Wind Turbine Aerodynamics Part B: Turbine Blade Flow Fields - Wind Turbine Technology: Fundamental Concepts of Wind Turbine Engineering - page 355

Early wind turbine field testing aimed at understanding rotational augmentation affirmed
the importance of blade geometry with respect to rotational influences [Madsen and Chris-
tensen 1990]. A wind tunnel experiment showed that blade geometry coupled with blade
rotation maintained blade lift under conditions in which lift otherwise would significantly
decline [Barnsley and Wellicome 1992]. Subsequent wind tunnel research determined that
rotational augmentation was most active at the inboard portion of the turbine blade [Ronsten
1992]. Analytical modeling of rotational augmentation furnished better comprehension of
the aerodynamics underlying this phenomenon, and provided foundational predictive capa-
bilities for design and analysis [Eggers and Digumarthi 1992, Snel et al. 1994, Du and Selig
2001; Corten 2001, Tangler and Selig 1997, Corrigan and Schillings 1994].
Inboard Locations on Turbine Blades
As shown in Figures 5-46 and 5-47, inboard radial locations exhibit rotational augmen-
tation more strongly than do radial locations farther outboard on the blade. In Figure 5-46,
C n at 0.30 R is plotted as a function of LFA , for both the stationary (parked) and rotating UAE
Phase VI blade with its NREL S809 airfoil. All data correspond to an axisymmetric operat-
ing state, with the rotor plane normal to the wind vector.
With the blade held stationary and a free-stream wind speed, U ¥ , equal to 20 m/s, the
Reynolds number, Re , at station 0.30 R equals 0.98 x 10 6 . At a U ¥ of 30 m/s Re equals 1.46
x 10 6 . These Reynolds numbers encompass the range for the 11 fastest rotating blade data
points, thus establishing a consistent baseline against which the rotating blade data can be
compared. For the stationary blade, C n increases progressively with LFA , reaches a relative
maximum of 0.94 at LFA = 32.2°, and thereafter remains approximately level.
To disclose hysteresis effects, if any, the stationary blade normal force curve for each
wind speed consists of two data sets: One corresponding to increasing LFA , and the other for
decreasing LFA . These two curves exhibited close agreement, testifying to the absence of
hysteresis effects for stationary operation, and disqualifying hysteresis as an explanation for
rotational augmentation.
Figure 5-46. Comparison of sectional normal force coefficients, C n , at an inboard sta-
tion for stationary (parked) and rotating operations of the UAE Phase VI wind turbine,
as functions of the local inflow angle, LFA . Differences between rotating and stationary C n
values were all consistently larger for inboard locations