Environmental Engineering Reference
In-Depth Information
Fig. 8.4 Schematic diagram
showing thermal control
system using direct optical
sensing
8.4 Blade Thermodynamics
Different areas on wind turbine blades need significantly different levels of thermal
power and flux for de-icing (Fig.
8.5
). Thus, optimally distributing the required
thermal energy across the blade can save a significant amount of thermal energy in
de-icing systems. The highest heat flux required is at the leading edge of the tip of
the blade. As such, a preliminary step in designing a suitable thermal actuation
method for de-icing is calculating the heat flux requirement for this most critical
region of the blade. Figure
8.5
shows the calculated heat flux requirement (due to
convection loss) versus nondimensional chord position x/c and nondimensional
span-wise radius r/R
tip
at a given atmospheric condition. The required parameters
are angle of attack AoA of the blade, blade geometry, angular velocity x of the
blade, and wind speed u
w
. Figure
8.5
shows the characteristics and behavior of the
convection loss for laminar flow over the blade. Negative values of x/c are for the
lower blade surface, positive values of x/c are for the upper blade surface, and x/
c = 0 is leading edge of the blade. The peak value of heat flux at the leading edge
of the blade tip for this simulation is about 1,800 W/m
2
for a 500 KW wind
turbine. This calculation uses experimental local values of Nusselt numbers at
different Reynolds numbers and Prandtl numbers for a NACA 63421 airfoil using
experimental convective heat correlations [
18
]. This airfoil has a nonsymmetrical
profile which creates different trends below and above the stagnation point for
Nusselt numbers and the convective heat transfer coefficient [
18
].
The convective coefficient of heat transfer (h) at each point is calculated as
h
¼
Nu
x
k
air
c
ð
8
:
1
Þ
where Nu
x
is the local Nusselt number on the airfoil, k
air
is the thermal conduc-
tivity of air, and c is the chord length of the blade.
The convective loss heat flux (q
conv
) is calculated as
