| Résumé | In order to take advantage of economies of scale, hydrokinetic energy (HKE) developers typically deploy multiple turbines within rivers or the marine environment in array or farm configurations. Successful planning and design of turbine array deployments requires an understanding of turbine wake hydrodynamics, wake interactions within arrays, and the performance of turbines within array fields, to enable quantification of the extractable power and impacts on the surrounding environment. However, consistent and reliable methods for predicting the power generation capabilities of turbine arrays and the total extractable power from a given site remain elusive. Numerical hydrodynamic models show considerable promise as tools to support hydrokinetic energy resource assessment, turbine array site selection, array design and impact assessment. For example, Computational Fluid Dynamics (CFD) models provide a means to analyse the high frequency motions and complex geometries associated with turbine-fluid interactions at the scale of individual turbines. CFD models can be integrated with numerical models that solve free surface flow equations to study the interactions between turbine arrays and hydrodynamics at coastal region or river reach scales. However, numerical models remain subject to limitations and require calibration and validation to provide confidence in their predictive capabilities, and to quantify uncertainty. This paper presents preliminary work whereby a set of large scale physical models was utilized to document the magnitude and spatial distribution of the velocity deficit at a high resolution upstream, and within the downstream wake, of simplified representations of cross-flow turbines modelled as porous rectangular plates. Subsequent phases of the research will include using this experimental data to calibrate and validate a CFD model of the porous plates, and conducting scale model experiments using more realistic, moving cross-flow hydrokinetic turbines to support improvements in CFD modelling techniques for HKE applications. |
|---|
