| Resumen |
A soft histogram feature-extraction algorithm (SH-FEA) was developed, to extract damage-sensitive information from measured-response data of a tie-bar component of the main rotor hub of a Lynx helicopter. The feature-extraction algorithm was based on fuzzy-set theory. The SH-FEA consisted of normalizing the incoming signal to the range of 1-1 and obtaining its histogram. Feature vectors were extracted, using the SH-FEA for every cycle of the 4 measured signals for the test under consideration, which consisted of 1296 cycles of GAG load testing until tie-bar failed. Six fuzzy bins were also defined, to obtain the soft histograms, with centers at -1, -0.6, -0.2, 0.2, 0.6, and 1, along with a width of 0.4. A reference vector was calculated by averaging the feature vectors corresponding to cycles 11-60, with the available extracted feature vectors. |
