When is a GW not a GW?

A report in the Guardian today included a rather curious statement. Covering a speech by Labour Party leader Ed Milliband at Whitelee Wind Farm near Glasgow, Damian Carrington claimed the following:

The windfarm, which spreads out across low, heather-clad hills, currently has 140 turbines and will add another 75 turbines soon, giving it a capacity equivalent to more than half a nuclear power station (about 550MW).

Now, it is true that the capacity of most nuclear power plants is somewhere around 1000 MW, however one would expect that a journalist who regularly writes about energy would understand that 550 MW of wind was not equivalent to 550 MW of nuclear.

So, how much of a 1 GW nuclear power plant would 550 MW of wind be equivalent to? For a start the (“nameplate”) capacity of a plant is its maximum output, not its average output.  Wind power, as everyone knows varies. The key issue here is capacity factor, the ratio of actual output against its “nameplate capacity.” The US for example has the world’s largest nuclear fleet and had an average capacity factor of 89% in 2011. (figures for other countries available here)

What about Whitelee Wind Farm? Based on this marketing leaflet for the extension of Whitelee, its current capacity factor is about 30%. So, instead of the wind farm’s capacity being the equivalent of half a nuclear power plant, as the Guardian claimed, it is the equivalent of less than a quarter.

None of this is anti-wind, it is simply pro-arithmetic.

About these ads

11 thoughts on “When is a GW not a GW?”

  1. I almost completely agree. I wish people would talk about actual electricity generated instead of “capacity”.

    I am very pro-nuclear but, like you, also very pro-arithmetic so one thing I might just say is that the UK has pretty woeful capacity factor for its nuclear plants which seems to hover between 50% – 65% (and it is the UK that is being talked about here)

    http://www.decc.gov.uk/assets/decc/11/stats/publications/dukes/5955-dukes-2012-chapter-5-electricity.pdf

    (table 5.10 on P. 144)

    The Guardian article is misleading, but maybe not as gratuitous as you think.

    Steve

    1. Steve

      I agree that you can argue about what nuclear capacity factor to use. The UK has very low current capacity factors by global standards. The real comparison should be on likely capacity factors for new build, where something between 80 and 90% would be a reasonable assumption. Though it may be less as they are likely to partially run as load following plants to accommodate high levels of renewables.

      The big problem with the Guardian article though is that the author seems oblivious to the fact that capacity does not tell the whole story. It is an embarrassing mistake for a journalist, but I have come to expect this of Damian Carrington, as he appears to make a living out of making daft pronouncements on energy.

    2. I agree. I once tried to do some digging, but I’ve never found a good explanation of why the capacity factor is so low for UK nuclear. As of earlier this year, we now have no Magnox plants, 7 AGRs, and 1 PWR. Is there something about AGRs that makes them poor in this respect?

      I also have tried to find out what is the standard capacity factor of Sizewell B. As it is the only Pressurised Water Reactor in the UK and as we are planning to build new PWRs, I thought its capacity factor would be a not uninteresting number. But I can’t find that information anywhere (Sizewell B doesn’t know, and DECC won’t tell). I’m guessing it is quite poor because otherwise they would shout about it. Sizewell B was closed for about 6 months in 2010.

      I am surprised to hear that new nuclear might run as load following plants, I would have thought that we are a such long long way away from the output of nuclear and renewables combined being above even night-time demand, that both would be on full blast all the time. Do you know where I could find any information about this?

      Regards Steve

      1. Good question about UK power plants. France actually provides production data on every plant in France. UK doesn’t appear to. Writing a post at the minute looking at load following in France.

        UK load following may or may not happen. Wind levels above 30% will result in occasions where wind is providing close to 100% of power. Currently it’s unclear if nukes will have to accommodate this by reducing output or if the wind just won’t all be fed in to the grid. More a question of the economics. However the planned EdF and Hitachi reactors can load follow. Both can go from 60 to 100%.

      2. “Wind levels above 30% will result in occasions where wind is providing close to 100% of power”

        this is,ironically, the one instance when “capacity” is a useful number. Max wind capacity (both onshore and offshore) was about 6GW in 2011 and nuclear about 9GW. If the lowest demand is 30GW, I would have thought we are ages away from this being an issue and requiring any load reduction of nuclear power plants.

        Regarding load following (also, I’m not sure if I’m using that right, “load following” is slightly different from just reducing load at night, I think), Rod Adams, who tweeted your blog this morning, had an interesting discussion of load-following a few weeks ago.

        http://atomicinsights.com/2012/11/atomic-show-190-nuclear-plant-performance-during-hurricane-sandy.html

        in it, the point was made that one of the reasons that US nukes have such high capacity factors is that they sometimes run when no one wants the electricity, and they actually pay people to take it. They do this because they get paid by how high their capacity factor is and by reducing the output they would also reduce their bonus.

        Regards Steve

      3. If you want detailed information about nuclear power plants, just one word : PRIS
        http://pris.iaea.org/PRIS/CountryStatistics/ReactorDetails.aspx?current=263

        Well opposite to EDF in France, you won’t get the load information for every quarter of an hour, but the yearly one should be all what you want.
        When comparing nuclear against renewable, it’s quite an error to just use directly the raw load factor. First nuclear can do load following which means that sometimes it does not produce at the max *because* the power is not needed. If they had a FIT, could sell power at fixed price all the time, and nobody would request a production reduction to help balance the grid, you’d see a slightly different number. Also it’s possible to choose when you close, for example EDF always closes the reactors for reloads in summer, when there’s less demand. This mean the unavailability that can be planned can be made to match seasonal demand variation. So whilst it looks bad on the load factor, it’s not that bad at all in practice.

        So in the case of Sizewell-B, the “Energy Availability Factor” represent the time where it could have generate at max load if it had been useful for the grid, and the difference between the “Operation Factor” and the “Energy Availability Factor” the time where it was really unexpectedly unable to generate power. And that number is really low. However we still don’t have the data about the times where it was shut down for planned maintenance / reload, and any occasion where it had to be shutdown to repair an unexpected problem. But since the operation factor is actually quite good, it seems there wasn’t many unexpected repair shutdowns.
        BTW since the reference power is slightly higher in 2011, the long 2010 shutdown was obviously for an uprate/systems enhancement.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s