|Theme||Visible||Selectable||Appearance||Zoom Range (now: 0)|
Results obtained from numerical modelling of wave and tidal currents and the resulting turbulence parameters at tidal energy sites in the Fall of Warness, which is a region consented by the Crown Estate for deployment of tidal stream devices in the Orkney Islands, Scotland, are reported in this paper. The software suite MIKE 21/3, which is a coupled wave-tidal flow model, has been used for this purpose. The coupled wave-current model is driven by boundary inputs of spatially and temporally varying wind, wave and tidal elevations. Turbulence closure is achieved using a two-equation (k-ε) turbulence model. The coupled model has been calibrated and validated with field measurements of waves and tidal currents acquired by Acoustic Doppler and Current Profilers (ADCPs) deployed in the Fall of Warness. The results indicate that the coupled model works well and the predicted wave-current parameters provide very good match to site measurements at different depths of the water column. The model outputs such as significant wave height, peak wave period, mean wave direction, tidal current speed and its direction, Turbulent Kinetic Energy (TKE) and its dissipation rate and Eddy viscosities are presented and discussed. These parameters will find its use in the design of tidal turbine components and its supporting structures.
Hydrodynamic loads on both fixed and floating tidal stream turbine components, e.g., rotor, supporting structures and moorings etc, need to be carefully determined when the machines are designed to operate in conditions where both waves and tidal currents co-exists. Unsteady flow due to turbulence, wave-current interactions, and variation of flow characteristics with depth can cause unsteady blade loading, resulting in fatigue. High-end computational modelling tools such as CFD software packages may have the ability to simulate wave-current interactions, however, its application to realistic site conditions with complex bathymetry and large computational domains covering several square kilometres, may not be feasible due to computational expenses and it may not even fully reproduce wave-current scenarios, especially when the interaction of directional waves, currents, and turbulences are to be modelled. It is well established that wave loads combined with different flow turbulent intensities, originating from wave-current-turbulent interactions, are the main contributors to fatigue failures of turbine blades. Wave-current induced turbulence affects tidal turbine power production in several ways, specifically through power performance effects, impacts on turbine loads, fatigue and wake effects, and noise propagation; it is therefore important to develop an enhanced understanding of wave-current-turbulence interactions in tidal energy research.
X-band radar provides a spatial backscatter results over a large area. This shows wave features to be clearly visible over a large area providing an advantage over standard in situ measurements. This paper suggests a new method of quantifying surface elevation in an almost real-time method by applying a second order Stokes waves in shadow regions (troughs). The initial results show the artificial wave trough method having an improvement in phase and magnitude when compared to independent in situ measurements. This method provides a better representation of the surface elevation. Once refined, the real time surface elevation can be used as boundary conditions for a short-term wave-forecasting model.
The use of X-band radar as a method of wave and current measurements has been around for a number for years (Alpers & Hasselmann, 1982; Young, Rosenthal, & Ziemer, 1985). This is due to the interaction between the sea's surface and electromagnetic wave that allows wave images to be collected over an area of several kilometers in all directions. Previously, these images were referred to as sea clutter as they are a byproduct of a ships navigation radar. A high-resolution spatial and temporal map of wave condition and surface currents can be created when an X-band radar is optimized to receive this clutter information.
The applications of X band radar to measure waves and currents is common, however, in a few cases this has been extended measure bathymetry. (Ludeno et al., 2015; Tenthof van Noorden, 2015; Trizna, 2001) and uses a dispersion relation filter to track the change in wave profiles as then interact with the seabed.
A large number of studies had originally focused on the analysis of radar data to extract wave spectra (Borge, Hessner, Jarabo-Amores, & de la Mata-Moya, 2008; Gangeskar, 2000; J. C. Nieto Borge, Reichert, & Dittmer, 1999; Seemann, Ziemer, & Senet, 1997). This normally uses the Signal to Noise Ratio (SNR) and a three dimensional dispersion relation filter with additional measurement coming from in situ wave sensors i.e. wave buoy or a ship's inertial measurement unit (IMU). This provides a good agreement when compared against standard wave measuring sensors. The calculated spectra however, is only capable to produce an output over a given period of time and for a given area. This provides good phase-averaged wave quantification but omits phase-resolving feature that are important for short-term wave predictions. In terms of resource assessment, the radar-derived spectra provides a low spatial and temporal resolution map, this remains a substantial advantage when compared to traditional 1-dimensional wave recordings from fixed locations.
A new method for simulating a frequency independent absorption within DHI’s Mike 21 Boussinesq wave (BW) model is presented. This provides an increase in the accuracy of the simulation of wave processes around a hypothetical WEC array. Multiple monochromatic wave simulations are combined to represent a wave spectrum. Wave device characteristics are then simulated using porosity layers. A frequency dependent porosity for each device is then applied based on data taken from an experimental study. This method is tested for nearshore shallow water devices where the wave energy disturbance is quantified for flat and varying domain gradients.