Tidal stream energy has gained attention as a source of predictable and renewable energy. Devices resembling underwater wind turbines, placed in fast tidal streams, have been proposed to extract this energy. Arrays of many such devices will need to be deployed to deliver a significant amount of energy to the electricity grid. One consequence of energy extraction is that the array provides a resistance to the tidal stream, which may change the local and far field hydrodynamics, which in turn affects the power available to the array. Array-scale hydrodynamic changes affect the flow presented to the devices, which in turn affects the total resistance the array provides to the flow. This thesis is concerned with the interactions between device, array, and the tidal stream resource, to better understand the power potential of turbine arrays. Linear momentum actuator disc theory is employed to describe the operation of an idealised turbine array partially spanning a wide channel. The model is comprised of two quasi-independent sub-models, an array-scale model, describing flow phenomena around the array, which provides the upstream boundary condition to the device-scale model, describing the flow around a device. The thrust applied by the array is the sum of the thrust applied by the devices, completing the sub-model coupling. The numerical simulation of arrays in depth-averaged simulations is then investigated using the two-scale concept developed in the analytic partial-array model. It is shown that the device-scale flow must be modelled with a sub-grid scale model in order to correctly describe the unresolved device-scale flow and hence estimate the power available to an idealised array. Turbulence modelling in depth-averaged simulations of turbine arrays is also discussed. Temporal variations in tidal amplitude and strength mean that generator capacity will need to be economically matched to the available resource. As device performance may consequently depart from the relationship derived in idealised models when power capping is employed, blade element momentum theory is modified to parameterise tidal turbine performance during power capping. The array-scale effect of power capping is studied in depth-averaged simulations, in which it is shown that a significant reduction in device thrust may occur during power capping, reducing the impact of energy extraction from the tidal stream.

This study presents a numerical investigation of a generic horizontal axis cross-flow marine turbine. The numerical tool used is the commercial Computational Fluid Dynamics package ANSYS FLUENT 12.0. The numerical model, using the SST k-w turbulence model, is validated against static, dynamic pitching blade and rotating turbine data. The work embodies two main investigations. The first is concerned with the influence of turbine solidity (ratio of net blade chord to circumference) on turbine performance, and the second with the influence of blockage (ratio of device frontal area to channel crosssection area) and free surface deformation on the hydrodynamics of energy extraction in a constrained channel. Turbine solidity was investigated by simulating flows through two-, three- and four-bladed turbines, resulting in solidities of 0.019, 0.029 and 0.038, respectively. The investigation was conducted for two Reynolds numbers, Re = O(10^5) & O(10^6), to reflect laboratory and field scales. Increasing the number of blades from two to four led to an increase in the maximum power coefficient from 0.43 to 0.53 for the lower Re and from 0.49 to 0.56 for the higher Re computations. Furthermore, the power curve was found to shift to a lower range of tip speed ratios when increasing solidity. The effects of flow confinement and free surface deformation were investigated by simulating flows through a three-bladed turbine with solidity 0.125 at Re = O(10^6) for channels that resulted in cross-stream blockages of 12.5% to 50%. Increasing the blockage led to a substantial increase in the power and basin efficiency; when approximating the free surface as a rigid lid, the highest power coefficient and basin efficiency computed were 1.18 and 0.54, respectively. Comparisons between the corresponding rigid lid and free surface simulations, where Froude number, Fr = 0.082, rendered similar results at the lower blockages, but at the highest blockage an increase in power and basin efficiency of up to 7% for the free surface simulations over that achieved with a rigid lid boundary condition. For the free surface simulations with Fr = 0.082, the energy extraction resulted in a drop in water depth of up to 0.7%. An increase in Fr from 0.082 to 0.131 resulted in an increase maximum power of 3%, but a drop in basin efficiency of 21%.

The addition of renewable energy in a previously diesel-powered off-grid micro-grid results in what is known in the field as a Stand-Alone Hybrid Renewable Energy System (HRES). Such initiative is a near-term target of both federal and provincial governments in Canada. Not only does it reduce environmental hazards like leaks, spills and air pollution, but the combination of renewable energy and fossil fuel generators can increase stability and lower the cost of electricity. It is deemed a crucial step towards a clean energy future, but also a necessity in the reconciliation process with Indigenous Peoples of Canada - many of who inhabit off-grid communities. The addition of renewable energy can greatly increase the independence of a community by reducing reliability on external diesel suppliers and creating job opportunities. To be successful, HRES need to be carefully planned; the variable and uncertain behaviour of natural resources add a level of complexity to the preliminary design stage. Energy systems are therefore simulated and optimized to estimate the lifecycle cost by determining the nature and capacity of their components and their operational strategy. Chapter 2 goes over the preliminary design stage of two HRES in British Columbian communities. Many modelling tools are available, ranging from full-factorial and linear optimization techniques that can solve single-objective problems, to meta-heuristic algorithms. One of the distinctions between different HRES modelling tools is the foresight horizon being used. Linear programming tools commonly have a perfect foresight over the typical year analysed, for both demand and natural resources. This can lead to an overly optimistic prediction of the lifecycle cost of a system when the reality of implementations comes with uncertainties. On the other hand, tools that use myopic foresight, or no knowledge of future parameters, can lead to pessimistic lifecycle cost estimates since the demand and power output of certain renewable energy technologies, like solar panels, can be known within a few hours. The purpose of Chapter 3 of this thesis is to guide readers towards the right tool in the context of energy system modelling for the preliminary design of HRES. It was found that the degree of importance of choosing the appropriate foresight approach is a function of renewable energy penetration, autocorrelation, and storage capacity. A system with a high renewable energy share, a low short-term (few hours) autocorrelation, and an optimal storage size will result in the highest NPC difference between the two methods. When planning for long-term HRES design, the choice of the foresight horizon can either be representative of a lower/upper cost boundary (perfect and myopic foresight respectively) or of the real-time predictability of the power output of the chosen renewable energy power source. The use of energy system modelling tools is often reserved for highly qualified personnel and is therefore costly for prospective communities. To improve community readiness with minimal investment, a simple alternative to energy system modelling is proposed in Chapter 4 for the integration of tidal stream turbines in British Columbia. A series of three logical conditions was demonstrated to inform on the viability of a project in terms of cost reduction in comparison to the business as usual scenario. These conditions were found to also be useful for determining the minimum scale, or the economic break-in scale, for a tidal stream turbine given a remote community. In this context, communities are found to be best described by the local price of diesel fuel as an easily accessible metric to represent the current cost of electricity, their electrical load scale, and the local tidal current resource. Ten British Columbian communities were selected to validate the results by comparing the set of conditions to a complete energy system modelling approach and four were found to reach savings of 10 % or more as compared to the business as usual scenario. The long-term objective of this work is to provide remote communities with an integrated, affordable, and turnkey solution for the displacement of diesel in their energy systems. The next steps in achieving this include augmented optimization tools to quantify and capture non-monetary value so that the modelling and multi-criteria decision-making steps of the design process can be bridged together. / Graduate

Co-location of offshore wind turbines at sites being developed for tidal stream arrays has been proposed as a method to increase capacity and potentially reduce the cost of electricity compared to operating either technology independently. This research evaluates the cost of energy based on capital expenditure and energy yield. It is found that, within the space required around a single 3 MW wind turbine, co-location provides a 10-16% cost saving compared to operating the same size tidal-only array without a wind turbine. Furthermore, for the same cost of electricity, a co-located farm could generate 20% more yield than a tidal-only array. These results are based on analysis of a case-study site in the Pentland Firth. Wind energy is assessed using an eddy viscosity wake model in OpenWind, with a 3 MW rated power curve and thrust coefficient from a Vestas V90 turbine. Three years of wind resource data is from the UK Met Office UK Variable (UKV) 1.5 km numerical model and corrected against a 400 m Weather Research and Forecasting (WRF) model run over the site. Tidal stream energy is modelled using a semi-empirical superposition of self-similar plane wakes, with a generic 1 MW rated power curve and thrust based on a full-scale, fixed-pitch turbine. Coincident tidal resource data is from the Forecasting Ocean Assimilation Model (FOAM) at 7.5 km resolution and correlated with a 150 m ADvanced CIRCulation model (ADCIRC). Wave parameters are corrected from ERA-Interim data with six months of wave buoy data. Multiple tidal turbine array layouts are considered, with maximum tidal energy generated for a staggered array with spacing of 20 tidal turbine diameters, Dt , streamwise and 1.5Dt cross-stream. However, cheapest cost of electricity from the tidal-only array, was found for a single row of turbines, due to minimal wake effects. Laboratory experiments were undertaken to validate the superposition wake model for use with large, shared support structures. Two rotors mounted either side of a central tower generate a peak wake velocity deficit 70% greater than predicted by superposition. This was due to high local blockage and a complex near-wake structure, with a corresponding increase in tower drag of 9%. Further experiments evaluated the impact of oblique inflow on turbines yawed at +/-15 degrees. These results validated a theoretical cosine correction for thrust coefficient and characterised the centreline wake drift with downstream distance. Extreme environmental loads for a shared support structure, compared to structures for wind-only and tidal-only, have also been modelled. A non-linear wave model was used to represent a single wave form with 1% occurrence for each hour of time-series data. Overturning moment about the base of a shared support, with one wind and two tidal turbines, was found to be 4.5% larger than for a wind-only turbine in strong current and with turbines in different operational states. Peak loads across the tidal array were found to vary by 2.5% and so little load reduction benefit could be gained by locating a shared support in a more sheltered area of the array.