Environmental Engineering Reference
In-Depth Information
Validation of these models continues, using turbulence data from sites where tur-
bines have experienced wind damage. Data from U.S. sites were collected at the Paciic
Northwest Laboratory. Turbine wake turbulence data are currently shared under an
International Energy Agency (IEA) agreement amongst eight countries [Milborrow and
Ainslie 1992]. A better understanding of small-scale turbulence is considered a major key
to improved fatigue life and performance in future wind turbines.
Aerodynamics
While the fundamental aerodynamics of wind turbines was well understood over half
a century ago, adequate analytical tools were not readily available to assist designers to
rapidly and accurately predict power performance. In addition, the characteristics of
hundreds of different airfoils for airplane wings, propeller blades, and helicopter rotors,
well-documented in wind-tunnel test reports, were less useful to wind turbine blade
designers than had been expected. Trends in aviation led to relatively thin airfoils, while
wind turbines (with less concern for weight and more for extremely long fatigue life)
require moderate-to-thick airfoils that are proportionately stronger while retaining good
performance. In addition, the
lift and drag
characteristics of most available airfoils had
been measured thoroughly for a relatively narrow range of
angles of attack
, only slightly
beyond the angles of
stall
(+/-18 to 20 degrees or so). Yet, portions of wind turbine blades
often operate at angles of attack well beyond stall.
The fundamentals of wind turbine aerodynamics, started by Prandtl and Betz, were re-
derived and extended by the initial work of Wilson and Lissaman at Oregon State
University and AeroVironment, Inc. [Wilson and Lissaman 1974, 1976]. A compendium
of aerodynamics design methods (as well as methods for other aspects of wind turbine
design) was developed at the Massachusetts Institute of Technology [Miller
et al
. 1978].
From this base, persistent in-house and contract research at the NASA Lewis Research
Center, the Solar Energy Research Institute, and Sandia Laboratories led to continual
improvements in aerodynamic modeling tools and understanding of rotor aerodynamic
issues. Early models based on
single stream tube
analysis and simple
blade element
theory
were replaced by computer codes using
multiple stream tubes
and
lifting line
models.
These, in turn, have been giving way to improved multiple stream tube models,
lifting
surface
models, and models incorporating three-dimensional low and turbulence effects.
At irst, both aerodynamic analyses and wind tunnel tests consistently under-predicted
the power output and blade loads encountered in the ield, for all but the smallest wind
turbines. Causes of these discrepancies included
three-dimensional low
,
inlow turbulence
,
and
dynamic stall
during rapid changes in angles of attack.
Airfoils speciically tailored to the needs of wind turbine designers were developed in
the 1980s, including
natural laminar low
airfoils, new moderate-to-thick airfoils for
increased strength and stiffness, and airfoils with
lower sensitivity to roughness
. Accumu-
lated dirt and insects have caused signiicant energy losses in many wind power stations.
Development of advanced airfoils continues today, aimed particularly at increased energy
capture in low-to-moderate winds and limited output power and loads in higher winds.
Structural Dynamics and Fatigue Life Design
The limitations of analytical models for wind characteristics and aerodynamic behavior
were aggravated in the mid-1970s by serious limitations in the state-of-the-art for predicting
structural dynamic loads and fatigue life. Widely-used structural analysis codes such as
NASTRAN
were useful for predicting
natural frequencies
and static loads, but were totally
inadequate for calculating
aeroelastic
dynamic loads, in which wind forces are coupled to
Search WWH ::
Custom Search