The predictability of and variation in wind energy

Wind power is like Ginger Rogers. You know what I mean. It isn’t judged by the same standards as other kinds of electricity generation.

I’m speaking specifically of the reliability of, or variation in, wind over time. Many people live in places where they personally experience highly variable wind, or at least, think they do, so it is easy to assume that wind generators would be sometimes running on full, sometimes standing still, in a more or less random and unpredictable way, but this is not necessarily true. There are regions where wind is much more consistent than people might imagine, though of course it is always somewhat variable. In fact, a bigger problem with wind may be not so much the variation, but the fact that in some regions it is out of sync with energy demands. In some temperate zones, wind may be weak during the day but stronger at night when electricity demands are low (but this can vary from region to region, and seasonally; there are places where winds tend to come up during the day and calm down at night)

The thing is, we often hear about how wind varies and thus wind can’t make up more than a certain percentage of our energy production. But this statement, while probably partly true until much better batteries are invented, ignores two things: 1) other forms of electricity generation also vary; and 2) we may have to live, in the future, with variable amounts of electricity. Oh, and 3) we can probably fix this all with a DC smart grid, but I’ll leave that for another time.


Regarding this first point, let’s compare wind to nuclear. A nuclear power plant like the one nearest me gets turned off now and then to replace fuel rods. This is not too often, and in theory this is done on a highly predictable schedule, but it is still a large amount of variation. That power plant, I assume, at best goes from normal high capacity to zero capacity for a while then back to high capacity. While wind may vary unpredictably, investing the cost of one smallish nuclear power plant in one small nuke plant vs. dozens of wind farms distributed across a very large area of the Dakotas and Minnesota will get you less difference in variation over time than one might expect.

As I noted, this variation in generation by a nuclear power plant is very predictable, so it doesn’t really count as variation. Variation you know is going to happen can be planned for. Predictability is the problem. But, that happens too with nuclear power plants. Whenever there is a tornado warning at the aforementioned plant, it gets powered down. That is not predictable. All sorts of other things happen at nuke plants that cause them to be powered down to some extent as well.

The point here is simple: When considering alternative forms of energy generation, it is not fair to say “wind is unpredictable and variable” and leave it at that, because it is not the case that all other forms of energy are perfectly “predictable and non-variable.”

I won’t say much about the second point, but it is obviously true. You can’t rely on being able to do whatever you want whenever you want it with whatever technology you want if we live in a world where doing these three things is causing us to lumber steadily towards self-inflicted Armageddon. Technology is not going to fix everything. We will have to make some adjustments. Please expect that and start growing accustomed to the idea now.

There is another form of variability that I want to mention: Catastrophic failure. If you build a whole bunch of wind farms across the Dakotas and Minnesota and spend the same amount on a small nuke plant somewhere in that region, catastrophic failure will have different patterns. Really bad storms, a rare earthquake, or terrorism (Canadians sneaking over the border and blowing up our stuff, presumably) will not take out all, or even most, of the wind farms, but a total “melt down” figuratively or literally, of the nuke plant will put the long term kibosh on that method of making electricity.

It is very difficult or even impossible to figure in the likelihood of catastrophic failure of any of these systems, but we can compare the form of such failure, and the worst case scenario for hundreds of wind generators distributed over three states is very different than the worst case scenario for a single nuclear power plant.

A few facts that might help frame this discussion in proper scale:

The annual generation of power at the aforementioned nuclear power plant is said to be 4,192 GWh. The plant is rated at 610 MW. That is the smaller of two plants in Minnesota, the other with an annual generation of 8,914 GWh (two 548 MW reactors).

A wind turbine may have a capacity anywhere from about 1.5 MW to over 7 MW (the latter are very rare) with the largest in design but not yet built closer to 10. Realistically, it would take about 600 to 1000 wind generators to equal this nuclear power plant. In Minnesota, the total wind energy capacity is about 2196 MW with about 10% more being constructed. This translates into 657 GWh annually, compared to 13,106 GWh annually for nuclear. The current plan in Minnesota is to more than triple wind capacity by 2021.

All the above numbers are based on capacities and estimates that may not fully add up. Another way of making the comparison is to look at recent estimates of the total percentage of energy generated by various sources statewide. At present, about 57% of our energy is from coal, 23% from nuclear, 9.4% from wind, 4.8% from natural gas and the remaining few percent from other sources. That nearly a fourth of our energy is produced by three nuclear reactors at two plants is both impressive (that’s a lot of electricity!) and somewhat concerning (that’s a lot of pressure to keep the plants working, and a big problem if one breaks).

Sources: Minnesota Energy Facts, and Wikipedia.

Share and Enjoy:
  • Twitter
  • StumbleUpon
  • Facebook
  • Digg
  • Yahoo! Buzz
  • Google Bookmarks
  • LinkedIn

19 thoughts on “The predictability of and variation in wind energy

  1. The annual generation of power at the aforementioned nuclear power plant is said to be 4,192 GW/h.

    Pedantic unit check. Are you sure it’s not GWh (total energy produced)? That’s how annual generation is usually measured (and the numbers work out to the plants working at about 78% capacity).

    As I said, this is just me being a pedant.

  2. There is yet another way of coping with energy supply fluctuations: Build large energy users (e.g. aluminium plants) at more than the needed capacity, and switch them on and off depending on how much energy is available (so that the average Al output stays the same). The additional cost (larger capital investment) is to be compensated by paying less for electricity (buying it only when it is cheap). Stuff like this is AFAIK being discussed here in .de, but I don’t know the results.

  3. Greg,

    To add to your ‘unrelaible wind’ point:

    â??According to the Global Wind Energy Council, some 50 countries are home to more than 10 MW of installed wind power capacity, compared to 30 countries operating commercial nuclear reactors.â?

    But nuclear is needed in order to provide baseload power for renewables, right?

    In response to that, this is a great quote from Siegmar Gabriel, then-Federal Environment Minister of German:

    â??If someone declares publicly that nuclear power would be needed in the baseload because of fluctuating energy from wind or sun in the grid, he has either not understood how an electricity grid or a nuclear power plant operates, or he consciously lies to the public.â?

    Source: Clean Technica

    Repeat cut and paste from my comment on your last Fukushima update (but apropos):

    “According to the most recent issue of the “Monthly Energy Review” by the U.S. Energy Information Administration (EIA), renewable energy has passed a milestone as domestic production is now greater than that of nuclear power and is closing in on oil.

    During the first quarter of 2011, renewable energy sources (biomass/biofuels, geothermal, solar, water, wind) provided 2.245 quadrillion Btus of energy or 11.73 percent of U.S. energy production. More significantly, energy production from renewable energy sources in 2011 was 5.65 percent more than that from nuclear power, which provided 2.125 quadrillion Btus and has remained largely unchanged in recent years. Energy from renewable sources is now 77.15 percent of that from domestic crude oil production, with the gap closing rapidly…

    Among the renewable energy sources, biomass/biofuels accounted for 48.06 percent, hydropower for 35.41 percent, wind for 12.87 percent, geothermal for 2.45 percent, and solar for 1.16 percent.”


    One way nuke apologists try to make nukes seem indispensible is by comparing nuke output to the output of renewable sources (and reduced consumption, and smart grid efficiencies) individually, and then waving their hands, presto changeo, ala-kazam– ‘We can’t replace nukes with any ONE of those!’ The arguments for petroleum and natural gas are similar.

    Of course, as rational consumers, we get to select “All of the Above” from the list of renewable alternatives, and reduced usage. Not one watt from nukes necessary.

    In fact, if we diverted all the nuke, coal and oil subsidies to renewables– think at least 12 billion *A YEAR– and the use of nukes and carbon for electricity generation goes away in under five years:

    “Whereas the government spends around $8 billion a year on oil subsidies and $3.3 billion on nuclear power subsidies, it only spends $400 million on incentives for all renewable power combined, said Pfund…

    Over the lifetime of oil, coal and nuclear, government subsidies have been exponentially higher cumulatively than they have been for the nascent renewables sector. As a percentage of the federal budget, subsidies for oil and gas have always been at least 25% more than the support for renewables, and at times as much as 10 times more.”

    and furthermore…

    “The cost of wind power generation has dropped to the point that onshore wind is now competitive with coal-fired electrical power in certain regions, according to a report by Bloomberg New Energy Finance.

    “The latest edition of our Wind Turbine Price Index shows wind continuing to become a competitive source of large-scale power,” Michael Liebreich, chief executive of Bloomberg New Energy Finance, said in a statement.

    The Bloomberg report, which analyzed confidential data provided by 28 major purchasers of wind turbines, found that expanding scale, improved efficiency and over-capacity among wind energy hardware manufacturers have combined to push the average price of onshore wind turbines below one million Euros per megawatt.

    “For the past few years, wind turbine costs went up due to rising demand around the world and the increasing price of steel. Behind the scenes wind manufacturers were reducing their costs, and now we are seeing just how cheap wind energy can be when overcapacity in the supply chain works its way through to developers,” Liebreich explained.

    This news comes on the heels of reports that a Southern California utility is preparing to buy solar power for less money than gas-fired electric power over the life of a 20-year contract.”

  4. After the issue when the wind died a few years ago leading to blackouts ERCOT (the Texas grid) hired a couple of meteorologists and now does routine forecasts of wind (you can see them at the grid information link on the site. If you go to the board of directors link you can see monthly reports on the grid operations report on how good they do. For example I just checked the real time status and saw 7160 average wind output in the last hour on a demand of 29543 gw, yielding a 24% wind penetration. (I suspect transmission constraints are limiting the last 1000 gw of turbines).
    The information online there puts the lie to some of the anti-wind folks. Yes there used to be problems, but they figured out how to avoid them.
    Solar in Tx actually works pretty well as a summer peaking source, as hot days have a lot of sunshine in the summer (max load from all the A/C)

  5. Current large-scale nuclear is not variable enough. Shut it down and it stays hot for days/weeks. They lose money every night for lack of load. We already have large-scale (building-sized) molten salt batteries that can smooth out wind production. Mr. Muschall’s comment(#3) is on the right track. Aluminum metal is encapsulated energy. The real problem is that all this is not economic with NatGas suddenly being so cheap. How long will that last?

  6. Can you check again the actual energy production for wind in your closing paragraphs there, Greg:

    – Installed wind is 2196 MW
    – Delivered energy from wind is 657 GWh

    That calculates out at about 3.4% average power (compared to peak production), which is abysmally small. At that return wind would be completely uneconomic.

    In response to Roland: if you invoke storage to solve wind generation variability, in fairness you should also allow this to reschedule overnight nuclear production – especially when you’re talking about heat storage.

    On Al plants: I would guess that such processes don’t take well to being switched on and off, although being ramped up and down might be a feasible load balancing mechanism.

  7. Joffan, that is pretty much what we are doing here in Minnesota, and it is not far from other states. I think the idea is to go to 10 or 20 % over the next decade or two.

  8. The trouble with all this interminable wrangling over how to prevent global warming runaway is that it has all become a kind of religion. There are the green integrists whose god is ‘renewables’ and whose devil is anything with the word ‘atomic’ attached, the heathens who blaspheme and call for atomic power everywhere, and the mass of quiet folks who vote to close down everything nuclear but hate the destruction of the countryside with windfarms and fields of solar panels. Like with all religious arguments, one never gets to a reasonable consensus.
    Lets take a look at the problem to be solved. According to the overwhelming majority of climate experts, if we don’t reduce our CO2 output by something like 80% by 2050 we risk unstable warming runaway where we loose control over the temperature rise and may end up with catastrophic destruction of the environment.
    To get an 80% reduction of CO2 pollution by 2050 would imply that Europe and the USA would have to have stopped burning fossil fuels by 2030, followed by India and China by 2050. To do this a crash program of construction of clean electric generators would be necessary. Clean is the important word, not renewable. Renewable generation will be necessary one day, long into the future, when the world energy consumption gets to around 500TW, else the generation of electric power alone will become a source of global warming. At the moment we generate around 20TW, so we have a 150 years or so before renewables become a necessity.
    Clean generation could be done with carbon capture on coal and gas with underground storage, but the technical problems and logistics are enormous. Few people imagine it could be done in 20years.
    Wind power has the disadvantage that to get, say, a 50% contribution, many 100’s of 1000’s would need to be installed along with gigantic energy storage batteries of some kind. There are several reports from competent scientisrs which have used the consumed fuel measurments of large national electricity suppliers with and without wind contribution. They have all concluded that the lowered efficiency of closed circuit gas turbines with cycling due to intermittent wind gives an overall fuel consumption with intermittent wind which is greater that that with no wind contribution. There was no reduction of CO2!
    Direct solar electric generation will one day be the ‘renewable’ solution, but today the efficiency is so low and the electric storage problem so great that only a minor, variable, contribution can be hoped for within the next 30years. Nuclear, the only practical solution, being the devil in person for an influential sector of the public, is out, at least until the first sign of an imminent climate catastrophy appears, which will perhaps be too late.
    We just have to hope that the climate scientists have been too pessimistic !

  9. It is funny, Dennis, how you say all these specific (unreferenced) things about all the different types of energy but you make no argument whatsoever about nuclear. You just state that it is the only solution.

    Jesus saves!

    Anyway, I’d love to see references for the studies you site.

    I’ll be reviewing a book here shortly that everyone must read. Addresses everything you bring up in your comment.

  10. You’re right Greg, one should always include references but my comment was getting a bit too verbose.
    Here are a couple of interesting URL’s. The first two are the reports on the fuel consumption with and without wind power. The one from the nuclear org on CO2 emissions and safety is, of course, from a pro nuclear source. However I have seen similar figures(which I can no longer find) from an independant safety group where Hydro was given the worst safety record (mainly from the Chinese dam accident where 26000 were killed) and coal a close second (from the calculated deaths among the worlds population) due to the dreadful pollution from the chimneys. Chernobil was the only nuke accident with any serious death or injury, mainly from the thousands of workers sent with poor protection to clean up the mess. Three mile island killed and injured nobody; Fukishima killed nobody and injured only a few safety workers.
    The last URL is interesting, showing the case of the underground natural reactor in africa which heated the rock to several hundred °C over millions of years without shifting. Indicates that underground storage of nuke wastes could be done safely. Of course not all points in my rambling ‘comment ‘are covered.
    The figure for the number of wind turbines needed to cover world electric demand is simple arithmetic. Just take the MWh consumption/year of any country,multiply by 4 (the average output is about 25% of rating) and then divide by 8600MWh (the average output of a 1MW rated turbine over 1 year. Add 10% for maintenance and then look at the size of the batteries to store that energy for, say 5days without wind!
    If you need other references (eg low efficiency of polar panels of around 18%)or info on the imperceptibily slow advances being made in carbon capture, let me know.

  11. You’re worrying me now Greg. I suppose you mean the turbine calc.
    Lets try the arithmetic for Germany, which I will take as about 80GW average level of consumption.
    So we get 80000 MW x 8600h in the whole year =688 million MWh for the year, 688 **6/8600 gives 80000 1MW turbines x 4 for the 25% wind efficiency(a very generous figure for Germany-is more like 15%), giving 320000 turbines. If the turbines are 2MW beasts one would only need 160000 of them.
    Of course nobody is proposing such a system. I believe 20% wind is more the idea so one would need to build 65000 rated at 1MW.I guess thats what the germans are doing at the moment.

  12. Dennis, what I’m referring to is this: You are giving us a four part equation. It is an inequlaity

    Good things about non-nuclear – bad things about non-nuclear < good things about nuclear - bad things about non nuclear in which you are counting up all the bad things about non nuclear but nothing else, in order to make your very biased case. People who read this blog are much, much smarter than you must think they are! What you say the Germans are doing is actually what we are doing here in Minnesota as well. Some of us are working on speeding up that effort. You know, to fill the gap between now and when a safe form of nuclear can be developed, tested, deployed in a limited way for a couple of decades and then if it works used more widely. Because we know that nuclear technology as it now stands is not safe, and is very expensive. Which is too bad. I want my free clean energy!

  13. OK, I see what you mean. I thought you were talking about my arithmetic!
    At the beginning of my first comment I said that global warming discussions are like religious ones, with no consensus at the end. This is what we are now seeing between you and I. For example you say above that nuclear technology is not safe, but the numbers in the URL I gave you shows that is one of the safest. I read a report the other day from E-ON, the big german energy producer. As you know the germans are stopping all nuclear generation very soon as a political comitment . In 2010 80% of german electricity was produced by non-renewables, with 40% nuclear and the rest brown coal and gas. I guess they are intending to replace most of the nuke with coal and wind, which, if the reports in my URL’s are not completely wrong, will mean a big increase in CO2! 2010 was a worldwide record year for CO2 pollution, and with everybody condemming nuclear I’m sure that when the world economy picks up again the figures will be even worse.
    In the end, Greg, there will be a reality check on all this as the % CO2 in the atmosphere and the world temperature anomaly increase and even the polititions start getting worried about their kid’s future.
    Many thanks for this excellent blog and for the exchange of ideas.

  14. At the beginning of my first comment I said that global warming discussions are like religious ones, with no consensus at the end. This is what we are now seeing between you and I.

    Please don’t drag me into your religious experience.

    J Doug: It is called “Pumped Storage” and it has been used for years. It is messy and big and inefficient but there have been times it has been determined to be useful.

  15. Wind energy are cost effective than electric energy. This Is one of the answer of to reduce global warming and protect our environment from pollution.

Leave a Reply

Your email address will not be published.