Optimisation du système de surveillance des hélicoptères pour l'amélioration du diagnostic et de la maintenance. (Optimization of fault diagnosis in helicopter health and usage monitoring systems)

A Health and usage Monitoring System (HUMS) anticipates discrepancies in therotorcraft drive-train, giving the operator an opportunity to perform correctivemaintenance before any damage becomes critical. In addition to usage spectrum analysis,a HUMS deploys vibration monitoring as a means to detect propagating faults. Thismethod consists in comparing in-flight vibration recordings to a normal state baseline. Arecurrent problem with this approach is that this baseline is aircraft specific and subject tochange between major overhauls, forcing the operator to relearn the baseline on regularbases.This study presents methods for evaluating the time-progression of the drive-trainvibration signature. By studying fluctuation in vibration signature over time, it is possibleto detect events such as maintenance actions and fault propagations without any aircraftspecific baseline. Several progression analysis methods are tested, both parametricmodels and filter-banks. Finally, progression analysis is used as a basis for faultdetection, and is shown to produce better results than traditional methods.