SlantNet: A Lightweight Neural Network for Thermal Fault Classification in Solar PV Systems
Electronics,
Год журнала:
2025,
Номер
14(7), С. 1388 - 1388
Опубликована: Март 30, 2025
The
rapid
growth
of
solar
photovoltaic
(PV)
installations
worldwide
has
increased
the
need
for
effective
monitoring
and
maintenance
these
vital
renewable
energy
assets.
PV
systems
are
crucial
in
reducing
greenhouse
gas
emissions
diversifying
electricity
generation.
However,
they
often
experience
faults
damage
during
manufacturing
or
operation,
significantly
impacting
their
performance,
while
thermal
infrared
imaging
provides
a
promising
non-invasive
method
detecting
common
defects
such
as
hotspots,
cracks,
bypass
diode
failures,
current
deep
learning
approaches
fault
classification
generally
rely
on
computationally
intensive
architectures
closed-source
solutions,
constraining
practical
use
real-time
situations
involving
low-resolution
data.
To
tackle
challenges,
we
introduce
SlantNet,
lightweight
neural
network
crafted
to
classify
efficiently
accurately.
At
its
core,
SlantNet
incorporates
an
innovative
Slant
Convolution
(SC)
layer
that
utilizes
slant
transformation
enhance
directional
feature
extraction
capture
subtle
gradient
variations
essential
detection.
We
complement
this
architectural
advancement
with
thermal-specific
image
enhancement
augmentation
strategy
employs
adaptive
contrast
adjustments
bolster
model
robustness
under
noisy
class-imbalanced
conditions
typically
encountered
field
applications.
Extensive
experimental
validation
comprehensive
panel
defect
detection
benchmark
dataset
showcases
SlantNet’s
exceptional
performance.
Our
achieves
95.1%
accuracy
computational
overhead
by
approximately
60%
compared
leading
models.
Язык: Английский
Current- and Vibration-Based Detection of Misalignment Faults in Synchronous Reluctance Motors
Machines,
Год журнала:
2025,
Номер
13(4), С. 319 - 319
Опубликована: Апрель 14, 2025
Misalignment
faults
in
drive
systems
occur
when
the
motor
and
load
are
not
properly
aligned,
leading
to
deviations
centerlines
of
coupled
shafts.
These
can
cause
significant
damage
bearings,
shafts,
couplings,
making
early
detection
essential.
Traditional
diagnostic
techniques
rely
on
vibration
monitoring,
which
provides
insights
into
both
mechanical
electromagnetic
fault
signatures.
However,
its
main
drawback
is
need
for
external
sensors,
may
be
feasible
certain
applications.
Alternatively,
current
signature
analysis
(MCSA)
has
proven
effective
detecting
without
requiring
additional
sensors.
This
study
investigates
misalignment
synchronous
reluctance
motors
(SynRMs)
by
analyzing
signals
under
different
conditions
operating
speeds.
Fast
Fourier
transform
(FFT)
applied
extract
characteristic
frequency
components
linked
misalignment.
Experimental
results
reveal
that
amplitudes
rotational
harmonics
(1xfr,
2xfr,
3xfr)
increase
presence
misalignment,
with
1xfr
exhibiting
most
stable
progression.
Additionally,
acceleration-based
proves
a
more
reliable
tool
compared
velocity
measurements.
findings
highlight
potential
combining
enhance
SynRMs,
improving
predictive
maintenance
strategies
industrial
applications
Язык: Английский
Misalignment identification and uncertainty quantification of rotor systems using Laplace prior-enhanced sparse Bayesian learning
Journal of the Brazilian Society of Mechanical Sciences and Engineering,
Год журнала:
2025,
Номер
47(7)
Опубликована: Май 17, 2025
Язык: Английский
Leveraging Digital Twins and AI for Enhanced Gearbox Condition Monitoring in Wind Turbines: A Review
Applied Sciences,
Год журнала:
2025,
Номер
15(10), С. 5725 - 5725
Опубликована: Май 20, 2025
Wind
power
plays
a
significant
role
in
sustainable
energy
production,
but
the
reliability
of
wind
turbines
depends
heavily
on
integrity
their
gearboxes.
Gearbox
failures
can
lead
to
downtime
and
operational
disruption.
In
this
context,
paper
provides
an
overview
evolution
gearbox
monitoring
techniques,
culminating
emergence
digital
twin
(DT)
technology.
We
explore
application
DT
technology
condition
monitoring,
focusing
two
critical
components:
bearings
gears.
This
includes
comprehensive
review
methodologies
involving
model-based
approaches
data-driven
techniques
using
signal
processing
(SP)
artificial
intelligence
(AI)
algorithms.
address
challenges
“learning
with
minimal
knowledge”
propose
framework
for
effective
Finally,
we
discuss
future
research
directions
potential
contributions
advancing
field
through
continued
development
implementation
DT-based
solutions.
Язык: Английский