Wind power has long been treated as the Cinderella of energy technologies in the United States when it comes to spending public money on research and development (R&D). Last year was no exception. Compared to a national research budget for nuclear of more than $960 million, the wind research centre at the National Renewable Energy Laboratory (NREL) got $49 million. Even among renewables, wind has been short-changed. Solar got $168 million and $200 million went to biomass.
The federal government's spending priorities are a bone of contention in the wind industry. "Of all of the renewables, wind energy has the greatest potential for making a difference. So why aren't we pouring money into it?" asks Sandy Butterfield, chief wind turbine engineer at NREL, which runs the National Wind Technology Center (NWTC) outside Boulder, Colorado. He is among around 100 NWTC researchers working on gaining a better understanding of wind power technology, from the generation of electricity in low wind speeds, to solving chronic component failure problems, to enabling wind turbines to provide better grid support.
Much more could be done to buttress the industry if more funds were available, argues the wind lobby. But NREL's paymaster, the US Department of Energy (DOE), is unconvinced. The wind industry has reached a point of maturity where it has the resources to pay for its own R&D, says the DOE's Steve Lindenberg. He mentions Vestas, Siemens and Suzlon. Commercial interests are driving substantial investment by wind companies in advancing the technology and other governments are putting money into R&D, something the DOE watches "on a routine basis," says Lindenberg. "It's more than a billion dollars of federal taxpayer money that's been put into wind research over the last thirty years by the federal DOE and that's a substantial investment that has turned into a very credible and capable industry."
But Butterfield says more spending is needed if a US industry is to compete with that of Europe and Asia. He notes that half the wind capacity installed in the US last year came from foreign suppliers, including from a string of newcomers, among them Acciona from Spain and Repower, Fuhrländer and DeWind from Germany. The other half came from America's GE Energy (43%) and Clipper Windpower (7%). Main components are also nearly all made abroad. Butterfield tells of an executive at a major US utility he overheard expressing annoyance at having to buy all his replacement gearboxes in euros.
Butterfield believes Barack Obama could do more to encourage the established US engineering industry to diversify into full-scale wind turbine manufacturing, including main components. "When Obama goes to a fastener company as an example of a wind turbine manufacturing company in the US, what's wrong with that picture?" he remarks, referring to Obama's pre-inauguration trip to Cardinal Fastener in Ohio (Windpower Monthly, February 2009). "GE and Clipper are in the business, but why isn't Caterpillar in the business, why isn't John Deere? Why not GM and Chrysler and Ford and some of these who really know how to build high quality, high manufacturing and high production rates?"
Most recently, the NWTC has seen its budget grow (chart). Spending rose from $49 million last fiscal year to $55 million for the fiscal year 2009, the largest budget since 1981. Another rise is expected for the fiscal year 2010, which starts on October 1. Congressional debate will take place in the summer. Since the centre's launch in 1978, however, it has often been hard-pressed financially. It struggled through most of the 1980s and 1990s, with annual budgets sometimes below $10 million. "Almost no work occurred for ten or more years," says NREL senior engineer Jim Johnson. Democratic administrations have not necessarily been more generous than when a Republican is at the helm. Even so, the centre has provided the turbine industry with useful results on research into areas such as turbine noise and aerodynamics, he says.
Hopes are that wind will now benefit from Obama's $787 billion stimulus package, signed into law in February. It adds $16.8 billion to the $1.9 billion previously budgeted for the DOE's Office of Energy Efficiency and Renewable Energy -- which sponsors NREL -- giving the office 11 times last year's funding. The office, which is being given some say into how the money should be spent, is expected to announce details on allocations as early as this month.
While NREL waits for the 2010 budget and any allocation to wind from the stimulus package, lab officials are reluctant to state clearly how much more funding is needed, lest they be seen as dictating to the DOE. "They don't like that at all," says Johnson. The American Wind Energy Association (AWEA), however, has already drawn up a wish list for $441 million of annual funding for research into wind (table). Of this, the largest chunk, $108 million of national spending, plus $116 million from individual states and private industry, would go to improving wind turbine technology and reliability. Other top items include a combined $72 million for integrating wind energy into America's existing power system and $31.5 million to research transmission grid technology.
A main objective is to improve wind power plant efficiency by increasing production at any given site. NREL has high ambitions to boost the capacity factor from 35% now to 45% by the use of technology able to reap more energy from thin air. A 10% reduction in turbine installed cost and improved reliability are also top goals. AWEA says the spending is critical to meeting the DOE's target of providing 20% of electricity from wind energy by 2030, up from about 1.5% today. The association also calls for $45 million for offshore R&D as part of the annual $441 million, an area which Johnson wants to see NREL more involved in. That figure represents a compromise between factions within AWEA supporting much higher levels of offshore wind research and others saying onshore research should be prioritised, says AWEA's Aaron Severn.
Winds and wakes
Butterfield worries that calls for more public R&D money could provide ammunition to critics who say the variable supply of electricity from wind plant makes them an unreliable part of the energy mix. "I don't think anyone is saying that wind energy is fatally flawed, but I think if most people were asked if we can improve it, we would say absolutely," he says. More accurate forecasting tools for wind power production, understanding of local weather patterns and characteristics of wind turbine arrays will all contribute to integration of wind on the power system, says Butterfield.
He draws a parallel between R&D in the car and energy industries. "If we stopped in the 1960s when the US had the best cars in the world, we'd still be driving 1960s cars," he says. "But we've made major improvements in performance efficiency, safety -- all these things that make the cars credible today compared to where they were fifty years ago. That is true of coal plants, nuclear plants, gas plants, aircraft engines and right down the list. If we had stopped when we had gotten something that was commercially viable, well, we wouldn't be where we are today." Laboratories, he says, are better suited to certain kinds of research than wind turbine manufacturers, which are under commercial pressure to return profits to their owners.
Butterfield has some personal preferences for spending. He would like to see NREL expand its work on gearboxes and other critical drivetrain components (box). In particular, he also believes a greater understanding of the wind resource is needed. Better knowledge about high-altitude atmospheric conditions would allow developers to improve output forecasts from wind plant, while understanding of meteorological conditions closer to the ground could improve wind plant layout and turbine reliability. Research into wake effects at large wind plant is still needed, he admits. While the industry has a fairly good understanding of how a single turbine interacts with wind flows, more research is needed into how the complex flow of air within an array of machines affects performance, he says.
That view is echoed in the industry at large. "As you get bigger and bigger, what's going on deep in the array, no one has really done a comparative study understanding the entire flow field," says Eric White at New York-based wind forecasting company AWS Truewind. "When you look at reliability, wakes and stresses and how turbines up front are affecting wind that is then banging on the following turbines behind -- is reliability really going to be the same through the rows?" asks White.
Johnson says NREL would do well to boost research into cutting the cost of wind turbines by reducing their physical mass. He notes that as turbines increase in mass, cost rises faster than rated power. NREL is already working on how to reduce the mass of towers and blades. Johnson says manufacturers currently build wind turbines with excess mass to protect drivetrains against damage from strong gusts. Researchers envision enhanced wind prediction systems, coupled with near instant pitch control that could together eliminate imbalances in the rotor plane and reduce wear on the drivetrain. Faster acting, more precise pitch control could allow for less mass and therefore lower overall cost, he says.
Last year, about a third of NREL's $49 million wind budget was spent on power system integration issues and another third went to improving the performance of wind turbines and their components. Slightly less than 15% was spent each on collaboration with industry and special interest groups to publicise the benefits of wind and power generation in low wind speeds. The remaining 7% funded research into distributed wind power (chart). One small group at the lab measures wind resource around the country and throughout the world. A travelling group called Wind Powering America promotes adoption of wind by state and local agencies. In addition to research and technology promotion teams of about 25 members each, NREL also gets help in wind research from the government's Sandia National Laboratory, which has sites in New Mexico, California and other locations.
One project launched in response to industry pressure is an ambitious attempt to answer why bearings in wind turbine gearboxes consistently fail, leading to gearbox breakdowns (Windpower Monthly, September 2007). To understand the forces causing the problem, full drivetrain testing is being done on a testbed able to run at up to 146 rpm at the input shaft, enabling a few months of testing to simulate 30 years of operation. The effects of turbulence and high transient loads can also be induced. "We can test everything but the hub," says Johnson. "That includes the low-speed shaft, the bearing, the gearbox, the brake, the generator, the nacelle, everything -- even the cooling system."
Research into turbine blades is also an NREL speciality, with one group dedicated to aerodynamics theory. A recent research project involves analysis of how new scimitar-shaped blades from Californian glass fibre company Knight & Carver respond to bending and other causes of fatigue. The company's 17.5 metre prototype blades are designed to reduce load on machines through a variation on the sweep and twist of a typical unit, allowing for changes in velocity further out on the blade. Three of the blades were installed on a turbine at a wind farm in Tehachapi, California, and the fourth was attached to a turbine at the Colorado lab.
Larger blade testing is to shift to Massachusetts and Texas through a DOE-funded partnership with the private sector. The growth of blades has outstripped the existing capabilities of NREL's testing facility, which is far from deepwater ports facilitating easy shipment of large blades. DOE will provide each of the test facilities up to $2 million in capital equipment and technical assistance for development and operation. Universities, government agencies and corporate partners will provide additional funds. The test centres have not been built yet.
And the rest
While testing components might be the lab's most recognisable activity, it represents "probably a tenth of the laboratory's work," Johnson says. Among other things, NREL provides advice to utilities on transmission network issues; studies existing and new generator models, also at the design stage; and writes computer programs covering areas from aerodynamics, to turbine control, to wind flow modelling. The quirkily named Colorado Renewable Energy Collaboratory links NREL with three Colorado universities and other parties to produce skilled engineers to fill R&D posts across renewable energy fields. Johnson says that although these efforts receive less attention than hardware testing, they are nonetheless reflected in weighty papers presented at conferences and in data sets and computer applications that outside engineers can download online for use in design and production of wind turbines -- meaning efforts not so visible to the public also bear fruit across the wind industry.