In the past several years, a number of new wind farms have begun commercial operation. Industry sources have estimated that more than 900 megawatts (MW) of wind capacity was under construction in 1999. A major portion of this capacity was constructed outside of California, the birth place of the wind power industry in the United States.⁽¹⁾ While the economics of wind turbine technology is improving, it is generally not yet competitive with fossil fuels.⁽²⁾ Just as the outlook for wind improves, it can also improve for other energy sources. Thus, despite the encouraging portrayal of wind turbines, they face uncertainty in the future. This paper looks at the forces behind recent wind energy development.

Current Status and Recent Events

In 1997, wind power generation capacity of 1,579 MW produced 3,254,117 megawatthours (MWh) of electricity.⁽³⁾ More than 99 percent of generation was by independent power producers, and nearly all of it was located in California. During 1998 and 1999, wind farm activity expanded into other States, motivated in part by financial and regulatory incentives and, in the case of Iowa and Minnesota, State mandates. Iowa, Minnesota, and Texas each had capacity additions exceeding 100 MW that came on line in 1999 (Table 1). During 1999, wind farm capacity that came on line consisted of state-of-the-art wind turbines manufactured primarily by Zond, a subsidiary of Enron Wind Corporation (392 MW); NEG Micon (325 MW); and Vestas (159 MW).⁽⁴⁾ Less than 32 percent of new wind power construction was located in California in 1999.

A number of recent events have triggered an interest in wind energy. Significant interest has arisen in the ability of renewable energy to survive as a viable energy source, compared with less expensive fossil fuels, as the electric power industry moves from a regulated to a competitive environment. Because renewable energy sources are generally perceived to be more environmentally benign than other energy sources, much recently enacted and/or proposed Federal and State legislation on electric competition contains provisions encouraging consumption of renewable energy. Hence, in those instances, electric restructuring may actually promote renewable energy use rather than restrain it. Wind energy, which is more economically competitive than most other renewable energy options, should benefit most from this effort.

Another event that increased interest in wind energy was the expiration of the federal production tax credit for any projects beginning operation after June 30, 1999. This tax credit was established by the Energy Policy Act of 1992 and provided a 1.5 cent per kilowatthour tax credit for the first 10 years of the project's life. Since all projects in operation by June 30, 1999, would be eligible for the tax credit, most of the capacity that came on line in 1999 came on by that date. Although the credit actually expired, it was reinstated in December 1999, it is retroactive to July 1999, and extends until the end of 2001. The current schedule for new capacity is less ambitious than 1999, but substantial (Table 1). A total of nearly 400 MW of new wind power construction (including a significant share of repowered capacity in California) was expected for 2000.

Additionally, in June of 1999, the Secretary of Energy announced the start of a new initiative, "Wind Powering America." The stated goal of this program is to have 80,000 MW of wind power generation capacity in place by 2020 and have wind power provide 5 percent of the Nation's electricity generation.⁽⁵⁾ Year-end 1998 wind power capacity was about 1,698 MW,⁽⁶⁾ so this goal represents an enormous increase in capacity additions. The initiative is mentioned here because of its potential importance and the attention it is drawing to wind energy. However, the full impact of the program on wind energy will be over the long-term future and is a concern more so for the Energy Information Administration's (EIA) Annual Energy Outlook, and less so for this paper, which covers the recent past and near-term future.⁽⁷⁾

Another long-term impact on renewable energy sources is concern over global warming and formulating a policy to reduce greenhouse gases in accordance with the Kyoto Protocol. A United Nations conference with representatives from more than 160 countries met in Kyoto, Japan, in 1997 to negotiate binding limits for greenhouse gas emissions for developed nations. Carbon dioxide is the major greenhouse gas. The target for the United States is to reduce carbon dioxide to 7 percent below 1990 levels in the 2008-2012 time frame. Adopting a carbon tax to accomplish this goal would increase the price of fossil fuels (particularly coal) but have little impact on the cost of renewables, which have zero or net zero carbon dioxide emissions. Assuming a carbon tax is imposed, analysis indicates that an increase in the consumption of renewable energy, led by wind, would make a significant contribution to achieving the targeted level of reduced emissions.⁽⁸⁾ The next United Nations Conference of Parties (COP) meeting to develop strategies to achieve the goals of the Kyoto Protocol was held in November 2000 in the Hague, Netherlands.⁽⁹⁾ No significant agreement was reached at that time, but future meetings are expected.

This paper is divided into two main sections followed by an appendix. The first section includes a technical discussion of expectations for wind turbine performance and efforts to improve it. The second section provides an overview of the world in which the wind power industry is developing. This discussion includes a broad view of the impact of electric power industry restructuring, as well as Federal and State incentives. These two main sections are supplemented by an Appendix of State Wind Profiles that takes a snapshot of the status of electricity restructuring in each State, the type of incentives or green power programs available to wind, and status of wind energy development through 2000. References are included so more current information can be obtained as needed.⁽¹⁰⁾

Wind Turbine Performance

The following sections provide an overview of the turbine technology being installed in today's wind farms. These turbines have generation capacities at or above 225 kilowatts (kW).⁽¹¹⁾ The discussion examines (1) wind resource issues and related siting considerations, (2) factors affecting wind turbine performance, (3) physical and operational characteristics of wind farm turbines and (4) operation and maintenance (O&M) considerations. The discussion focuses on wind farm turbines manufactured by NEG Micon, Vestas, and Zond, as they represent most of new installed capacity in the United States. The discussion indicates that each of their designs is equally adaptable to a variety of wind farm sites. The discussion shows how O&M considerations can be managed to ensure that the cost of O&M for a wind farm can be controlled and minimized.

A major caveat in evaluating information presented in this section is the availability of data. Performance data on operating wind turbines are frequently proprietary and extremely closely guarded. Thus, although some historical data are available, the data used in this chapter are often based upon engineering sources and not actual commercial operational performance data.

Factors Affecting Wind Turbine Performance
Wind Resources and Wind Turbine Machine Basics ⁽¹²⁾

Winds are created by atmospheric temperature and pressure variations caused by the sun heating air during the day, so general wind patterns coincide well with electricity demand during the daytime. During nighttime, temperature variations are lessened; therefore, winds are less severe. Although geostrophic winds (or global winds) winds determine the prevailing direction and magnitude in an area, the surface winds (up to an altitude of 100 meters) such as sea breezes and mountain winds are key factors in calculating the usable energy content of the wind at a particular site. Wind direction is influenced by the sum of global and local effects; when larger scale winds are light, local winds may dominate the wind patterns.

The wind resource is seldom a steady, consistent flow. It varies with the time of day, season, height above ground, and type of terrain. An area's surface roughness and obstacles are also important determinants in wind resource. High surface roughness and larger obstacles in the path of the wind result in slowing the wind by creating turbulence. Wind speed generally increases with height above ground.

A wind turbine converts the force of the wind into a torque (turning force) that turns the turbine blades, which are connected to the shaft of an electric generator. The amount of energy that the wind transfers to the blades depends on the density of the air, the blade area, and the wind speed. Wind speed determines how much energy is available for conversion to electricity. For wind farm applications, developers seek sites with an annual average wind speed of at least 7.0 meters per second (15.7 miles per hour), measured at a wind turbine hub height above ground of 50 meters (164 feet).

Wind power density, measured in watts per square meter of blade surface, is used to evaluate the wind resource available at a potential site. The wind power density indicates how much energy is available for conversion by a wind turbine. The power available at a given wind speed varies with the cube (the third power) of the average wind speed.⁽¹³⁾ Wind power developers think in terms of ranges of wind power density, termed wind power classes. Sites with a wind power class rating of 4 or higher are preferred for large-scale wind plants (see Table 2), which have installed capacity of at least 10 MW.⁽¹⁴⁾ For any given wind power class, the wind power density range and wind speed range increases with hub height; a hub height of 50 meters is the approximate hub height for utility-scale turbines. For instance, NEG Micon turbine hub heights range from 40-55 meters for 600 kW and 750 kW turbines, to 49-80 meters for their 900 kW to 1.5 MW turbines.⁽¹⁵⁾ Depending on rotor diameter, Vestas turbine hub heights range from 35-65 meters for their 600 kW and 660 kW models, to 60-100 meters for their 1.5 MW and 1.65 MW models.⁽¹⁶⁾ The Zond turbine hub height is 53 meters for their 750 kW turbines, with an optional 65 meter height for the 48 meter and 50 meter rotor diameter versions of the 750 kW turbine.⁽¹⁷⁾

The goal of wind turbine design is to convert as much of the power in wind, illustrated by the wind power classes in Table 2, into turbine generator power output. The power curve for a wind turbine shows this relationship of wind speed to turbine power output by plotting turbine power output (e.g., kilowatts) as a function of wind speed (e.g., meters per second). Power curve values vary among turbines because turbine design approaches differ. The impact of design on power curve values is illustrated by comparing the wind speeds at which various turbines achieve rated power. For instance, the Zond Z-48 turbine achieves 750 kW rated power output at a lower wind speed (11.6 meters/second) than does the NEG Micon Multi-power 48 (16 meters/second) (Table 3). The shape of the power curve also varies with turbine design. For instance, the NEG Micon Multi-power 48, which uses a generator that operates at constant speed, produces less than 750 kW output at wind speeds less than or greater than 16 meters/second (Table 3), the speed at which it achieves rated power. In contrast, the variable speed generator used in the Zond Z-48 design enables the turbine to maintain rated output of 750 kW over the range of wind speeds listed in Table 3, starting with 11.6 meters per second (the speed at which it first achieves 750 kW output), because the generator speed varies with wind speed to maintain rated output. Power output per unit of rotor swept area offers a way to compare performance among wind turbines. Restated, the goal of wind turbine design is to obtain the highest value of power output per unit of rotor swept area (Table 3) for the lowest capital cost.

Siting Factors Affecting Wind Turbine Performance

Several performance factors contribute to the selection of a wind farm site. Choosing a terrain with the least number of obstacles, least roughness, and the most expansive views is generally a good practice. The orientations of trees and shrubs and erosion patterns along a terrain provide clues to prevailing wind directions.

Meteorological data, preferably spanning periods greater than 20 years, are used to screen potential sites. Meteorologists collect wind data for weather forecasts and aviation, and that information is often used to assess an area's potential for wind energy. However, wind speeds and wind energy are not measured with great enough precision when monitored for weather forecasting to enable placement of turbines within a site. For example, wind speed is influenced by surface roughness, obstacles, and contours of the local terrain. The impact of these factors may be estimated when screening for potential wind farm sites.

Land conditions, which affect the cost of site preparation, are a factor in wind farm economics and in site selection. The earth must be able to withstand the combined weight of a tower foundation and the tower, turbine, and rotor. The earth and geography leading to and including the site must be accessible to large, heavy trucks and cranes used to haul wind turbine components on to the site and to install the turbines. The cost of building a road to the site must also be factored into site selection.

Connection to the electric grid presents other issues that must be addressed when choosing a wind farm site. Grid connection may be a component of total project cost, depending on the terms of the wind electricity purchase agreement between the wind farm developer and the electric utility. For example, the Southwest Mesa Wind Energy Project in Texas uses 700 kW NEG Micon turbines, which produce 600 volt electricity.⁽¹⁸⁾ Electricity travels from the turbine to a field transformer to the wind farm substation to the utility transmission line. Therefore, the following transmission capital must be included in the project cost: field transformers, substation, and transmission lines to connect each element, ending with connection to the utility line. Congestion on the regional transmission system is also a consideration. It would be undesirable to locate a new wind farm where the transmission system would not accommodate the power generated.

Once a potential site is selected, meteorological data are measured at points within the site as part of wind turbine "micrositing." Micrositing refers to the actual placement of turbines within a wind farm site to optimize electricity production.

Capacity Factor

Capacity factor is defined as the actual annual wind farm energy output, in kilowatthours, divided by the rated maximum turbine output, in kilowatts, times 8,760 hours/year. For a 100 kW turbine producing 175,000 kWh in a year, the capacity factor would be:

Capacity Factor

Factors affecting the magnitude of the capacity factor include wind resource intermittency, the wind farm site's wind speed distribution, turbine design, and turbine reliability. The degree of wind resource intermittency may vary both daily and seasonally. For a given turbine design, turbines sited where the wind resource is more intermittent will have a lower capacity factor. The wind farm site's wind speed distribution, and the associated average annual wind speed, affect annual electricity output. The annual electricity output for a wind turbine increases with average annual wind speed, since more hours of operation at a higher wind speed mean a higher average kilowatt power output from the turbine. Thus, for a given turbine design, wind farm sites with higher mean wind speeds have higher capacity factors. Historical data show wind farm capacity factors in the range of 25 percent to nearly 36 percent (Table 4). An objective of turbine design is to maximize annual power output, which would increase the capacity factor. Higher capacity factors, compared to Danish data and DOE 1997 baseline data for class 4 winds, are projected for the Zond Z-750 Series turbines (Table 4) because the Zond Z-750's variable speed generator design, taller tower, and larger rotor swept area enable a greater amount of wind energy to be converted to electrical energy. Finally, an increase in turbine reliability would be reflected in an increase in the capacity factor.

Annual electricity production can be estimated from the turbine's power curve, which plots kilowatt output as a function of wind speed.⁽¹⁹⁾ Alternatively, electricity production from wind turbines may be estimated by statistical means.⁽²⁰⁾

Contrary to conventional steam or nuclear power generation, the wind turbine with the larger capacity factor may not have an economic advantage over a wind turbine with a lower factor. For example, compare two wind turbines with the same rotor diameter but different generator capacities in a location with daily wind gusts or seasonal wind variations that are above the mean daily or seasonal speed. The turbine with the larger generator may be more economical because it enables higher power output, thus more electricity, when the wind turbine can take advantage of higher wind speeds. This strategy would tend to lower the capacity factor, using less of the available capacity of a larger generator. However, the strategy is economical if the value of the electricity production can be increased more than the incremental cost of the larger turbine over a smaller capacity turbine. The value of the electricity depends on daily or seasonal variations in electricity price. For instance, increased electricity production from a larger turbine has more value if produced during peak, rather than off-peak, periods of a utility's load curve.