PEER REVIEW
20
th
Australian International Aerospace Congress, 27 February- March 1 2023, Melbourne
ISBN number: 978-1-925627-66-4
Please select category below:
Normal Paper
Student Paper
Young Engineer Paper
Accelerated Gearbox Degradation Monitoring using a
Combination of Vibration and Acoustic Emission Features
Cristobal Ruiz-Carcel
1
, Andrew Starr
1
1
Cranfield University, Cranfield, Bedfordshire MK430AL, United Kingdom
Abstract
Commercial HUM systems usually rely on vibration measurements to assess the condition of
shafts, gears, and bearings in rotorcraft transmission systems. Recent research has shown that
acoustic emissions (AE) can be advantageous in the detection of mechanical faults in helicopter
gearboxes. In particular AE has been successful in detecting very early small defects on bearings
and gears, increasing sensitivity during the fault detection process and providing extra time for
maintenance planning. Additionally, knowledge about the gear degradation process can be used
to define rules that combine AE and vibration sensing technologies, adding robustness to the
fault detection process by minimising false alarms.
In this study, a mixture of AE and vibration signal features were used to monitor accelerated
natural degradation in an experimental gear rig. The suggested signal processing technique first
defines individual alarm levels for each feature based on statistical analysis of the signals during
healthy operation. Then proposes knowledge-based rules to merge them into a single health
indicator for fault detection. This combined approach has the benefit of producing an early
fault warning, and increases the confidence in the fault detection procedure as the features reach
threshold values systematically, reducing the chances of false alarms.
Keywords: acoustic emission, vibration, feature extraction, natural degradation, data fusion
1. Introduction
Health and Usage Monitoring Systems (HUMS) typically use predefined vibration features to
assess the condition of rotorcraft transmission systems [1]–[3]. Commercial HUMS make use
of different vibration analysis methods to detect faults in bearings, gears and shafts. Condition
Indicators (CI’s) are key vibration features extracted from the acquired vibration signals, which
can be related to specific mechanical faults [4]. In HUMS a range of different CI’s are extracted
from vibration data to characterize component health.
Acoustic emissions (AE) are defined in the field of machine monitoring as transient elastic
waves produced by the interface of two components or more in relative motion [5]. Typical AE
sources include impacts, crack growth, friction, turbulence, material loss, cavitation, leakage
etc. Its main benefit against vibration analysis and oil analysis is the capability to detect faults
earlier due to the high sensitivity offered by AE [6]. On the other hand, the main drawback of
AE is the difficulty in processing, interpreting and manipulating the acquired data [7], [8].
Vibration-based gearbox monitoring is well established, however the application of AE to this
field is still in its early stages [6], [9] and it is difficult to see it implemented in commercial
tools. In the area of HUMS some research has been carried out in recent years to prove the
