Abstract
Utilising tidal currents as a renewable energy resource is presently under consideration to meet the requirements of increasing worldwide energy demand and the need to reduce carbon emissions. In this respect, in-stream tidal devices are proposed to convert the kinetic energy of currents into useful extractable power. In order to extract a useful amount of energy from tidal currents, the proposed devices need to be deployed in an array or farm-like format. Due to the thrust exerted by the devices within an array, the natural flow regime will inevitably be changed. In light of this, this study aims to estimate the maximum power that can be extracted by tidal turbine arrays and assess the far-field effects of energy extraction in the designated areas around the UK for various array configurations.
In this thesis, the ocean tides are modelled using the long wave equations, commonly referred as the shallow water equations (SWEs). A numerical solver based on a Runge-Kutta discontinuous Galerkin finite element method is employed to solve the SWEs. One main advantage of the discontinuous Galerkin method is that it approximates the solution individually at each element, which allows for discontinuities within the solution system while ensuring mass conservation locally and globally. The selected numerical solver has been verified against several benchmark tests. It is then modified to include a line discontinuity to represent the effect of tidal turbine array(s) in a coastal basin. The algorithm implemented in the numerical solver involves a sub-grid model, which is based on Linear Momentum Actuator Disk Theory (LMADT) to approximate the local flow-field in the presence of the turbines. This near-field approach allows the flow velocity at the turbine to be estimated with a greater accuracy. As the power available to the turbines is related to the velocity at the turbine blades, the characterisation of the designated tidal site as a resource using LMADT may be more accurate than previously proposed methods. An additional advantage of using LMADT is that it provides a distinction between the power extracted by the turbines and the total amount of power that is removed from the tidal stream, including the wake mixing losses. The methodology employed in this thesis has been applied to two tidal basins around the UK; the Anglesey Skerries (a headland) and the Bristol Channel (an oscillating bay). A comprehensive unstructured triangular finite element model has been constructed to simulate the naturally occurring tides at these regions. The constructed model has then been validated against field measurement.
The validated model is used to conduct parametric studies, which evaluate the importance of tidal array locations, configurations and operating conditions on the available power at the Anglesey Skerries and the Bristol Channel sites. The parametric study aims to evaluate a realistic upper limit of available power at each site considered. This study also provides a unique analysis to examine the potential tidal farm interactions by deploying several tidal arrays at both Anglesey Skerries and the Bristol Channel.