The Future of Wave Energy Devices

The future of wave energy devices in the UK is largely driven, at present, by the direction taken by Wave Energy Scotland (WES - link). Device developers utilise the support and resources available from WES to take their devices to the next level. WES is a publicly funded body dedicated to the development of wave energy – they use specific funding schemes to support the development of novel wave energy converters, power take-off systems, structural materials and control systems. Within each of these calls is a staged approach to technology development, with the objective of proving that a technology is suitable for wave energy, offers a potential reduction in LCOE and can be significantly de-risked to attract external investment.

Recent WES landscaping studies have illustrated the way in which they are working on their ultimate goal of getting wave energy to become a competitive form of energy generation in the UK market. Landscaping studies on mooring solutions and electrical connections were designed to address issues with device CAPEX and try and form a unified solution. While other studies on alternative generation technologies and very large scale WECs highlight the importance of PTO integration and the potential for multi-MW devices.

Ocean Energy Systems' (OES) have assembled data from many different developers and therefore many different device types into a single report (ref), so it serves as a guide rather than something applicable to all technologies. The information presented in the figure (below) shows that, at present, device costs and OPEX dominate the costs for wave energy devices. However, there are other important factors (grid connection, moorings & foundations) that contribute to project costs and taking steps to address this is a logical move for WES, as 15% of project cost is not insignificant. It is also important to note that better mooring design can increase performance of wave energy devices as well as reducing cost - providing a boost to LCOE.

OPEX costs in the early phases of wave development are a significant proportion of the project cost since device reliability is still being developed and understood. The unfortunate 'catch-22' is that a lot of reliability development cannot be done with desktop studies, it requires (often expensive) specialised lab testing or field testing. And while specific test facilities do exist (DMaC (link) at the University of Exeter for example), field testing is often far more beneficial. Field testing opens up the opportunity to capture issues that so often arise at interfaces, that would not be captured by component testing and also integrates the full marine environment, removing any of those 'unknown unknowns' that had not been identified in the development process. We appreciate this is a costly process, but getting devices deployed and testing is going to be the most effective way to improve reliability and start to bring down OPEX.

It should also be noted that the OPEX costs for wave energy devices are not fully understood as there are no long term WECs out there to provide data to validate modelling and assumptions.

WEC Cost Breakdown

Data Source: International Levelised Cost of Energy for Ocean Energy Technologies - OES

Alternative Technologies

Image Source:

Alternative Generation Technologies

The inconsistent oscillatory nature of wave motion means that it is not ideally suited to the most common form of electrical energy extraction used the world over - the rotational magnetic generator. These devices are more comfortable running at a fairly high constant speed - hence the application in coal and gas fired power stations and (often via a gearbox) wind turbines. With the oscillatory motion of the waves, there is likely a more economical use of material and technology to convert the forces and motion observed by a device into electrical energy.

Moreover, there are many different WEC designs, operating in different ways. Some will look to extract work primarily from the wave forces with little displacement, while others will focus on using strain to extract useful work from the waves. The figure presented is for illustration purposes but illustrates the different device types and depicts a means to identify suitable technology matches to wave energy converters (based on their force-strain characteristics).

Finding a powertrain more suited to the forces and motion of a wave device could offer vast improvements to performance and, if a route to market exists, cost reductions too. Without undertaking a detailed study of our own, we feel there may be some potential in di-electric materials used as flexible membranes (for example in OWC or bulge wave devices); however, these materials tend to output high voltage when operating at the scale of ~100kW and the power electronics to result in grid compliant power is quite bespoke and therefore potentially costly.

In summary, on the face of it whilst conventional rotational generators are not ideally suited for the motion of a wave device, alternative hydraulic and pneumatic solutions do not offer huge cost savings at present. So, is it worthwhile looking for some alternatives? Yes, there is no harm in trying to identify technology used in other industries or under development at universities that has the potential to be applied to wave energy. It is a big landscape to explore thus will require a broad, collaborative approach and a means to quickly quantify the potential of a technology. It will certainly be interesting to see the results of the WES landscaping study, where more time and consideration can be put into the different technologies and their application to wave energy.

Very Large Scale Devices

Large scale generation of 10MW+ is the direction that wind energy has taken - maximising the power out per foundation - should this be the direction explored by wave energy?

Starting with the basics, if a single device is used to capture 10MW of power, this has to be achievable in common (but sizeable) sea-states, let's say 100kW/m. To get 10MW from this sea state requires a capture width of 100m before accounting for conversion losses, so let's say 200m.

By a well-known theorem in the wave energy literature (equation 9.26 in "The Applied Dynamics of Ocean Surface Waves" [Singapore: World Scientific], C.C.Mei, 1989), capture width is related to the far wave field produced when the device moves in still water: it is the power radiated in the direction of the waves to be captured, divided by that lost in all other directions. Hence an axisymmetric device (e.g. a heaving buoy), with directionally-uniform wave radiation in still water, can have a capture width of no more than (wavelength/2π). This is the well-known point-absorber theorem, which is normally quoted as allowing a very small axisymmetric device to reach this limit, but at the same time, it is a limit on the capture width of very large axisymmetric devices.

A device with dipole-like wave radiation in still water (e.g. a pitching buoy), can have a capture width of no more than (wavelength/π). Since the wavelength of a 100 kW/m sea-state is approximately 100m, this rules out a very large class of WECs from further consideration as the required power simply cannot be generated from a feasibly sized device.

To achieve the required captured width of 200m, the WEC must be able, in still water, to focus a beam of waves in the direction back towards the waves to be captured. A single, very long un-segmented Salter’s Duck, for example, is capable of producing such a narrow beam, but only in one direction. In all wave sites, wave heading is always spread over a range of directions simultaneously, typically 30 degrees. This consideration rules out any WEC which is a single rigid body.

Pelamis Wave Energy Device

Image Source: Subsea World News

Large Wave Energy
Carnegie CETO Point Absorber WEC

Image Source: Carnegie Clean Energy

What remains are WECs which are multiple bodies, and in the first instance it is useful to restrict attention to multiple bodies in a line. There are two possibilities:

  1. A line parallel to the incoming wave crests, such as a segmented Salter’s Duck
  2. A line at right angles to the incoming waves, such as an attenuator device. There is a theoretical limit to this configuration, which arises as the forward bodies start to shield the rearward ones from incoming wave energy. There is an approximate analytical formula for this effect in the literature e.g. a 600m long device producing 500 kW in waves of 3.5m Hs is losing about 30% power at its tail end, through shielding. However, this is for waves exactly head-on – the losses reduce rapidly with wave heading, being negligible for waves 15 degrees or more from head-on. Such a heading could in principle be achieved with a secondary tail mooring line (as fitted to Pelamis, for example), so the attenuator configuration is theoretically capable of producing 10MW, if made sufficiently long.
  3. The multiple bodies need not be in line, but can form a 2-D sheet, or network, on the water surface, with a common PTO. This configuration can conveniently be considered as formed originally from a line of devices, that has repeatedly been cut in two and doubled up. If the original line was broadside to the waves, such doubling-up will clearly introduce shielding of the lines that are now placed down-wave, and so reduce power capture. If the original line was at right angles (or some lesser angle) to the wave crests, the tail of the device will see increased shielding at each doubling-up, because part of the device has moved closer to it, and thus shields it more effectively. Thus 2-D sheets or networks are inferior to lines, from a power output point of view.

To summarise, theory suggests that there are very few devices operating at the waterline that can offer 10MW from a single device. Devices below the waterline start to sacrifice some of the energy in the wave and thus will need to be even larger to achieve such large power ratings, but have not been studied in great detail here.  This is a completely theoretical analysis and does not detract from the attractiveness of multi-MW arrays of separate devices, but does propose some limits on single device types which is perhaps not so applicable to other forms of offshore renewables (wind and tidal). For the time being wave energy has shifted somewhat to focus on the deployment of smaller devices to maintain the learning rate, improve key areas such as reliability and survivability, without the need for very large investment. These devices are targeting more niche applications such as fish farms and offshore platforms where renewable energy can be offered in comparison to diesel generators.

Of course, the energy market may change with likely increasing prices of fossil fuels, increasing demand and the UK grid coming within 2% of capacity in recent winters. If the lights went off, the demand for all forms of renewable sources would increase and wave energy needs to be de-risked and ready to provide electricity to the UK grid

2 comments on “The Future of Wave Energy Devices Is large scale generation possible?Add yours →

  1. Courageous analysis but may have missed the mark on 2 key points (WEC suitability of direct drive rotary PTOs and scale economies of non-axisymmetric WECs);
    1. DIRECT DRIVE ROTARY ELECTRIC PTOs: The report states that the “inconsistent oscillatory nature of wave energy is unsuited to rotary PM generators”. Ultimately, all WEC power will most likely be generated by rotary generators (probably with PMs), even if an intermediate PTO step (adding cost and reducing efficiency) is utilized. Linear electric generators are unlikely to see large scale WEC use due to poor utilization of expensive PMs, and inherent air gap and “end-stop” problems”. The CorPower HiDrive/WaveBoost (multi pinion w linear rack gear) is certainly a better solution than linear generators, and uses rotary generators. WECs with rotary vs linear PTO inputs have an inherent advantage by not having to make the linear to rotary conversion at some stage. The CorPower HiBoost can certainly be used with rotary input PTO WECs simply by replacing their circular rack gear with a linear one but even this system is probably over-kill for rotary input PTO WECs which can utilize today’s high torque (20 mmNm), low RPM (10-20 RPM) wind turbine PM generators as is (without rotational rectification) with or without single stage gearboxes. Power smoothing and conditioning can best be done with modern power electronics.
    2. MULTI-MW SCALES FOR NON-AXISYMMETRIC WECs: We agree with your conclusion that there is a theoretical maximum capture width for axisymmetric WECs limiting practical capture width and scale to 1/2 MW (or less). This conclusion is moot, however, because axisymmetric WECs have negative economies of scale (float area or volume and hence CAPEX increase exponentially with buoy diameter), so why even consider MW+ scale axisymmetric WECs??

Leave a Reply

Your email address will not be published. Required fields are marked *