A Self-supervised Learning Approach for Anomaly Detection in
Rotating Machinery
Fabrizio De Fabritiis
1,2
, Konstantinos Gryllias
1,2,3
1
Department of Mechanical Engineering, Division LMSD, KU Leuven, Celestijnenlaan 300, Box 2420, 3001 Leuven, Belgium
2
Flanders Make@KU Leuven, Celestijnenlaan 300, Box 2420, 3001 Leuven, Belgium
3
Leuven.AI – KU Leuven Institute for AI, B-3000, Leuven, Belgium
ABSTRACT
Early fault detection in rotating machinery needs careful
expert analysis of vibration data for monitoring a component
state. Online methods that automatically set a threshold and
raise an alarm when the vibration signature is anomalous
are meant to efficiently manage key assets in a preventive
maintenance plan.
In recent years a focus has raised on data driven methods
in parallel with the increasing attention towards machine
learning and, particularly, deep learning models. In this re-
gard, for rotating equipment components, an important aspect
relates to labelled data scarcity for supervised training. On
the other hand, the advent of the Internet of Things allows
to gather data from multiple assets with relevant information
on the asset state itself. Self-supervised learning methods in
deep learning application are currently tackling this problem.
Investigating Self-learning approaches to integrate domain
knowledge and learn relevant features from unlabeled data is
therefore important for condition monitoring applications.
In this paper a methodology is proposed based on cycle
consistency representation learning for training an embedder
network on univariate unlabeled data. In order to learn a
distance metric in the embedding space the original data are
transformed to generate sequences of augmented inputs to
enforce learnable pattern similarity in the augmented pairs. A
differentiable cycle-consistency loss is chosen to maximize
the numbers of augmented pairs in the learned embedding
space that have minimum features distance. The pretext
task in the described self-supervised setting aims to train a
Fabrizio De Fabritiis et al. This is an open-access article distributed under the
terms of the Creative Commons Attribution 3.0 United States License, which
permits unrestricted use, distribution, and reproduction in any medium, pro-
vided the original author and source are credited.
feature extractor for discriminating dissimilar samples in the
embedding space by a distance metric and to provide a useful
representation for down-stream tasks.
The paper analyzes the performance of the approach for
anomaly detection in rotating machinery. The methodology
is tested on vibration data provided by the Center for Intelli-
gent Maintenance Systems (IMS), University of Cincinnati,
considering different accelerated life test campaigns. The
data were collected to monitor the fault development in bear-
ings and the model shows how the learned embedding space
discriminates effectively anomalous samples from normal
ones in the degradation stages of the bearings.
1. INTRODUCTION
In rotating machinery, critical mechanical components
are commonly monitored with vibration sensors directly
mounted onto the machine. Incipient faults are detectable
observing the variations in the vibration pattern (Henriquez,
Alonso, Ferrer, & Travieso, 2013) and the context in which
the machine is operating. While displacement sensors and
velocity sensors are preferred for specific applications, piezo-
electric accelerometers are widely adopted in most cases
for being affordable, small and sensitive to a wide range of
frequencies (Bogue, 2013).
Early fault detection in rotating machinery relies on model-
driven (Jalayer, Orsenigo, & Vercellis, 2021) or data-driven
(Liu & Gryllias, 2021) methods. A methodology for early
fault detection is effective if it enables the identification of
anomalies in e.g. vibration signals which are symptoms of
the dynamic forces originated from the initial development of
a defect in the monitored component. As an example, assess-
ing the state of gears and bearings is critical for drivetrains
in complex systems, such as wind turbines. A sudden failure
1
