| Autor: |
Mauricio, Alexandre, Talon, Arnaud, Agathe, Vercoutter, Janssens, Karl, Gryllias, Konstantinos |
| Předmět: |
|
| Zdroj: |
EA National Conference Publications; 2023, p238-246, 9p |
| Abstrakt: |
The components of an engine's helicopter gearbox are vulnerable to fatigue and therefore Health and Usage Monitoring Systems are intended to be developed, focusing towards early, accurate and on time detection of degradation's initialisation with limited false alarms and missed detections. The main aim of HUMs is to enhance the helicopters' engine operational reliability and functionality and to improve flight airworthiness. Bearings are one of the components of essential interest of helicopter's engine drivetrains, as they support the rotating components or gears, and early degradation detection is necessary to prevent sudden breakdown. On the other hand, bearing signals are often masked by noise and other vibration sources, which further challenges their early detection. Over the recent decade, several cyclostationary tools have been proposed in order to extract the bearing health state information from vibration data. These methods are based on the demodulation of the signals using either the Hilbert Transform or the Spectral Correlation and/or Coherence, in tandem with band pass filtering using band selection tools. In this paper, the performance of several cyclostationary-based indicators targeting to early bearing degradation detection is evaluated on a special dataset, including vibration and particle data, captured on a dedicated test bed of a helicopter engine drivetrain during a lifetime test. A bearing with an inner race indented in three positions was mounted on the test bed and was run until its end of life, when the full surface of the inner race was spalled. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
