170 J.M. Brown et al. Fig. 15.19 Comparison of singe healthy case to the average PSD of all healthy data 0 1 2 3 4 5 10−5 100 105 Healthy Case Comparison Motor Speed: 8636.1 [rpm], 143.9 [Hz] Normalized Frequency Intensity [(m/s2)2] Healthy Case Average Healthy Case Motor Freq Fig. 15.20 Comparison of PSD of propeller with asymmetric tip damage to healthy average 0 1 2 3 4 5 10−5 100 105 Asymmetric Tip Damage Motor Speed: 8641.2 [rpm], 144.0 [Hz] Normalized Frequency Intensity [(m/s2)2] Damaged Case Average Healthy Case Motor Freq Fig. 15.21 Comparison of PSD of propeller with symmetric tip damage to healthy average 0 1 2 3 4 5 10−5 100 105 Symmetric Tip Damage Motor Speed: 8450.4 [rpm], 140.8 [Hz] Normalized Frequency Intensity [(m/s2)2] Damaged Case Average Healthy Case Motor Freq The average was also compared to several damage cases. In Fig. 15.20, the propeller with asymmetric tip damage shows multiple peaks at off-harmonic frequencies that are easily distinguished from the healthy case. Figure 15.21 also shows the PSD of the ARS acceleration signal for the symmetric tip damage case. This is less easily distinguished than the asymmetric case, but still distinctly different than the healthy data. Both cases show the potential for damage metrics based on the deviation in the frequency domain. Quad-Rotor For the quad-rotor, each damaged test has three systems of healthy components to compare to the damaged component. Although each motor rotates at a different speed for a given time, using the ARS signal normalizes the frequencies with each motor’s individual frequency as discussed before. Similar to the single rotor platform, the healthy arms are dominated by the motor frequency and its harmonics. Due to the coupling between arms and the increased complexity of the quad-rotor system, the healthy signals exhibit many other small peaks. Figure 15.22 is a PSD of each of the four arms on the quad-rotor for the same asymmetrically damaged tip case demonstrated on the single rotor platform. Figure 15.23 shows the PSD of one of the structural damage tests performed through loosening of bolts. Both types of damage produce intense low frequency and off-harmonic vibrations that are not present in the arms with healthy components. In addition, the clear detection of structural damage will be extremely useful to alter or abort the flight of an MR vehicle before a motor mount fails catastrophically. For both the single rotor and quad-rotor platforms, frequency analysis of the vibration data along a single in-plane accelerometer shows distinctly different trends for the damaged case. Both platforms indicate deviation in the frequency could be used in determining damage metrics.
RkJQdWJsaXNoZXIy MTMzNzEzMQ==