Energy Answer Not Blowin' In The Wind
The wind is notably capricious, varying in strength and direction in seemingly random ways. The output of wind turbines, which capture the energy of the wind and turn it into electrical power, also varies unpredictably. Leveling out these fluctuations is the biggest obstacle to wind power serving as an effective alternative to conventional power sources. A study, just published in PNAS Online, claims that wiring together offshore wind farms, from along the entire length of the US East Coast, could provide a steady power source for the area. This has led to proclamations by a number of green power advocates that America's energy problems can be solved by wind power. Unfortunately, steady to a scientist is not the same as steady to a power grid engineer, or consumers.
The potential energy of wind is tremendous. One estimate puts it at nearly five times as much as the world's entire existing electricity demand. To environmentalists and climate change alarmists who fear the emission of CO2, wind offers a seemingly perfect energy source: it requires no dirty fossil fuels or hated nuclear power; no drilling, mining, or uranium enrichment; and wind's carbon footprint is essentially zero. The catch is that wind is notably unreliable, and if there is one thing consumers cannot abide, it is unreliable electrical power. A study, to be published in the Proceedings of the National Academy of Sciences (PNAS), and made available online, claims to have tamed the wind and smoothed out its inconsistent behavior. Motivation for the study is stated in the paper's first paragraph:
The world’s wind resource for electric power is larger than the total energy need of humanity. For surface winds over land globally, Archer and Jacobson estimate the wind resource at 72 terawatt (TW), nearly five times the 13 TW world’s demand for all energy. In a more detailed regional estimate, Kempton et al. calculated that two-thirds of the offshore wind power off the U.S. Northeast is sufficient to provide all electricity, all light-vehicle transportation fuel, and all building heat for the adjacent states from Massachusetts to North Carolina.
The study, “Electric power from offshore wind via synoptic-scale interconnection,” was authored by Willett Kempton, Felipe M. Pimenta, and Dana E. Veron, all from the Center for Carbon-free Power Integration at the University of Delaware, and Brian A. Colle, from the School of Marine and Atmospheric Sciences at Stony Brook University. “As wind power becomes a higher proportion of all generation, it will become more difficult for electric system operators to effectively integrate additional fluctuating power output,” the study says. “Power fluctuations are important if wind is to displace significant amounts of carbon-emitting energy sources.”
The group studied wind data from 11 meteorological stations along 1,550 miles (2500 kilometers) of coastline off the Eastern United States. “Instead of just looking at the statistics of connecting turbines,” said lead author Willett Kempton, an electrical engineer and energy policy analyst, “we also decided to look at the meteorology.” They compiled five years of wind data from the monitoring stations, whose locations are shown in the figure below.
Offshore meteorological stations selected for this study
Existing projects for offshore wind power are authorized at the state level and end up connecting to existing land grids. According to the authors, this creates problems managing the fluctuating power from each individual project. They claim that “the variability of wind power is not as problematic as is often supposed, since the electric power system is set up to adjust to fluctuating loads and unexpected failures of generation or transmission.” This is done using the traditional mechanisms to smooth out power grid fluctuations: reserve generators, redundant power line routes, and ancillary service markets. But the existing grid was not built to handle large amounts of highly variable generation capacity—existing power sources are both steady and adjustable by the operators, not controlled by the whims of nature.
As an alternative, the authors propose that all the offshore generators for the entire east coast be connected in their own power grid, to be managed by its own Independent System Operator, an ISO in power grid speak. A wind farms located in the Northeast might be operating at full capacity under gale-force winds, while the southeastern portion of the grid languishes under sunny skies and tepid breezes. Theoretically, an “Atlantic Transmission Grid,” could balance the fluctuations from the scattered wind farms and provide reliable power sufficient to meet the needs of all the adjacent states.
In the study, the researchers estimated how much power offshore wind farms could produce if they had been placed at the same locations as the monitoring stations—which would be the case under current wind-farm configurations. Then they calculated the combined power output of the farms if they were all connected into a single grid. As expected, each station's power output varied, depending on how the wind was blowing. Simulating a power grid connecting all the stations showed a leveling of this fluctuation.
Capacity factors for 11 stations in May 1999 and November 1999.
Note that the term CF stands for “capacity factor,” calculated as a generator's average output divided by its rated capacity. For example, a 1 MW turbine with a capacity factor of 35% (0.35) will not produce the fully rated 8,760 MWh in a year (1 × 24 × 365), but only 1 × 0.35 × 24 × 365 = 3,066 MWh, averaging to 0.35 MW. Of course, all power plants have capacity factors, depending on energy source, technology, and purpose. Here is what the Renewable Energy Research Laboratory says about sources other than wind:
Typical wind power capacity factors are 20-40%. Hydro capacity factors may be in the range of 30-80%, with the US average toward the low end of that range. Photovoltaic capacity factors in Massachusetts are 12-15%. Nuclear capacity factors are usually in the range of 60% to over 100%, and the national average in 2002 was 92%.
Reportedly, linking the wind farms showed “a tremendous amount of smoothing” of power output. As the wind data showed, Kempton added, the quick swings between high- and low-power generation periods that are characteristic of individual wind farms slowed down dramatically within the simulated grid, varying over days instead of hours or even minutes. By creating a wind-power grid, he says, “you can make a rapidly changing and unsteady source of power a slowly changing and stable one.” Well, sort of.
One month of power fluctuation.
If you look at the actual figures, about 20 times a month the output of individual stations S2 and S10 changed by over 50% in the span of a single hour. The “smoothed” grid changed by no more than 10% of its capacity in any single hour, which is certainly an improvement. But in power grid terms that is positively terrible. I am talking about reliability and availability—not power quality, which also must be met but is not the issue here. A reliable power grid must deliver adequate power when it is called for.
Historically, loss-of-load probability (LOLP) has been used as the single most important metric for assessing overall reliability.† LOLP is a projected value of how much time, in the long run, the load on a power system is expected to be greater than the capacity of the generating resources. LOLP is really not a probability but an expected value. The common practice was to plan the power system to achieve an LOLP of 0.1 days per year or less, which was usually described as “one day in ten years.”
In practice, “one day in ten years” has been proven to be not acceptable. The Northeast blackouts of 1965 and 2003, and the New York City blackout of 1977, resulted in major changes to power system planning and operating procedures. The economic impact and public outcry caused by these events have prompted power suppliers and regulators to try to prevent their recurrence, even though they occurred more than ten years apart.
Furthermore, LOLP characterizes the adequacy of generation to serve the load on the system. It does not model the reliability of the transmission and distribution system or the duration characteristics of outages (e.g. one long outage as opposed to several shorter ones). Even at this somewhat superficial level, it is obvious that the researchers' “smoothed” wind grid cannot meet expected reliability standards.
Free Power At What Cost
In theory, there are any number of alternative green energy sources that could power the world: solar, geothermal, ocean waves and wind. But what about the cost? According to the PNAS study, the cost to link existing or planned offshore wind farms is minimal. Of course, they assume that the wind farms themselves are already paid for. Here are the calculations from the report:
As an approximate cost comparison, a total of 2,500 MW of offshore wind generation has been approved or requested by states from Delaware to Massachusetts. Connecting them by a 3 gigawatt (GW) HVDC submarine cable would require 350 miles of cable. At early European offshore wind capital costs of $4,200/kW and submarine cable capital costs of $4,000,000/mile, the installed costs of planned offshore wind generation would be approximately $10.5 billion; the connecting transmission would add $1.4 billion.
Before proceeding we need to take note of the haphazard mixing of electrical power and energy in most media reports regarding wind and other alternative energy sources. Electrical power is measured in watts (W), named after the eponymous James Watt, while energy is given in watt-hours or, more commonly, kilowatt-hours (kWh or more correctly kW·h, but most everyone leaves the “·” out). Using SI units for energy and power, one joule of energy is the work done moving against a force of one newton over one meter's distance. One watt of power is the rate at which work is done when an object is moving at one meter per second against a force of one newton.
The watt is used to measure electrical power as well. In electrical terms, work is done at a rate of one watt when one ampere flows through a potential difference of one volt.
Electrical consumption, which is energy, is given in kWh or some multiple of watt-hours. To get kilowatt-hours, take wattage and divide by 1000 to get kilowatts and then multiply by the number of hours of use. If a 100 watt light bulb is turned on for one hour, the energy used is 100 watt-hours or 0.1 kilowatt-hour, or 360,000 Joules (60 min/hour · 60 sec/min · 100 J/sec). This same quantity of energy would light a 40-watt bulb for 2.5 hours. Generating capacity is usually given in terms of power, and is listed in megawatts (MW) or gigawatts(GW). Rated capacity is the maximum demand or load a generator can support.
In 2008, according to the US DOE's Energy Information Agency (EIA), annual net electric power generation decreased for the first time since 2001. Electricity generation and sales were adversely affected by the weakening economy, dropping 0.9% from 4,157 billion kilowatt-hours (kWh) in 2007 to 4,119 billion kWh in 2008. Summer peak load fell by 3.8%, from 782,227 megawatts (MW) in 2007 to 752,470 MW in 2008. Winter peak load, which is always smaller than summer peak load, increased in 2008 by 0.9%, from 637,905 MW in 2007 to 643,557 MW in 2008. Nationally, the contiguous US experienced an average temperature that was the coolest in more than ten years, so the downward overall trend may not continue.
Optimistically using the 2008 Summer peak load and dividing by the 2,500 MW rated power of the approved offshore wind farms yields a national requirement of an additional 300 times more capacity. Multiplying $12 billion per 2,500 MW installed (cabling included) by 300 yields a total price of $3,612 billion. But that estimate is wildly optimistic.
The average CF of a wind turbine is around 0.4, so to actually generate that much power you need to divide the total cost by the CF, yielding $9,030 billion. And this is only for a grid that is 90% reliable. To raise the generation capacity up to the expected CF of 0.4 across a typical month would require four times the calculated amount of generation capacity. We are talking about more than 36 trillion dollars to even start approaching a wind only grid with acceptable reliability.
The future look of the American coastline.
Predictably, this interesting but not earthshaking research has caused a flurry of pronouncements gushing about powering the world with wind, primarily from those with heavy emotional investments in renewable energy. As reported in a Science Now news article, this is the first time a study has demonstrated that offshore East Coast wind can provide “a relatively reliable supply of smooth power.” That according to environmental engineer Mark Jacobson of Stanford University in California, who went on to call the findings important because the wind resources of the region are “tremendous and could theoretically supply all US electric power demand.” In theory, $36 trillion gets us to “relatively reliable,” what a deal.
There are good reasons why wind and solar energy enthusiasts always talk about providing enough power on average to supply some number of homes or businesses. The fundamental fact is that these forms of energy are intermittent, inherently unsteady, varying over time in unpredictable ways. Even with 4 or 8 times the required rated capacity in wind generators there will still be times when the power available will not meet demand. When this happens things will ether have to be shut down, presumably by a new expensive “smart grid,” or the power grid will suffer brownouts and rolling blackouts.
Also, do not expect the results for America's East Coast region to hold for other places around the world. In a 2008 study by Oswald et al. little benefit was found in pooling the output of wind farms in the UK. According to Kempton et al., “the lack of benefit seen by aggregating stations in the United Kingdom may be due in part to the roughly north–south orientation of the island, thus experiencing their east–west passage of frontal systems nearly simultaneously.”
The American coastline has a lot of wind.
Even over the North Sea, where much of Europe's wind generation capacity is located, the breeze sometimes fades. On one day in January, 2009, for example, a stalled high-pressure system becalmed many of Germany’s wind turbines. Only 113 MW flowed from the country’s 24 GW of installed wind turbines—less than one half percent of capacity. Denmark's solution to unreliable wind power is simple: when supplies run short they buy power from the reliable nuclear reactors in France. Perhaps Canada or Mexico will sell power to the US when the winds stop blowing.
Electrical engineers have been studying the problem of integrating variable power sources, like wind and solar, with the existing power grid for several decades. They have calculated that the maximum amount of variable energy that can be supported by a distribution grid and still maintain reliability is around 25-30%. Several countries have achieved relatively high levels of wind power generation: 19% of electricity production in Denmark, 13% in Spain and Portugal, and 7% in Germany and Ireland as of 2008. Even the US DOE's optimistic “20% Wind Power By 2030” plan doesn't see a future scenario where wind power contributes more that a manageable 20%. This primarily because the costs of dealing with intermittency rises sharply above 20%.
Officials from the Obama administration are hailing wind power as this week's energy solution. On April 6, 2010, Interior Secretary Ken Salazar told an open forum in Atlantic City that East Coast wind turbines, stationed offshore, could generate the same amount of electricity as 3,000 coal-fired power plants. “There is tremendous potential with wind off the Atlantic,” he said. The DOI also released a report concluding that shallow water wind energy alone could provide 20% of the electricity needs of almost all coastal states (again, note the 20% cap).
Cristina Archer, a specialist in wind energy meteorology at California State University, says the PNAS study findings are “amazing.” Kempton's team shows “that an uninterrupted power supply from winds along the most populated and most energy-demanding coastal area in the country, and perhaps in the world, is possible.” Possible to a scientist and realistic to an engineer are obviously two different things. This is why scientists should not be allowed to plan real world projects.
Bottom line: the wind will not provide the majority of the USA's electricity, or even the East Coast's. Do not misunderstand, I think that wind power has a place in the energy future, both in America and around the world. It can play a part in almost any country's effort to achieve energy independence, but the wind alone is not a viable solution. Fatuous statements by unrealistic green-power advocates does everyone a disservice. Claims that wind power can be free and reliable are incorrect—the wind is free, the power grid must be reliable, and wind power is neither.
Be safe, enjoy the interglacial and stay skeptical.
Sunsets will still be pretty in the future, even with ugly wind turbines.
† Note: there are a number of ways to measure power grid reliability besides LOLP. For more about power quality and reliability see “Measurement Practices for Reliability and Power Quality,” by John D. Kueck et al., a report prepared for the US DOE by Oak Ridge National Laboratory.