Wind Turbine Aerodynamics Part A: Basic Principles - Wind Turbine Technology: Fundamental Concepts of Wind Turbine Engineering - page 336

Comparison of Theoretical VAWT Performance with Test Results

Theoretical predictions of the performance of the Sandia/DOE 17-m VAWT (similar to
that in Fig. 3-24) will now be compared with measured performance data [Worstell 1980].
The principal parameters of the Darrieus rotor on this machine are as follows:

Airfoil = NACA 0015; lift and drag data from [Sheldal and Klimas 1981]
R = 8.28 m (at z = 0) c = 0.61 m

H = 17.0 m A = 183 m 2

B = 2 W = 4.0 to 5.3 rad/s

N R = 1.5 to 2.0 ´ 10 6 at the rotor equator

r = 2.790 + 30.100 - z 2 0 £ | z | £ 4.548 m

r = 12.60 - 1.4826 | z | 4.548 £| z | £ 8.090 m

Test power coefficients are plotted versus tip-speed ratio in Figure 5-37, together with
vortex-model predictions [Strickland et al. 1979]. It can be seen that this type of aerodynam-
ic model is quite accurate. However, since an aerodynamic analysis based on vortex theory
is the most complex and time-consuming of the methods for predicting VAWT performance,
it is generally reserved for validating simpler models.
The test data and vortex theory predictions in Figure 5-37 are plotted in a dimensionless
power-curve form in Figure 5-38, as rotor-speed power coefficient, K P , vs. advance ratio,
J . We note that the region of maximum power coefficient ( J » 0.18 or l » 5.5) occurs at
low power levels, indicating that C P , max is not a critical parameter in determining the annual

Figure 5-37. Comparison of experimental and theoretical power coefficients for the
Sandia/DOE 17-m VAWT. Test data from [Worstell 1980]; theory from [Strickland et al.
1979]