PEER REVIEW
20
th
Australian International Aerospace Congress, 27-28 November 2023, Melbourne
ISBN number: 978-1-925627-66-4
Signal processing informed deep learning for failure
detection in a fleet of multi-stage planetary gearboxes with
limited knowledge about characteristic frequencies
Jan Helsen
1
, Fabian Perez
1,2
, Faras Jamil
1,2
, Jérôme Antoni
3
and Cédric Peeters
1
1
: Department of Applied Mechanics, Vrije Universiteit Brussel, Brussels, Belgium.
Email: jan.helsen@vub.be
2
: Artificial Intelligence lab, Vrije Universiteit Brussel, Brussels, Belgium
3
: Univ Lyon, INSA Lyon, LVA, Villeurbanne, France.
Email: jerome.antoni@insa-lyon.fr
Abstract
Condition monitoring of multi-stage planetary gearboxes is a complex challenge given the fact
that gears the large number of rotating subcomponents. Typically, the large number of gears
creates many harmonic excitations masking bearing signatures. Different state-of-the-art
harmonic removal methods, e.g. cepstrum liftering, are available. Such methods have been
shown to be automatable. However, exact characteristic frequency values are not always known
for such gearboxes in commercial systems. Estimation of gear teeth numbers has been shown
in literature. Bearing frequency determination is much more challenging. Deep learning
methods can offer a solution. Once the harmonic content is removed, focus can be on the
detection of modulations linked to bearing problems. Spectral coherence methods have shown
to be highly reliable for such detection. However, if no info is available about normal behaviour
in the coherence maps it is essential to detect which modulations are changing over time. This
paper investigated the use of deep learning auto-encoders trained on spectral coherence maps
as core component in an anomaly detection framework to identify changes in modulations. The
auto-encoders are trained with large sets of healthy data. In this way we maximally use available
data and avoid the need of large sets of labelled failure data. Typically, such data is not available
for most operators. To illustrate the methodology data of six offshore wind turbines is used.
Keywords: Planetary gear, condition monitoring, vibration, deep learning, physics-informed
Introduction
Condition monitoring (CM) targeting fast and accurate detection of problems is an important
aspect in a typical predictive maintenance strategy, since the logistics of spare parts and repair
equipment (e.g. crane vessels for wind turbines [1]) need to be optimized to avoid large
downtimes. Faults need to be detected early to allow for alarming. However, providing a general
alarm as a diagnostics feature does not suffice. CM methods need to be able to distinguish
between the fault types and pinpoint the analyst to the subcomponent that needs to be replaced.
Typically, the characteristic frequencies linked to the rotational fault signature are used for this
purpose. Nonetheless, many end-users of machines have limited access to the details of these
frequencies as they are not always disclosed by the manufacturer or different machine variants
can have slightly different subcomponents (e.g. bearings). In literature, methods exist to
estimate the characteristic frequencies linked to gear teeth numbers. An example is the method
of Sawalhi and Randall targeting the use of a fine-tuned harmonic-sideband cursor approach
[2]. For the estimation of bearing fault frequencies these methods are existing much less. An
