Monday 25 March 2013

Why do investors want high nameplate power ratings?

There is a lot of focus on the nameplate power ratings of wave and tidal projects. A good example of this is Aquamarine’s choice of name for their second prototype. The ‘Oyster 800’ is not the most obvious sequel to the ‘Oyster 1’. The name does however draw attention to the 800kW power rating, which happens to be higher than the 750kW of competitor Pelamis. One could almost imagine that there was a race of some sort!

In a sense, there is a race. Both companies are courting the same limited investment pool. They are competing not only with each other, but with tidal energy too. One of the things many investors are very keen on, is big name-plate ratings.

The focus on power rating has interested me for some time, and I've had the opportunity to discuss this with a wide cross section of people in the business. Based on these discussions, here is my understanding of why there is a preference for large nameplate ratings:

  1. Technology maturity/readiness is one of the key ways that investors judge the risk and value of a marine energy start-up. Wave and tidal developers have been racing to be ready for investment. All the protocols on technology readiness levels (TRL) associate progressively higher TRLs with progressively larger scale prototypes.  
  2. Wave is not commercially mature, and the closest thing we've got to it is offshore wind. So as a starting point, it is somehow intuitive to imagine units with similar ratings, availabilities and capacity factors. At present, offshore wind is driving down costs by progressively upping the rating of turbines installed.
  3. As with offshore wind, there may be a cost of energy (CoE) advantage to having a smaller number of larger devices (lower balance of plant costs). We'll know better once we have sufficient operational data. In the absence of this data, it is nevertheless worth noting that most in big industry believe large nameplate ratings will bring down the CoE.
  4. Big industry will only invest in wave energy if the potential global market is big. This is, to my mind, the main reason for the focus on high power sites. The most cost effective device for high power sites will likely be larger, and have a larger power rating, than the optimum for lower power sites.
  5. For an unproven device, there is no certainty about anything much - annual energy , capacity factor etc. Nameplate rating is however something that you can be certain about - perhaps this explains the focus on rating?

The TRL protocols were never intended to be prescriptive. There has been a growing awareness that if developers and investors were to use the TRL protocols as road-maps, they could be missing something along the way. At the recent ICOE, two papers suggested additional criteria, for use along with measures of readiness, to guide development of wave energy technology. Weber suggested a parallel measure of performance; Fitzgerald noted that TRLs track functional readiness, and suggested a parallel measure of life-cycle readiness.

Recent talks given by wave energy developers and their industrial partners (e.g. RenewableUK wave and tidal 2013) suggest that investors are no longer impressed by claims of 'look at the peak powers we can do for 10 hrs a year'. The focus is now firmly on CoE. 

Pelamis is an interesting case in point. The P2 had the same rating as the P1, despite being larger – it had a higher capacity factor. The Pelamis website hints at CoE improvements. Pelamis probably have enough operational data by now to make a good judgement of what direction to move in for a more cost-effective capacity factor, so it will be interesting to see what the next generation Pelamis design looks like.

Image credits:
Behemoth rubber duck by Trey Ratcliff, part of Florentijn Hofman’s Rubber Duck project:


  1. Very interesting! I got thinking about this point on a public tour of Pelamis in 2007. Individual hydraulic motors on view (from a P1 machine?) were rated at 400kW for 6 seconds of every minute. The difficulty of bridging the gap between abstract rating/durability figures like this and real-world duty cycles starts to explain why wave power needs a long-term development programme.(Sadly I didn't see the generator nameplate.)

    In summary, the nameplate power, if it is truly the maximum power that the device can convert, might correlate with a couple of other relevant parameters:
    - Maximum stress (loading) (in the broadest possible sense) within the power conversion system/chain, from which can be derived design targets for components/subsystems
    - The necessary power rating of the power tranmission system to the grid (cable, transformer, switchgear)
    These are "sizing" factors that influence the direct manufacturing/installation cost because they are driven by "peak" power levels. But as you pointed out, CoE comes in to determine the true output, i.e. the MWh or £££ per year. It seems odd that investors were fixated on the peak power and neglected the CoE which has such a large influence on their payback. Some of the financial backers of wave power are actually large plc-grade institutions.

  2. I feel I ought to come to Pelamis's defence here in case someone else reads this out of context! By 'rated at 400kW for 6s of every minute' I presume you mean you saw some power traces from sea trials that hovered at 400kW for roughly 6s of every minute and gave less power the rest of the time? (stop me if I'm wrong...)

    The trace you saw could be 1 of 2 things: either the 1st stage, where joint motion is damped and energy extracted is stored in pressurised hydraulic fluid; or the 2nd stage, where captured energy is converted to electricity. If it was the first stage, then you observed the challenge facing every wave energy developer: the large gap between average and maximum powers. If it was the second stage, then the control system determined how long to stay on a given power output. The length of time spent at maximum power depends on how energetic the sea state was at the time and on operational details (e.g. was the control algorithm prioritising a smooth power output?) One trace of a few hours is not enough to judge capacity factor (ratio of average power generated when device is available for generation, to nameplate power) which is typically calculated for one year.

    Another thing to keep in mind is that Pelamis have a sort of snake gym program: they progressively up the sea states that they collect operational data in. Back in 2007 I'm guessing they would not have been testing in the types of seas where the mature device might generate rated power almost continuously.

    For clarity, when I say 'rating' or 'nameplate power' I am referring to the second stage.

  3. By the way, more comments on the Linked-In group that got me thinking about this topic in the first place:



Note: only a member of this blog may post a comment.