Environmental Engineering Reference
In-Depth Information
Blade
(Aero-Thermo
Plant)
Blade Temperature at Each Thermocouple
Existence, Type, and Thickness of Ice at Each Optical Sensor
Fig. 8.13 Closed-loop control schematic diagram. T
d
= 2 C is the desired blade surface
temperature and e
j
(t) is the error signal between the desired blade temperature T
d
and the actual
blade temperature at the jth thermocouple
takes less than 2 s in the loop. As such, we use a closed-loop control update rate of
0.5 Hz. This is sufficiently fast for the desired thermal response bandwidth of
0.001-0.01 Hz for the actuation network. Both the control system software and
OFDR software have been developed in LabVIEW [
29
].
The result of the numerical ice detection algorithm is sent to the closed-loop
control system at 0.5 Hz. The command voltage output of the control software is
amplified through a custom fabricated multichannel op-amp circuit and fed to the
distributed resistors on the blade.
8.8 Computational Model Validation with Experiments
Accurate prediction of the aero/thermodynamic response of a rotating blade
covered with ice/snow and distributed electrothermal resistors is complex due to
the coupled nonlinear dynamics of turbulent wind, the ice/water phase change, and
multiple distributed heat sources on the blade. Experimental validation is required
to achieve a reasonably accurate dynamic model for such a complex system. In this
section, we compare the transient thermal response of a composite blade under-
going distributed thermal actuation using a computational model in ANSYS [
30
]
with results from an experimental run performed using our test setup. The
geometry of the hollow blade with distributed resistors is modeled in SolidWorks
[
31
]. This geometry is imported into the ANSYS Workbench environment for
computational simulations (Fig.
8.14
). The transient thermal module in ANSYS is
used to compute the temperature variation as a function of time on the blade for
different heater layouts. In the simulations explained in Sect.
8.9.2
, a variable time
step is implemented with a minimum time step of 0.05 s and maximum time step
of 0.5 s. In the ANSYS analysis setting, heat and temperature convergence modes
are
activated
to
achieve
better
stability
and
convergence
in
the
numerical
simulations.
For our experiment, a temperature sensor is placed between two of the middle
resistors in the second and third row to record the variation of the temperature
(Fig.
8.15
). Figure
8.16
shows the experimentally applied input voltage to all of
