Bathymetric effects on marine hydrokinetic array performance

Abstract

Approximately 16% of the globally generated electricity comes from conventional hydropower installations. Recent technological improvements in marine hydrokinetics (MHK), and a global demand for increased renewable energy, are enabling this technology to become a major contributor in the global energy market. MHK devices convert the kinetic energy, or energy of motion, from waves or water currents into electricity that is then transferred to the electrical grid. Wave energy converters (WECs) capitalize on the oscillatory motion of ocean waves, while current energy converters (CECs) use river, tidal, or ocean currents to generate electricity and often resemble wind turbines. Unlike wind, water currents are less intermittent, and, in the case of tidal currents, highly predictable. At present scales, individual CEC and WEC devices alone are not powerful enough to make hydrokinetic power economically feasible. Therefore, deployment of arrays of marine hydrokinetic devices is the most cost-effective method for these devices to become a major contributor in the energy market. In addition to device design and operational conditions, how these devices are deployed within a site determines their potential for power generation. The power generated depends upon interdevice proximity, where it is generally assumed more electricity is generated as the spacing between each device increases. However, most array performance studies of current-energy converters do not consider bed topography and are either inside smooth walled channels or deep, open waters. The research presented here explores the impact of site bathymetry on array performance since each deployment location is likely to have a significant impact on the optimality of an array layout. It is first shown that without boundary constraints the performance of an array does improve as the inter-device spacing is increased. Uniquely though, the normalized power and loading on an array is not transferable from an unconstrained domain to simple, sloping beds. The effects of this demonstrate the need to consider the topography of real world locations for general array designs.