Читать книгу Wind Energy Handbook - Michael Barton Graham - Страница 106

The turbine considered in this section is stall regulated and is run at constant rotational speed. More detail about this method of operation will be discussed in the next section, but the main feature is that there is, theoretically, a unique power output for a given wind speed.

When the turbine was under test, the chosen rotational speed was 44 rpm. Energy output and wind speed were measured over one‐minute time intervals and the average power and wind speed determined. The test was continued until a sufficient range of wind speeds had been covered. The one‐minute average results were then sorted in ‘bins’ 0.5 m/s of wind speed wide, and a fairly smooth power vs wind speed curve was obtained, as shown in Figure 3.60.

The turbine has a diameter of 17 m and would be expected to produce rather more power than shown above if operated at a higher rotational speed.

From the data in Figure 3.60, the C_P ‐ λ curve can be derived. The tip speed of the blades is (44π)/30 rad/s × 8.5 m = 39.2 m/s, the swept area is 8.5².π = 227 m², and the air density was measured (from air pressure and temperature readings) at 1.19 kg/m³_.

Figure 3.60 Power vs wind speed curve from the binned measurements of a three blade stall‐regulated turbine.

Figure 3.61 Comparison of measured and theoretical performance curves.

Therefore,

(3.96)

The mechanical and electrical losses were estimated at 5.62 kW, and this value was used to adjust the theoretical values of C_P. The resulting comparison of measured and theoretical results is shown in Figure 3.61.

This comparison looks reasonable and shows that the theory is reliable, but the quality of the theoretical predictions really relies upon the quality of the aerofoil data. The blade and aerofoil design are the same as given in Section 3.11.

Figure 3.62 Measured raw results of a three blade wind turbine.

One last point should be made before classifying the theory as complete: it would be as well to look at the raw, one‐minute average data before it was reduced down by a binning process; this is shown in Figure 3.62. In the post‐stall region, there seems to be a much more complex process taking place than the simple theory predicts, and this could be caused by unsteady aerodynamic effects or a bistable separation condition.