How useful is tidal-stream energy for electricity supply?

Matt James Lewis1, James McNaughton2, Grazia Todeschini3, Michael Togneri3, Ian Masters3, Matthew Allmark4, Tim Stallard5, Simon P Neill6, Alice Goward Brown7 and Peter E Robins1, (1)Bangor University, Bangor, LL59, United Kingdom, (2)EDF Electricité de France, Paris, France, (3)Swansea University, Cardiff, United Kingdom, (4)Cardiff University, Cardiff, United Kingdom, (5)University of Manchester, Manchester, United Kingdom, (6)Bangor University, Bangor, United Kingdom, (7)Bangor University, School of Ocean Sciences, Menai Bridge, United Kingdom
Renewable energy could increase access to electricity, bringing communities out of fuel poverty, and provide grid-connected communities with a low carbon secure energy supply. However, temporal variability exists in non-thermal renewable energy sources, such as clouds for solar-PV and gusts for wind energy. The change to temporally variable renewable energy sources, from controllable and predictable thermal power stations, is a therefore a challenge for electricity grid system engineers. Due to the regular periodicity of the tide, tidal-stream energy is often stated as predictable. Here, we use various metrics to understand the temporal variability and asses the quality of tidal-stream electricity. Direct high-resolution measurements of power and electricity from a 1 MW tidal turbine (Orkney Islands, UK), and incoming hub-height tidal current velocity, were interpolated to a common timeseries at 0.5 Hz. Variability of shore-side measured voltage was found to be well within acceptable levels (∼0.3% at 0.5 Hz) and fine-scale temporal variability did not significantly affect resource yield estimates (<1%). Tidal power temporal variability, at sub 10-minute scales, was low (standard deviation 10–12% of rated power), with decreasing power variability for higher flow speeds. Fine-scale tidal current variability (i.e. turbulence) followed a normal distribution, and turbulence intensity levels reduced for higher mean flow speeds. Measured power variability was driven by this flow variability, allowing a statistical model to successfully predict power at 0.5Hz using broad-scale tidal resource model data. The synthetic power variability model reliably downscaled 30-minute tidal velocity simulations to power at 0.5 Hz (85% skill and 14% error), using a “t-location” distribution of observed fine-scale power variability in combination with an idealised power curve. The predictability and quality of tidal-stream energy was therefore considered high, which may be undervalued in future electricity system design. Moreover, tidal energy appears suitable for off-grid renewable energy systems as control measures and storage solutions can be designed to provide a firm supply, perhaps cheaper than some estimates of Levelized Cost of Energy would suggest.