The economics of wind power

By Ulrich Decher, Ph.D.

It is often stated that since no one can charge money for the wind, wind-generated electricity is free. This is not true. A modern wind turbine, which can generate 2 megawatts of electricity (MWe) when the wind is blowing, costs about $3.5 million installed. Five hundred of these turbines installed at a wind farm, to be able to generate 1000 MWe, would cost $1.75 billion. Add in other costs, such as for operation and maintenance (O&M) and transmission lines, and the total sum could match the approximate $4 billion required to build a nuclear plant.

All of these costs need to be recovered from customers or taxpayers. So, the cost of wind-generated electricity is not free.

Diablo Canyon nuclear power plant

A typical wind farm would generate electricity about 30 percent of the time, and not necessarily at times when electricity is needed. There is a very big difference between intermittent sources of electricity, such as wind farms, and baseload sources, such as nuclear power. The argument that nuclear power also has down times is true, but these refueling and maintenance outages are largely planned during times of low electricity demand (during spring and fall).

As I mentioned in Fitting Wind onto the Electricity Grid, my recent ANS Nuclear Cafe post, wind turbines by themselves do not add electrical capacity to a grid. They must be paired with other generators of equivalent power to compensate for wind variations and for the stability of the  electricity grid.

This pairing—wind and backup—has limits because of the huge rapid variability of wind that must be compensated for by the backup power source. It is estimated that this pairing can account for only 20 percent of the capacity of the grid. This means that wind can be only 6 percent of the generation (.20 x .3). This limit has already been reached in Europe by countries such as Germany and Denmark.

Wind power fuel tradeoff with natural gas

Since wind power is a fuel saver, one of the questions that might be asked is exactly how much fuel is saved, or put another way: What is the economic tradeoff between wind farms and the fuel saved, such as in a natural gas power plant?

A simplified comparison shows that the worth of the natural gas saved is less than the cost of building and operating a wind farm. The details of the cost tradeoff are shown at the end of this article.

There are some additional costs that make the comparison even worse:

  • Transmission losses. Since the transmission lines from a remote wind farm are likely to be longer, a wind farm may need to be larger to provide the same amount of power as the backup. For example, if we assume a 10-percent electricity loss per 100 miles, a wind farm 500 miles away needs to be double in size.
  • Transmission line cost. A remote wind farm will need expensive transmission lines to deliver the electricity. For example, a proposed new 12 000-MW high voltage transmission line connecting wind sources in New England would cost $19 billion–$25 billion[1]. Transmission line cost may not be directly born by the power provider, so these costs may be hidden from any direct cost comparisons, but ultimately they are still paid for by the consumer or taxpayer.

Pickens

An illustration of how the pairing of wind and natural gas has failed recently due to economics was provided by T. Boone Pickens, when he tried to send wind-generated electricity from Texas, which he called the “Saudi Arabia of wind”, to California. His attempt at promoting natural gas by pairing it with wind seemed like a good idea and got much television advertisement (his emphasis was on the wind portion of the pairing, as it seemed a more popular idea). His strategy, however, depended on gas prices at $9 per million BTU. The price has since dropped to $4 per million BTU.

There appears to be no economic justification for windmills when paired with natural gas. If the price of natural gas is low, then the worth of the saved fuel does not compensate for the cost of the wind farms. If the price is high, then the use of natural gas is not competitive with other forms of power generation. Although natural gas prices without windmills may be competitive today, there have been price fluctuations by as much as a factor of two as recently as a few years ago.

Click to Enlarge

Wind power fuel tradeoff with oil-generated electricity

Wind farm on Maui

Another question that might be asked is how does this tradeoff compare when the electricity is generated with oil? In Hawaii, oil is the major fuel for electricity generation. Another favorable factor in Hawaii may also be that the wind generation capacity factor may be higher on these islands. A simplified cost tradeoff shows that there is indeed a cost advantage to backing up oil generation with windmills. Oil is such an expensive fuel that anything that reduces fuel consumption is well worth the cost.

The fact that oil is so expensive is the reason that it is seldom used in the continental United States for electricity generation. In Hawaii, however, there appear to be few other choices. This may change if small nuclear plants become available as a low-cost alternative.

Hydro backup

It should be noted that if hydro power is used to compensate for wind power, there is no compensating cost saving for the saved fuel. The saved fuel is the extra water that goes over the spillway and is wasted. It is cheaper to have no wind farms in this pairing and let hydro do the entire job of supplying the needed electricity. Here are some factors that limit wind generation on a hydro grid:

  1. Too much wind on the grid may violate the Endangered Species Act. Placing too much wind on the grid is actually a

    Salmon pool

    concern in California, as that state is negotiating with the neighboring Bonneville Power Administration (BPA) grid for renewable energy credits to meet its self-imposed Renewable Energy Standard. In order for the BPA to help meet California’s demand for wind-generated energy, it might need to decrease the hydro generation to the point that the excess water flow over the dams causes harmful effects to migrating salmon during the spawning season due to excess dissolved nitrogen [2].

  2. Too much wind on the grid may violate agreements to provide downstream irrigation needs. During drought situations, it may not be possible to turn down the hydro generation to let wind onto the grid and still meet irrigation needs.

Hydro plant

If there is no cost advantage or environmental advantage to placing wind on a grid with ample hydro, one may well ask why we are doing that. The answer is that we have passed laws in many states (Washington and California, for example) that do not count existing hydro into the legal definition of renewable energy. This may be surprising to many readers, as existing hydro certainly fits the definition of being naturally replenished. Existing hydro is certainly replenished as well as new hydro would be.

The BPA grid currently has 3000 MW of potential wind energy (when the wind is blowing). Assuming the above-mentioned price of a windmill, this means that consumers at the BPA have already spent at least $5 billion for wind-energy production without any obvious benefit. This potential wind capacity is expected to double by 2012, so BPA consumers are expected to spend another $5 billion without an obvious benefit.

The bottom line is that we have allowed laws to be passed that are harmful both to our pocketbooks and to the environment. Without the benefit of these laws, wind developers would have lost their legally mandated status and there would be no windmills on grids with ample hydro.

There is no free lunch

Wind-generated electricity is not free. The cost of fuel for any power plant is just part of the cost that a consumer needs to pay. Because the fuel cost is zero does not mean that the cost of the generated electricity is zero.

This is similar to the electricity generated by hydro. The cost of the water is zero, but the hydro-generated electricity is not zero. It includes O&M costs and the cost of building the hydroelectric dam.

For a nuclear plant, the fuel cost is not zero, but it is a relatively small portion of the generation cost. It is certainly smaller than the fuel cost in a natural gas plant, where the fuel cost is about 80 percent of the generation cost.

For power providers that use oil as fuel, it appears that wind generation is worth the fuel-cost savings.  Oil is not used extensively, however, because it is so expensive.

In conclusion, there appears to be no economic justification for building windmills except when low-cost alternatives are not available. This is especially true when windmills are placed on a grid with ample hydro, as there are no compensating fuel savings in that situation.

There is no free lunch.

Cost tradeoff of wind versus fuel  saved

Inputs:

  1. A 2-MW wind turbine costs approximately $3.5 million installed.
  2. The O&M cost of a wind farm is approximately 20-25 percent.
  3. The  maximum life expectancy of wind turbines is 20 years.
  4. The price of gas is about $4 per thousand cubic feet.
  5. The price of a barrel of oil is $80.
  6. It takes about 7.7 cubic feet of natural gas to generate 1 kWh of electricity (dividing the generation in Table 7.2a by the fuel consumption in Table 7.3a in these tables published by the U.S. Energy Information Administration ).
  7. It takes 0.00175 barrels of oil to generate 1 kWh of electricity (using the same tables as above).

Assumptions:

  1. The capacity factor of a wind farm is about 30 percent (land based).
  2. The a higher capacity factor of 45 percent is assumed for Hawaii.
  3. The average life of a wind turbine is 15 years.
  4. Interest costs for the wind farm are neglected.
  5. The cost of transmission lines are neglected.

Analysis

 

Cost of wind farms:

 

  1. A 1000-MW wind farm costs $1,750 million to install all the turbines (500 turbines  x $3.5M per turbine).
  2. For a lifetime of 15 years, the costs is $116 million per year (1,750/15).
  3. When including O&M, this increases to $145 million/year (116 x 1.25).

Electricity generated:

  1. The amount of electricity that a 1000-MW wind farm is expected to produce in a year is 2,630,000 MW-hrs for a 30-percent capacity factor (1000 MW x 365d x 24 h/d x .3).

Cost of natural gas saved:

 

  1. The value of the fuel saving in the backup 1000-MW natural gas plant is $81 million/year. (2.63 x 106 MW-hrs x 7.7 cubic feet/kWh x $4/1000 cubic feet x 103 kW/MW).

Cost of oil saved in Hawaii:

 

  1. The value of the fuel saving in the backup 1000-MW oil-fired plant is $552 million/year. (2.63 x 106 MW-hrs x [.45/.3] x 0.00175 barrels/kWh x $80/barrel x 103 kW/MW).

 

Conclusion:

  • The yearly natural gas fuel-saving cost benefit for operating a wind farm is less than the yearly cost to install and operate wind farms.  There is, therefore. no economic incentive to pair a natural gas plant with a wind farm, unless the price for natural gas goes up.
  • For a pairing of wind farms with oil-fired generation, there appears to be a significant savings. This is primarily due to the much higher price of oil versus natural gas for the same energy content. This is the reason oil-fired generation is not much used anywhere, except Hawaii, where there is not much other choice. At today’s prices, oil is 4.5 times more expensive than natural gas for the same extracted electrical energy (.00175 barrels/kWh x $80/barrel)/(7.7 cft/kWh x $0.004/cft)=4.5

References:

  1. Peter Wong, manager Resource Adequacy, ISO New England Inc. “An Overview of ISO New England and Operation of the New England Electric Power Grid,” given at Western New England College, Western Massachusetts Sustainability Symposium, October 24, 2009.
  2. Bonneville Power Administration Testimony before the Public Utilities Commission of the State of California, May 12, 2010 (See Page 8, second paragraph)

 

Ulrich Decher holds a PhD in nuclear engineering. He is a member of the ANS Public Information Committee and a contributor to the ANS Nuclear Cafe.

8 Responses to The economics of wind power

  1. Great article. My friend Glenn Schleede will be excited by it. It is relevant to MA where utilities have a choice between Cape Wind and Quebec Hydro. By way of background I was in the reactor physics section at Bettis Atomic Power Lab in my youth.
    John Bewick

  2. Roger Reynolds

    Excellent article. I have been trying to explain these concepts to my students and this article makes very clear, concise arguments.
    Thanks

  3. The cost of wind power will get cheaper with time as we learn to build them better. The cost of all of the other source of power will get more expensive as the supply get smaller and the demand rise. As the author reminds us that there is no free lunch we also need to calculate the health related cost of fossil fuel and nuclear fuel. With nuclear energy we still have build a repository but we don’t have the political will to use it so we store the nuclear waste in storage facilities that were not design for long term storage. As time passes we risk potential release of nuclear pollution. I don’t think we should expand our use of nuclear energy until we are political mature enough to deal with the waste issue.

  4. Bill Rodgers

    Mr. John,

    I am not sure how you can say that just because we will be able to build a better windmill in the future that for some reason power costs will automatically decrease.

    The future costs of windmill-generated power must factor in the costs of new transmission and distribution networks required to consolidate diffused wind power into useable power for the grid as Dr. Decher discusses. The wind power in New England is stagnant not because of expensive windmills but because of the billions required to build out the required T&D for windmills spread out over thousands of square miles.

    T. Boone Pickens did not walk away from his wind power plans for the Midwest because the project economics were unprofitable (they were not when long term natural gas sales required to back up the windmills was included). He walked away because the politics of the massive infrastructure build-up required to support his grandiose wind plans were turning against him at the same time natural gas prices were decreasing due to the national economic situation.

    In other words, he walked away because he was going to be expected to fund more the transmission and distribution requirements for the massive windmill build-up out of his own pocket and he did not want to be in the T&D business, only the natural gas sales business. Mandated windmill power guaranteed T. Boone Pickens would be able to sell more natural gas for decades to come but the T&D aspect made it unwieldy for his business plan.

    So no matter how inexpensive a windmill is in the future on an individual basis and no matter if the costs to install said windmill drop, there are other economic and noneconomic costs associated with wind power that are not accounted for which is the point Dr. Decher is making.

    Secondly, we have technical solutions for spent fuel rods; recycling, new generation reactors that can use spent fuel as an input with reprocessing, etc. What we do not have is the political leadership to drive a true solution from prototype testing towards full-scale implementation. Instead, we are left to argue amongst ourselves how to store a perfectly good source of energy for thousands of years when there is no technical reason or need to do so.

    So it is not an issue of political maturity, it is an issue of political grandstanding and laziness as well as significant influence from lobbyists from non-nuclear oriented industries that is preventing us from having a sane energy policy which does not rely on the spin of the earth and the daily rising and setting of the sun to generate electrical power.

  5. Ernesto Faillace

    I loved this article and the “island scenario” angle got me to thinking about AREVA’s possible offers in it’s integrated nuclear/renewables energies portfolio. We talk a lot about clean energy parks where wind and concentrated solar intermittent sources could be paired with more baseload sources like nuclear and biomass steam electric plants. I started thinking about “scalable” marine clean energy parks after the publication of the FlexBlue concept of small (~100 MWe) submerged nuclear power plants, one or more of which could possibly be paired as backup or peak power providers to a multiple of the number of wind-turbine-generators (dependent on nominal wind turbine power rating, average availabilty factors, baseload vs peak electricity demands, etc.) in an off-shore wind farm. Since the FlexBlue concept is derived from submarine reactor designs, which must be able to respond quite rapidly to changes in propulsion demands, their potential as a backup of intermittent sources like wind or solar power should be evident, even if they cost somewhat more per MWhr basis than baseload-oriented reactors. Here the “gold standard” would be the MWhr cost of generating back up power with oil, which could go even higher in countries where there are financial penalties for CO2 emissions. Once costs (and feasibility) for the FlexBlue concept are better defined, it might be interesting to explore this potentially symbiotic pairing of two low-CO2 sources of electricity that would help smooth out the intermittency effect of wind power in an island situation, while also providing a safe, reliable and relatively low cost (when compared to oil-based generation, which is NOT getting any cheaper these days) backup power wind or solar power plants.

  6. Ernesto: I also like the idea of small nuclear plants for the islands of Hawaii. However, what role do you think that the wind farms would have? Why not run the nuclear plants all the time, especially if the power can be adjusted to meet the daily load. Wind farms would make the nuclear fuel last longer, but I do not think that would be economical.
    Also, I’m not sure that people would like to visit Hawaii and see a lot of windmills everywhere. That is not the natural beauty that attracts people to the islands.

  7. Ernesto Faillace

    Sure, having ONLY a series of small nuclear plant would probably be good enough… But it really depends on the global economics of the energy park. Without a good economic analysis of the amortized costs the electricity generated from the three systems (oil-fired, off-shore wind turbine, underwater nuclear), it is difficult to make the comparison. It is probable that the cost of a series of small nuclear plant(s) would be relatively expensive when compared to a single large nuclear plant, but still lower than oil-fired plants. Having more small nuke plants allows for gradual investments and has the benefit of staggering any maintenance/refueling outages to maintain a pretty good baseload level. And if one uses “floating wind turbines”, these can be placed further away from the coast to minimize “visual pollution” – of course, in Hawaii one might be concerned with what those frisky humpback whales might do to them… In any case, it is always a pretty good idea to have some sort of diversity in your power supply, as long as there is reasonabl economic justification. Since winds tend to blow more consistently at off-shore rather than onshore locations, the economics of an off-shore wind-power/submerged nuke energy park that shares the transmission line infrastructure (which is submerged, so it is also easier to permit compared to overland towers) might be comparable to a “nuke only” park. Note that additional synergies could be achieved if the anchor cables keeping the floating wind turbines from drifting away could also do some double duty as “fence posts” supporting monitoring equipment detect intruders and could even serve as “tent poles” to keep the protective netting in place over a submerged reactor.

  8. And how long do you think gas is going to be this cheap at $4? You need to factor in a couple of scenarios assuming oil price at $120, $140, $160 and gas at $8, $10, $12 because you can bet your nuclear brain that over the lifetime of a wind farm (is it 15 or 20 years, you have assumed both it would seem) we will see these prices for oil/gas. the Windmills would look damn nice then, wouldn’t they?