3.5 Small Wind Turbines
3.5.1 Introduction
In the last two decades, wind power utilization has emerged from a niche industry to an industrial sector with global significance. In the mid-1980s, wind turbines had an average rated power of 30 kW and rotor diameters of less than 15 m. Since then, wind turbines with a rated power of 5 MW and more and rotor diameters of more than 125 m have been developed. Today, multi-megawatt wind turbines of 2 MW and more dominate the market. But in the last few years the interest in small wind turbines has also grown again. Small wind turbines are used in a broad field of application ranging from very small mobile battery chargers rated at only a few hundred Watts to grid-connected systems with up to about 100 kW.
3.5.2 Turbine Size and Applications
Some useful parameters for small wind turbine classification are their physical size, that is, the rotor diameter or the rotor swept area as well as their electrical properties, usually the rated electrical power. According to the international standard for the design requirements, small wind turbines have a rotor swept area equal to or smaller than 200 m2, generating at a voltage below 1000 V AC or 1500 V DC (IEC, 2006). This corresponds to a rotor diameter of up to 16 m and a rated power of up to about 75 kW. Furthermore, it is useful to distinguish between larger commercial systems and smaller residential systems. Figure 3.5.1 depicts the size of three small wind turbine classes in comparison to a large wind turbine.
Figure 3.5.1 Small wind turbines divided into three size classes
Another way to differentiate small wind turbines is by their application, for example battery charger, grid-tied systems, or water pumper; type of installation, for example, free-standing or building mounted, or mobile design characteristics, for example, horizontal or vertical rotor axis, or number of rotor blades.
Figure 3.5.2 shows four examples of available small wind turbine models of different size and design used for different applications.
Figure 3.5.2 Example of small wind turbine models of different sizes and designs
The small wind turbine market is complex. Worldwide more than 180 manufacturers offer a huge variety of turbine models with different technical features. The quality of these turbines is very heterogeneous. Whereas some available turbines do not comply with international standards and are unsafe to operate, others have a successful track record. Studies show that small wind turbines can achieve high reliability and a life expectancy of 20 years and more (Kühn, 2007). According to current market reports, about 20,000 small wind systems with an installed capacity of about 40.000 kW are installed every year (AWEA, 2010; Fraunhofer IWES, 2011; Renewable UK, 2011). It is estimated that approximately 75% of all newly installed small wind turbines fall within the micro category (cf. Figure 3.5.1). These small turbines are predominantly employed in off-grid systems and used as battery chargers. At the same time small wind turbines of the micro category contribute to less than 25% of the annual capacity added. Small wind turbines that fall within the XS category are deployed in on- and off-grid applications, whereas turbines within the S category are mainly grid tied.
On-grid systems dominate the market in terms of installed capacity, with the United States, United Kingdom, and China being the biggest markets. Typical grid-connected small wind turbines are tower mounted and provide electricity for private homes, small businesses, and farms. Yet the use of small wind turbines in cities and the built environment is controversial and challenging for reasons of safety, structural engineering, vibration, noise, and the complex flow conditions on rooftops and around buildings.
Regardless of their size, small wind turbines have huge potential in remote applications, especially to be used to supply off-grid telecommunication towers and to be employed for rural electrification in hybrid power supply systems and mini-grids. But so far, these future markets are largely untapped. Other small wind turbines target niche applications such as research and measurement stations, water pumps, heating, and advertisement.
3.5.3 Turbine Design and Technology
Despite the variety of design concepts, many small wind turbines are surprisingly similar. Common features are the three-bladed rotor made from fiberglass-reinforced plastics with a horizontal axis directly driving a permanent magnet synchronous generator.
One of the most fundamental design characteristics is the number of rotor blades. For large wind turbines, three blades are state of the art. The same is true for most small wind turbine models, although two-bladed and multibladed machines are also available. The choice for a three-bladed rotor can be attributed to being the best compromise between cost, aerodynamic characteristics, and dynamic behavior (Manwell, McGowan, and Rogers, 2006). Some design criteria related to the number of rotor blades are briefly considered in this section.
The ratio of blade area to the rotor swept area is called solidity. Small wind turbines with fewer rotor blades have a lower solidity and use less blade material accordingly. This is a major advantage as it results in less tower head mass and lower production costs. Contrariwise rotors with fewer blades are more challenging from the engineering point of view because they require a more advanced airfoil design and slender blade shape. In addition, fewer blades allow more compact generators and gearboxes as rotor speed and torque are linked to the number of rotor blades. For example, a rotor with three blades instead of four compensates for the missing blade by rotating faster at lower torque. At the same time, the loss in efficiency is comparatively small. The difference between three and four blades is only about 1%. A two-bladed rotor is about 3–4% less efficient than a three-bladed rotor (Hau, 2005).
Generally, rotors with fewer blades also emit more noise due to higher rotor speeds. Two-bladed small wind turbines are generally easier to assemble and install than machines with three or more blades (Gipe, 2004). A particular drawback associated with one- and two-bladed rotors is structural dynamic problems induced by nonsymmetric forces acting over the rotor plane. Three-bladed machines run more smoothly. In some applications a multibladed rotor might be favored. Examples include mechanical wind-driven water pumpers using 24 or more blades to provide a high torque. Also some battery chargers in the micro category have a six-bladed rotor design to allow a good starting behavior, that is, a high starting torque at very low wind speeds.
Small wind turbines with permanent magnet generators usually come in different voltage options. Typical voltages of battery chargers are between 12 and 48 V, whereas turbines for grid connection generate at voltages higher than 100 V. The variable voltage output from the permanent magnet generator is rectified and fed into a battery via a charge controller. In grid-tied systems, an inverter performs the conversion of the rectified generator voltage to the frequency of the grid, that is, 50 or 60 Hz. Instead of a directly driven permanent magnet generator, some small wind turbine systems in the S category use an induction generator and a gearbox. In these cases, the generator is directly connected to the grid.
Similar to all current turbine models in the multi-megawatt class, the majority of small wind turbines are upwind machines, that is, the rotor position is in front of the tower. A characteristic feature of these small horizontal-axis machines is the passive yaw system with a tail vane to orient the rotor into the wind (see Figure 3.5.2). But there are also small wind turbines that do have downwind rotors. An advantage of this rotor concept is the possibility to free yaw without the need for a tail vane. Special attention should be paid to the so-called tower shadow effect. That is when the downwind rotor blades pass through the disturbed flow in the wake of the tower. If not properly designed, it can result in having a significant impact on the dynamic behavior of the complete small wind turbine as well as on noise emissions.
The overspeed control ensures that a wind turbine remains within design limits at anytime, especially in extreme wind conditions. There are two distinctive and popular overspeed control options with small wind turbines, both being passive mechanisms. The first is the furling mechanism, which is common with turbines that have a wind vane. These machines are controlled by means of minimizing the projected rotor swept area. The second control mechanism is passive pitch, which functions by turning the leading edge of the rotor blades into the wind (Gasch and Twele, 2002). The pitch angle of the blades is changed by means of spring assemblies and induced by the wind thrust. There are also small wind turbines in the XS and S categories that use active control and protection systems similar to those in larger turbines, for example active pitch and active yaw. The control systems of these turbines are more complex and require additional electronics, actuators, and sensors.
The current market share of small wind turbines with a vertical rotor axis is rather small. An advantage of the three-dimensional rotor is that it does not need to be aligned to the wind, that is, a yaw system is not required. Another more subjective criterion is the more appealing design of vertical axis machines. Drawbacks of vertical-axis machines include a more complex rotor construction and a relatively high specific tower head mass. Both facts lead to higher efforts for manufacturing, transport, and transportation and result in higher specific costs (€/kW or €/m2). They also require more massive support structures when compared to horizontal-axis turbines.
The tower supports the nacelle and rotor and puts the small wind turbine into the wind out of the effects of the terrain. The choice of the right tower type and of the optimal height strongly depends on the wind turbine specification and on the site conditions. The most common tower types are tube or lattice towers that are either freestanding or guyed. Up to about 30 m, height tilt-up towers are very popular with small wind turbines in the micro and XS category. Tilt-up towers allow easy installation and maintenance. The most common tower type for turbines in the S category is a freestanding tower with a concrete gravity foundation. Usually freestanding towers can be climbed for maintenance and smaller repairs. However, they require a crane for installation and major repairs.
3.5.4 Performance
A wind turbine independent of its design, number of rotor blades, and rotor geometry can capture a maximum of 16/27 or about 59% of the kinetic energy in the wind. This is known as the Betz limit. The theoretical maximum power output of a wind turbine is as follows:
(3.5.1) 
where:
| CP,Betz | is the maximum theoretical efficiency of the turbine, that is, 59% of the available wind power can be converted into electrical power; |
| ρ | is the air density in kg/m3, and at standard atmosphere the air density ρ is 1225 kg/m3 (temperature is 15°C, air pressure is 1013,35 hPa); and |
| A | is the rotor swept area of the wind turbine in m2. |
The wind power is proportional to the air density and to the rotor swept area or squared to the diameter of circular shaped rotor, respectively. As air is about 800 times less dense than water, wind turbine rotors need to be comparatively large. The formula also shows that the power available in wind is proportional to the cube of the wind speed, that is, when the wind speed doubles, the power in the wind increases eightfold. Typical operating wind speeds of small wind turbines are between cut-in at about 3.5 m/s to cut-out at about 15 m/s. At higher wind speeds, the wind power becomes too high and the wind turbine needs to be protected from destructive wind forces. The engineering challenge is to convert the maximum of the highly fluctuating wind power over the whole operating wind speed range. The so-called method of bins is helpful to describe the power performance of wind turbines. Using wind speed bin-widths of 0.5 m/s or 1 m/s, the power curve of a wind turbine gives the mean power output for each wind speed bin i. The power output of a wind turbine PWT,i in wind speed bin i is represented by the following equation:
where:
| CP,i | is the average power coefficient in wind speed bin i; and |
| Vi | is the average wind speed in wind speed bin i. |
Typical maximum power coefficients of small wind turbines usually range between 0.25 and 0.35, whereas modern large wind turbines achieve maximum power coefficients of more than 0.5. The generally lower power coefficients mainly result from simpler design and control systems. A problem with the power curves of small wind turbines is that they cannot be trusted at any rate. In many cases, nonstandard procedures and test conditions as well as inadequate measurement equipment are used to determine the power curves. It is therefore advisable to check for the plausibility of a small wind turbine's power curve before performing yield estimations.
The average specific yield for small wind turbines is considerably lower than that of large wind turbines. Besides the generally lower power coefficients, low towers (i.e., hub heights of typically 10–40 m) and the influence of obstacles on flow and turbulence are reasons for this. Most small wind turbines are installed near the place of electricity consumption, that is, near buildings that act as obstacles slowing and distorting the flow of air. Experience shows that besides choosing the wrong turbine model, there are three main reasons for disappointed operators: inadequate sites, overestimated wind resources, and too-short towers (Kühn, 2010). This highlights that proper planning is essential to ensure good performance and longevity, as bad siting can considerably reduce overall power output and high turbulent flow can significantly shorten the life expectancy of a small wind turbine.
Wind resource assessment, yield estimation, and siting are fundamental tasks when planning a small wind system. The average wind speed is the most important parameter for the characterization of the wind resource and is ideally measured at the site and at the hub height of the planned small wind turbine. In many cases a wind measurement campaign, common with large wind projects, is not carried out. The accuracy of the yield estimation, that is, the annual electricity production highly depends on the available data about the local wind regime. Because these data are often unknown (e.g., the site characteristics), the wind speed must itself be estimated, allowing only very rough estimations of the annual electricity production. Wind maps can give a useful hint about the general wind resources of a region. From Figure 3.5.3 it becomes clear that Scotland, for example, has a much better wind regime than the Tuscany region in Italy. However, local conditions and obstacles cannot be accounted for. Furthermore, the resolution of these maps is not high enough to be used for yield estimations. Also, most wind maps show the wind regime for greater heights that are not relevant for small wind turbines with relatively small hub heights. It is therefore advisable to conduct a measurement campaign whenever possible.
Figure 3.5.3 Wind resources at 50 m above ground level for five different topographic conditions from the European Wind Atlas (Troen and Petersen, 1989)
The wind resource can be assessed by measurement campaigns lasting at least over a period of one year. Wind data can be obtained by means of a mast equipped with an anemometer and wind vane measuring wind speed and wind direction. Wind speed increases with height above ground. This is called vertical wind shear. If data from the measurement height z1 are available, the wind speed V at height z2 can be roughly approximated by the Hellmann power law:
Where:
| V(z1) | is the measured or reference wind speed at height z1 in m/s; |
| V(z2) | is the wind speed at height z2 in m/s; and |
| α | is the wind shear exponent. |
The wind shear exponent varies with the roughness of the terrain and with the stability of the atmosphere, among other factors. Another parameter to describe the surface roughness is the roughness length z0, which is equivalent to the height in m at which the wind speed is 0 m/s. Table 3.5.1 shows some values for α and z0.
Table 3.5.1 Wind shear exponent α and surface roughness lengths z0 for different terrain types (Gipe, 2004, p. 41)
| Terrain | Wind shear exponent α | Roughness lengths z0 in m |
| Cut grass | 0,14 | 0007 |
| Hedges | 0,21 | 0085 |
| Trees, hedges, and a few buildings | 0,29 | 0,3 |
| Suburbs | 0,31 | 0,4 |
For yield estimation, it is relevant to know the distribution of wind speeds, for example, over a year. This is important because very high wind speeds are equivalent to very high powers available in the wind (compare Equation (3.5.2)). However, their contribution to the overall available wind energy is usually comparatively low. This is firstly because very high wind speeds occur generally less frequently. Secondly, small wind turbines will cut out at very high wind speeds or will operate at limited power.
The Weibull distribution is applied in wind data analysis to model the probability distribution of the wind speed, especially if no time series of the wind speeds are available. The Weibull distribution can approximate a variety of different wind regimes, using only two parameters. The scale factor A is a measure of the wind speed. The shape factor k describes the form of the wind speed distribution. A special case of the Weibull distribution is when the shape factor k = 3.4, resulting in almost normal distribution, that is, the wind speeds would be equally distributed around the average wind speed. In reality, however, this distribution is not very common. The Rayleigh distribution is another special case of the Weibull function. It has a shape factor of k = 2. This distribution offers a fair approximation for many wind regimes, for example, for locations in Continental Europe. It is also used by many manufacturers of wind turbines as well as in international standards to calculate an annual reference electricity production. The Rayleigh probability distribution has the advantage that it requires only the average wind speed in order to be calculated:
(3.5.4) 
Where:
| dV | is the width; and |
| Vi | is the average wind speed in wind speed bin i. |
The annual yield or annual electricity production can simply be estimated by applying the power curve of a small wind turbine to the wind speed distribution calculated with the Rayleigh distribution. Based on the Rayleigh distribution, the annual electricity production can also be estimated without power characteristics of a wind turbine. First the total specific power density of the wind in W/m2 can be calculated for each wind speed bin i by equation (3.5.3), without applying the Betz factor. Now, by using the Rayleigh distribution the annual specific power density and the annual wind energy density can easily be derived. Some example values are given in Table 3.5.2 for Rayleigh distributed wind speeds and standard atmospheric conditions. In the fourth column is the specific annual electricity output of a theoretical conversion rate of 20% (e.g., a constant average power coefficient of 0.2). The estimation of the annual electricity of a small wind turbine with a rotor diameter of 5 m (i.e., about 20 m2 rotor swept area) is 2620 kWh at 4 m/s average wind speed and almost doubles to 5140 kWh at 5 m/s average wind speed.
Table 3.5.2 Estimates of the annual electricity output of small wind turbines with an overall wind energy conversion rate of 20% based on a Rayleigh wind speed distribution and standard atmospheric conditions. Adapted from Gipe (2004, p. 58)
