How reliable are Britain’s wind farms?

Posted on Updated on

According to the Telegraph website, Drax (the owner of Britain’s largest biomass power plant) has attacked wind and solar power for their lack of reliability.

The Telegraph reports:

Dorothy Thompson, Drax chief executive, said the UK’s wind and solar farms were “increasing the instability of the grid”, with “mounting” cost implications, citing forecasts suggesting the bill for balancing supply and demand would rise from £1bn this year to £2bn by 2020.

Yet “aggregate wind farm and solar output in the UK can be expected to fall below 1pc of total electricity production with reasonable regularity”, she argued.

The industry lobby group Renewables UK, naturally, wasn’t too happy:

The comments prompted an angry response from wind industry body RenewableUK, which said Drax was “wrong to accuse wind of increasing instability on the grid” and “misleading to claim that wind output regularly falls to low levels”.

Now, this sounds like a factual matter worth looking at quantitatively. Does wind output regularly fall to low levels?

Given that Drax appear to have got their information about the reliability of wind farms from something I wrote perhaps I am in a position to answer.

Drax claims that wind farm output is expected to fall below 1% of electricity production with reasonable regularity. This is not what I said.

Instead I showed that aggregate wind farm output should be expected to fall below 1% of total installed capacity. This is actually a much stronger claim.

In effect, wind farms can be relied on for absolutely zero power. When electricity demand peaks we must be ready for wind farms to be producing more or less nothing. This is a simple fact, and no amount of hand waving and vague language from lobby groups can get around it. (Of course technological innovation in energy storage might change this, but that’s another story.)

The fact that wind farms can produce almost zero power is regularly denied. This is normally done by waving of hands. “If it’s not windy one place, it will be windy elsewhere”. Not so.

To see this consider what happened on the 16th June 2013. On that day maximum instantaneous wind farm output was 2722 MW.

That’s the maximum. The minimum was only 19 MW.

So, the maximum was 143 times greater than the minimum.

britain_bigDropBritain’s energy system must therefore account for the undeniable fact that wind farm output will go close to zero. Unless we want to risk blackouts every time this happens we will have to build vast amounts of back up capacity.

Without large scale storage, we effectively need to have an electricity grid that would function perfectly if the wind farms and solar panels were yanked out of it.

Note on data

Wind farm output data is easily available from Grid Watch. Official data is available from Elexon, however more processing is required to handle that data.

I produced the graph above in R, using ggplot2 and a custom theme.


24 thoughts on “How reliable are Britain’s wind farms?

    Joe Public said:
    July 30, 2015 at 1:25 am

    Very interesting.

    From mid-afternoon to past midnight, the UK’s entire fleet of (~5,500??) turbines provided less than 1% of UK demand. The nadir was when they were cumulatively generating just 0.06% of the UK’s total demand of 30,289 MW.


    Brendon said:
    July 30, 2015 at 3:38 am

    It appears from the Grid Watch data that on average across all 5-minute intervals in 2014, the UK was getting just over 7% of electricity needs from wind (median was about 6%). At the same time, pumped storage averaged 0.8%. So clearly for 7% wind, you don’t need vast amounts of storage, the troughs in wind production are being made up by other generating capacity and grid interconnections. And the troughs in renewable generation won’t be as low when more renewables (and more types) are added. So the question is what other grid capacity (and storage) do you need at higher levels of intermittent renewables: 10%, 50%, 80%, etc (and how do those renewables fill in each other’s troughs)? To me, “vast amounts of capacity” is not a satisfying answer.


      Robert Wilson said:
      July 30, 2015 at 7:24 am

      This is the waving of hands my post refers to.


        Brendon said:
        July 30, 2015 at 1:45 pm

        Nothing in my comment stated or implied that variable wind doesn’t need backup capacity or storage. The question I asked was how much is needed at high levels of renewable penetration. “Vast” is not an answer. And the answer is not 100% of renewable capacity, especially as more types of renewables are added. I’m interested in engaging in a dialogue, dismissive comments won’t enable that.


        Robert Wilson said:
        July 30, 2015 at 1:57 pm

        What are these types of renewables and how reliable are they at 7 pm in a sunless December when Britain’s electricity demand peaks?


      jbenton2013 said:
      July 30, 2015 at 9:54 am

      What utter nonsense. It doesn’t matter how much installed capacity you have on the system, if the wind isn’t blowing and/or the sun isn’t shining then there will be no output. Pretty obvious to even a small child.


    stefanthedenier said:
    July 30, 2015 at 8:35 am

    wind turbine produce low voltage electricity -> by the time goes to the main greed, most of electricity is lost in cable resistance – the main greed recessives enough electricity to run one transistor radio; one wind turbine to install costs over 5 million dollars. All that wasting money, for what?!:


      Robert Wilson said:
      July 30, 2015 at 8:38 am

      Indeed. We could have used that money to fund education in numeracy and critical thinking. Some people would have benefited.


    scienceofdoom said:
    July 30, 2015 at 11:23 am

    It seems that the more academic discussion on this topic is 100% in agreement – that backup is required for most wind power.

    Basically we can look at the statistics of wind across the generating region (output is proportional to wind speed cubed but maxes out at high speed and is “turned off” when wind speeds get too high). It’s found that some extended periods provide little energy. Some of these are unfortunately in times of maximum demand.

    The result is a “capacity credit”. In the UK’s case, based on a number of studies, adding 25GW of wind power (to 0.5GW) provides 4GW of capacity credit. That is, 25GW of wind power displaces 4GW of conventional power generation, not 25GW of conventional power. On the plus side you don’t need to run this “backup” most of the time. So there’s a cost and lots of studies have tried to quantify it. But the cost gets larger the more you increase the proportion of wind power in the grid.

    A little more in Renewable Energy I, and future articles will dig into the calculations in more detail.


      Robert Wilson said:
      July 30, 2015 at 11:28 am

      Advertising your own blog is not something I appreciate. Please stop doing it so openly.


        William said:
        July 30, 2015 at 8:59 pm

        The first comment, from GWPF, is a more blatant advertisement. A comment by a reader at ScienceOfDoom (SoD) links back to you, that is how I came to your site. SoD is well known for careful analysis of climate science.


      Andy Boston said:
      July 30, 2015 at 11:44 am

      16% capacity credit seems far too high to me. What you’re saying is that grid can depend to a very high level of certainty on getting at least 4GW of output from 25GW of wind at times of highest system stress, so much so that it can do without 4GW of highly dependable generation. I looked at this many years ago with some rigour whilst working for a utility and 6-8% was nearer the mark then. With more recent stats on periods of low output I suspect that was a significant over-estimate.


        scienceofdoom said:
        July 30, 2015 at 9:25 pm


        Was that for the UK? Each country has different values dependent on the correlation of wind speeds across the generation region (i.e., generally the larger the region the lower the correlation, or in layman’s terms at least some wind somewhere). Also the capacity credit depends on the actual penetration of wind, the excess in the grid of other energy, and the acceptable LOLP (loss of load probability).

        If it was the UK and the result was a lower value it isn’t surprising – there will be some assumptions somewhere in that 2003 National Grid study cited by the 2012 textbook (1st edition was 2007 and most studies in the 2012 edition were still pre-2007).


      scienceofdoom said:
      July 30, 2015 at 9:16 pm


      I apologize. I am new to your site.

      I posted graphs of the statistical calculation from a textbook in that article. If I give the textbook as a reference I don’t believe anyone else will be able to access it (unless they have academic access). Still, in future, I will post references to academic literature instead.


        Robert Wilson said:
        July 30, 2015 at 9:29 pm

        I have no problem with you linking to my blog. I have a problem with you basically posting comments here that are basically not overly subtle ads for your own blog


    edhoskins said:
    July 30, 2015 at 7:36 pm

    By 2014 European Union countries had invested approximately €1 trillion in large scale Renewable Energy installations according to Renewable Energy industry sources.

    This has provided a nameplate electrical generating capacity of about 216 Gigawatts, nominally about ~22% of the total European generation needs of about 1000 Gigawatts.

    The actual measured output by 2014 has been 38 Gigawatts or 3.8% of Europe’s electricity requirement, at a capacity factor of ~18% overall.

    However Renewable Energy production is dependent on the seasons, local weather conditions and the rotation of the earth, day and night.

    So the Renewable Energy contribution to the electricity supply grid is inevitably erratic, intermittent and non-dispatchable. It is therefore much less useful than dispatchable sources of electricity, which can be engaged whenever necessary to match demand and maintain grid stability. That 3.8% Renewable Energy contribution to the grid is often not available when needed and obversely its mandatory use can cause major grid disruption if the Renewable Energy contribution is suddenly over abundant.

    Accounting for capacity factors the capital cost of these Renewable Energy installations has been about €29billion / Gigawatt. That capital cost should be compared with conventional gas-fired electricity generation costing about €1billion / Gigawatt.

    The whole 1000 Gigawatt fleet of European electricity generation installations could have been replaced with lower capital cost Gas-fired installations for the €1trillion of capital costs already expended on Renewable Energy in Europe.

    In spite of their being virtually no costs for fuel, Renewable Energy installations can still cost up to 2 – 5 times as much to operate and maintain as conventional Gas Fired plant.

    When the capacity percentages are taken into account, the capital costs can be 15 – 50 times greater than Gas Fired plant. To date about € trillion (€1000,000,000,000) has been spent on the installation of Renewable Energy technologies in Europe.

    By Government and EU diktat, these costs have been extracted by extra charges imposed on utility bills throughout Europe. This is a very regressive form of taxation imposing more burdens on poorer people whilst leaving wealthier people who are able to pay less affected. It is also invisible in Government balance sheet as an industry imposition on consumers.

    These regressive “Green taxes” have already lead to significant fuel poverty throughout Europe.

    Increased energy costs are also impacting on European industries with many major corporations seeking more congenial manufacturing locations outside Europe to the detriment of the European economies.

    It is also questionable whether the Renewable Energy industry, when viewed “from cradle to grave”, including manufacturing, installation, connection and demolition effects, does in fact reduce CO2 emissions to any significant extent. The CO2 saved may never exceed the CO2 emissions generated to erect the total installation, and at a maximum the use of Renewable technologies only amounts to about a 4% saving in CO2 emissions,

    The USA has made significant CO2 emissions reductions over the past few decades by replacing Coal Fired generation with Gas Fired electricity generation with the feedstock provided by the fracking revolution. It is estimated that using natural gas for electricity generation as opposed coal burning saves about 30% of CO2 emissions. This effect in the USA has been assessed to have been more effective means of CO2 emissions reduction than all actions worldwide arising from the Kyoto protocol.

    The Renewable Energy industry could not exist without the Government mandated subsidies and preferential tariffs on which it depends. Therefore it is not a viable business proposition

    Viewed from the point of view of the viability of the nation’s electrical grid, Renewable Energy would never be part of the generating mix without its Government mandate and Government market interference.


      Robert Wilson said:
      July 30, 2015 at 7:38 pm

      Thanks for that rather long list of dubious assertions.

      Can you provide evidence that renewables cost more to operate than gas fired power plants? That’s a daft statement, so I look forward to being entertained by your evidence.


        edhoskins said:
        July 31, 2015 at 8:07 am

        just do the sums based on US EIA and Eurobserver (renewable supporter) data


        Robert Wilson said:
        July 31, 2015 at 8:20 am

        I ask you for evidence and you give me a homework assignment. But obviously you don’t have evidence to back up your dumb assertion.


    donoughshanahan said:
    July 31, 2015 at 11:45 am

    You are missing the point when you say

    “These examples show that we should never rely on one form of power generation or very few very large plants because it makes you vulnerable to systemic flaws. It does not only apply to windpower. ”

    As has been shown large areas of low or no wind can occur over large areas (see previous posts for e.g. on California wind etc). That means that most or all of the wind turbines will produce little power. It is a common (non independent) mode of failure.

    Now two coal power plants situated say 100 miles away will have few common modes of failure. Maybe drought, storms or no coal supply or something like that but it is not common. What that means is you can build both coal plants with some amount of overcapacity and get away with losing one. The more you have, the smaller the overcapacity. Essentially their modes of failure are independent of each other.

    But because wind turbines have a common and regular mode of failure, overcapacity does not help.

    Been around for years now and not yet out of virtual phase.


    Robert Wilson said:
    July 31, 2015 at 11:45 am

    Hand waving and sophistry of the highest order. The grid is designed to be able to handle individual generators going offline. This is completely different to wind farms, which go offline in bulk.


    donoughshanahan said:
    July 31, 2015 at 2:40 pm

    If you had read my comment, it would be clear. It is unlikely that any given storm or Jellyfish will knock out two plants simultaneously and it is rare. Indeed your links are to single events.

    However little to no wind for example across a wide area is reliably common. Hence the low capacity factor for wind.


      Robert Wilson said:
      July 31, 2015 at 3:50 pm

      ontspan’s comments have been removed. His behaviour was suspicious, plus WordPress was spamming some of them automatically which suggests he has been hassling bloggers a lot.


Comments are closed.