|
The pitch links of rotor blades are essential
hardware that provide direct control to rotorcraft. The
pitch-link loads undergo large changes in magnitude
as a result of flight conditions that range from those of
relatively benign level flight to those associated with
severe, complex maneuvers. In the present study, a
"complex" maneuver was defined as one that
involved simultaneous non-zero aircraft angle of
bank (associated with turns) and aircraft pitch rate
(associated with a pull-up or a push-over). Also, since
a typical rotor blade pitch link operates in a highly
dynamic environment, the pitch-link loads obtained
from flight tests have associated with them a greater
degree of uncertainty. Analytical prediction of pitch-link
loads is thus difficult, and methods that provide
accurate results are highly desirable. The objectives
were (1) to obtain physical insights into the nature of
complex maneuvers and (2) to apply neural networks
to efficiently characterize maneuver-related rotorcraft
blade pitch-link loads. The NASA/Army UH-60A
Airloads Program database was used.
Since existing load factors do not represent the
above-defined complex maneuvers, a new physics-based
parameter, the maneuver-load-factor (MLF)
was derived and used. The MLF includes both the
aircraft angle of bank and pitch rate, resulting in a
single parameter. Figure 1 shows the MLF and the
pitch-link loads variations with forward speed.
Approximately 80 test data points (with pitch-link
loads greater than 1,000 pounds) were considered.
The associated neural network application involved
five inputs, namely, the MLF and four standard
parameters. The neural network output was the peak
oscillatory pitch-link load. Figure 2 shows the
presently obtained finer correlation (±400 pounds
error-band). This correlation was obtained using a
single-hidden-layer, back-propagation neural network
with 10 processing elements; it was trained for
600,000 iterations and had a final root-mean-square
error of 0.07.
Point of Contact: S. Kottapalli
(650) 604-3092
skotapalli@mail.arc.nasa.gov
Back To Top
Previous Paper
Return to Global Civil Aviation
Next Paper |
|
Fig. 1. Maneuver-load-factor and pitch-link load
variations with speed (advance ratio).
|
|
Fig. 2. Pitch-link load correlation using physics- and
neural-network-based approach.
|
|
