Wind Turbine Aerodynamics Part B: Turbine Blade Flow Fields - Wind Turbine Technology: Fundamental Concepts of Wind Turbine Engineering - page 366

the absence of dynamic stall, while the second represents the vortex-dominated, dynami-
cally-stalled flow field. Both of these loading components are quantified by means of a set
of nonlinear differential equations containing a significant number of coefficients that need
to be extracted from unsteady experimental measurements.
Clearly, the Boeing-Vertol model formulation is mathematically simpler and requires
fewer empirical inputs than the ONERA model. However, comparisons of model outputs
with measured data have yielded apparently inconclusive results regarding relative model
fidelity. Bierbooms [1992] reported that ONERA model fidelity was superior to that of other
models that required fewer inputs. In contrast, Yeznasni et al. [1992] found that the ONERA
model generally deviated more from measured results than the Boeing-Vertol model. It
should be noted that both researchers were careful to stress the importance of experimental
data quality, as well as correct analysis and application of these data.
The Leishman-Beddoes dynamic stall model [Pierce and Hansen 1995, Leishman 2006,
Gupta and Leishman 2006] takes advantage of a more physics-based characterization of dy-
namic stall, which is implemented in three parts. The first part uses inviscid thin airfoil
theory to quantify low angle of attack aerodynamics, while the second part superimposes dif-
ferent analytical relationships to account for viscous effects as angle of attack increases. The
third part of the Leishman-Bedoes model replicates the formation, convection and growth,
and ultimate shedding of the dynamic stall vortex, as illustrated in Figure 5-55.
Frequency of Occurrence of Dynamic Stall
Rotational augmentation and dynamic stall, as described above, are not anomalous
events. Though operational turbines lack instrumentation that would enable specific detec-
tion, both events take place with some regularity during wind turbine operation, across the
operating envelope. Figure 5-58 illustrates the operating conditions during which these phe-
nomena occurred for the UAE Phase VI turbine during the NASA Ames wind tunnel test. The
square marked “Nom” (nominal) near the origin of the plot represents the range of yaw error
and test airspeed, U ¥ , that was not influenced by rotational augmentation or dynamic stall.
Rotational augmentation predominates for yaw error and U ¥ corresponding to the rectangular
region labeled “Rot Aug”, which lies immediately above the “Nom” square. Dynamic stall is
dominant for conditions demarcated by the large rectangular region marked “Dynamic Stall”
that fills the upper right portion of Figure 5-58.
Figure 5-58. UAE Phase VI wind turbine operating envelope, showing the observed
dynamic stall regime and typical gust and direction change equivalents.