups all effects of the same order together. This means that all 2nd order effects are accounted for in the model provided that the model used is of sufficiently high order). An introductory investigation into the application of a 2nd order Volterra model is also presented. Higher order system identification modelling provides a greater insight into
complex non-linear interactions which can be used to provide a more accurate description of the device behaviour or can aid in the further development of the hydrodynamic model.
| Date of Award | Dec 2009 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Trevor Whittaker (Supervisor), Kenneth Doherty (Supervisor), Alan Henry (Supervisor) & Matthew Folley (Supervisor) |