|
As part of NASA's overall goal to improve aviation safety, fundamental research is being conducted to support the development of systems that monitor critical rotating components, such as those found in helicopter gearboxes and aircraft engines. Damage detection based on observed patterns of surface vibration requires insight into the statistical properties of complex signals that are produced by interacting elements within the system, as well as the effects of in-flight maneuvering.
Table 1 shows the tendency of vibration signals to remain constant (i.e., stationary) over 34-second recording intervals. The extent of nonstationarity is dependent on both maneuver state and aircraft type, which other evidence suggests are related to vehicle weight and engine torque variations. Hence, these findings provide an essential link for developing damage-detection algorithms that are not deceived by nonstationarity into making costly "false-alarms."
In FY2000 we conducted the first flight tests of the Ames' OH-58C aircraft and made comparisons of vibratory signals with an identical transmission tested at the NASA Glenn Helicopter Transmission Facility. Table 2 shows the results of an effort to parse signal energy for transmission component sources from these two tests. This work reveals similarities and differences between real flight and test rig vibration signals, information that is necessary to develop damage-detection algorithms with low false-alarm rates and high fault detection. Work in this area was reported at the American Helicopter Society Annual National Forum in 2000.
Point of Contact: Edward Huff
(650) 604-4870
ehuff@mail.arc.nasa.gov
Back To Top
Previous Paper
Return to Revolutionary Aviation
Next Paper |
