Advancing Light Aircraft Health Monitoring with Flight Phase
Clustering
Oguz BEKTAS
1
, Jan PAPUGA,
2
, Sylvain KUBLER
3
1,3
SnT, University of Luxembourg, 6 Rue Richard Coudenhove-Kalergi, L-1359 Luxembourg, Luxembourg
oguz.bektas@uni.lu
sylvain.kubler@uni.lu
2
Evektor spol. s r. o., Leteck
´
a 1008, Kunovice, Czech Republic
jpapuga@evektor.cz
ABSTRACT
This paper addresses the clustering of flight phases of a light
aircraft for health monitoring using vibration data. The aim
is to improve diagnostic and prognostic functions. Grouping
condition monitoring data under similar operating conditions
is significant for predictive maintenance. Clustering also sup-
ports advanced analytics for fault detection and estimation of
remaining life. The proposed framework uses self-organizing
maps for flight phase clustering. The findings show that the
algorithm can recognize and classify flight phases in various
operational domains. Additionally, visualization of cluster
maps uncovers complex patterns and non-linear relationships
in sensor data under different flight conditions. As a follow-
up, analyzing the vibration properties within these estimated
clusters (regimes) provides insights from condition monitor-
ing data behavior during flight phases. The results confirm the
effectiveness of the method, but also confirm that determining
light aircraft regimes requires more focus due to their unique
flight patterns that are absent in commercial airliners. In this
context, this research has dealt with these unique patterns and
provided the foundation for a new model for clustering with
an attempt to contribute valuable insights into improving the
reliability and efficiency of light aircrafts.
1. INTRODUCTION
Clustering aircraft condition monitoring data across multiple
flight phases supports advanced analytics. Grouping data un-
der similar operating conditions is key to transfer valuable
data features to fault diagnosis and prognosis. Therefore, de-
termining flight phases has the potential to improve the air-
craft service life. In this regard, this study focuses on ob-
Oguz BEKTAS 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.
taining features from flight data by taking into account flight
phases. Although the literature have paid attention to var-
ious aspects of aircraft health monitoring (Kordestani, Or-
chard, Khorasani, & Saif, 2023), little is known about the
specific methods for flight phase identification when it comes
to light aircraft. Much of the existing research focuses on
general aviation or commercial airliners (Bektas, 2023; Lyu,
Thapa, & Desell, 2024). This leaves a gap in the litera-
ture regarding light aircraft. Furthermore, the integration
of advanced clustering techniques like self-organizing maps
(SOM) with further predictive maintenance analytics remains
under-explored. This gap is significant because the operating
conditions and maintenance needs of light aircraft differ from
those of larger aircraft. Addressing such a gap can improve
reliability while reducing maintenance costs. This study com-
bines SOM and vibration analysis to address this. This is a
novel methodology specifically designed for light aircraft.
Flight data monitoring is a routine of data collection and
analysis applied in commercial operations (Gavrilovski et al.,
2016). Therefore, it deals with big data where it is not pos-
sible to manually review all the collected information by hu-
man experts (Oehling & Barry, 2019). Instead, the literature
has witnessed various studies on the use of data mining and
machine learning techniques to analyze flight data. In par-
ticular, transient flight phases can reveal more information,
and clustering can help discover hidden insights from both
frequent and infrequent flight phases. For this purpose, a pre-
vious study by (Bektas, 2023) grouped the phases of flight
data according to the most important sensor readings. How-
ever, this study was for an airliner with regular flight regimes.
However, the same tool introduced for self organizing maps
(SOM) training by Wittek and Gao (Wittek & Gao, n.d.),
called Somoclu, can also provide advanced visual inspection
for light aircraft.
With such clustering mentioned above, flight phases can be
1
