EGU General Assembly 2020
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

The value of tidal-stream energy resource to off-grid communities

Matt Lewis1, John Maskell2, Daniel Coles3, Michael Ridgill1, and Simon Neill1
Matt Lewis et al.
  • 1Bangor University, Centre of Applied Marine Science, School of Ocean Science, Bangor, United Kingdom of Great Britain and Northern Ireland (
  • 2JMcoastal Ltd, 10 Station Rd, Preston, Lancashire, UK
  • 3SIMEC Atlantis Energy, Edinburgh, UK

Tidal-stream energy research has often focused on the applicability of the resource to large electricity distribution networks, or reducing costs so it can compete with other renewables (such as offshore wind). Here we explore how tidal electricity may be worth the additional cost, as the quality and predictability of the electricity could be advantageous – especially to remote “off-grid” communities and industry.

The regular motion from astronomical forces allows the tide to be predicted far into the future, and therefore idealised scenarios of phasing tidal electricity supply to demand can be explored. A normalised tidal-stream turbine power curve, developed from published data on 15 devices, was developed. Tidal harmonics of a region, based on ocean model output, were used in conjunction with this normalised tidal-stream power curve, and predictions of yield and the timing of electricity supply were made. Such analysis allows the type and number of turbines needed for a specific community requirement, as well as a resource-led tidal turbine optimisation for a region. For example, with a simple M2 tide (12.42hour period) of 2m/s peak flow, which represents mean flow conditions, a rated turbine speed of 1.8m/s gives the highest yield-density of all likely turbine configurations (i.e. calculated from power density and so ignores turbine diameter), and with a 41% Capacity Factor. Furthermore, as tidal current and power predictions can be made, we explore the battery size needed for a given electricity demand timeseries (e.g. baseload, or offshore aquaculture). Our analysis finds tidal-stream energy could be much more useful than other forms of renewable energy to off-grid communities due to the predictability and persistence of the electricity supply. Moreover, our standardised power curve method will facilitate technical tidal energy resource assessment for any region.

How to cite: Lewis, M., Maskell, J., Coles, D., Ridgill, M., and Neill, S.: The value of tidal-stream energy resource to off-grid communities, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-2747,, 2020

Comments on the presentation

AC: Author Comment | CC: Community Comment | Report abuse

Presentation version 1 – uploaded on 30 Apr 2020
  • CC1: Comment on EGU2020-2747, Rory O'Hara Murray, 04 May 2020

    Looks like you've been raiding the ReDAPT database? Have you come across any good examples of successful off grid tidal turbines, or any good potential locations?

    • AC1: Reply to CC1, Matt Lewis, 05 May 2020

      Hi Rory, thanks for the question. I don't know of many offgrid tidal energy projects, but Aquatera have been busy in Indonesia... It would be good to find some to see what power demands are needed..

  • CC2: Comment on EGU2020-2747, Sam Fredriksson, 05 May 2020


    Thanks for a nice and interesting presentation. At slide 4 the voltage variability is shown. It is hard to see any dependencies to the plots above. Since I reason that the voltage variability is of importance, could you elaborate this a little bit?


    • AC2: Reply to CC2, Matt Lewis, 05 May 2020

      Hi Sam, what a good question! thanks for the interest. I must admit that I do not understand the electricity side of things (co-author Grazia helped here); but it is measured at the shoreside and is a function of supply and demand - not just demand. I guess this is why it spikes on accelerating and decelerating broad-scale power. The flicker part (fine-scale) is removed through decouplng of generator with grid using capacitors. This raises interesting questions of whether tidal turbines should be (system inertia questions) in small-scale systems - maybe?

      • CC3: Reply to AC2, Sam Fredriksson, 05 May 2020

        Hi Matt,

        Thanks for your answer. The peaks do not realy seem to link to typical demand peaks either. Maybe something to study further in later studies.

  • CC4: Comment on EGU2020-2747, Onno Bokhove, 05 May 2020
    Dear Matt,
    Since there were (too) many questions atm, mine per comment.
    In which paper do you consider the batteries for storage, per your last slide?
    Also curious to see how you model batteries in this context.
    Kind regards, Onno
    • AC3: Reply to CC4, Matt Lewis, 05 May 2020

      many thanks for your question.

      the battery slide was to suggest that the analysis could be applied to battery size and design studies (I think), because we can now predict tidal energy supply for multiple years at sub-second scale resolution. This could help researchers in battery design and fatigue (although I don't know, so hope someone out there would like to collaborate), but on the broader scales - if the demand time-series is known (I took gridwatch templar data and normalised to the peak demand), then we can look at how much supply a tidal energy farm could provide - and thus how much storage would be needed to redistribute the electricity for times when needed (because the tide in UK progresses by ~1hour per day). Therefore, and although we fall short of attempting to do so, if a client approached us with their likely energy demand for a year, we could look into which turbine, and the numbers needed, and the size of storage (battery or fly wheel for example)... hope that made sense?

  • CC5: Variation from power curve, Simon Waldman, 05 May 2020

    Hi Matt,

    Thanks, this is interesting. I'm especially curious about your point on slide 5 that "•2Hz measured power curve very different to “idealised” used in resource assessment". Do you have any insight into why this is? Thoughts that spring to mind are a turbine that isn't aligned with the flow, or perhaps inertia in the rotor preventing it from responding quickly to turbulent velocity changes on a short timescale - but I'm speculating.


    • AC4: Reply to CC5, Matt Lewis, 05 May 2020

      thanks for the question, my fault for not being clearer - sorry.

      In a resource assessment, I use 15min mean flow speeds and a power curve like the red line: cut in speed with power being a cube of velocity and a constant Cp value, up to a rated speed and power. however 0.5sec frequency data shows real power not like this, with turbulence (both speed and direction) changing the Cp shall we say.... So does this realistic flow variablilty affect resource? our work suggests not because it is variability about the mean flow (thus aggregates to no real change in resource), thus ocean resource models can keep using mean flow time-series and not worry... but the quality of power (i.e. effect to a consumer from all that variability) might be an issue. If it is, we can downscale our resource models to provide sub-second frequency power time-series for these kinds of researchers to look at... 


      Hope that answers your question?