Sensors,
Journal Year:
2024,
Volume and Issue:
24(5), P. 1680 - 1680
Published: March 5, 2024
This
study
introduces
a
novel
nonlinear
dynamic
analysis
method,
known
as
beluga
whale
optimization–slope
entropy
(BWO-SlEn),
to
address
the
challenge
of
recognizing
sea
state
signals
(SSSs)
in
complex
marine
environments.
A
method
underwater
acoustic
signal
recognition
based
on
BWO-SlEn
and
one-dimensional
convolutional
neural
network
(1D-CNN)
is
proposed.
Firstly,
particle
swarm
(PSO-SlEn),
BWO-SlEn,
Harris
hawk
(HHO-SlEn)
were
used
for
feature
extraction
noise
SSS.
After
1D-CNN
classification,
found
have
best
effect.
Secondly,
fuzzy
(FE),
sample
(SE),
permutation
(PE),
dispersion
(DE)
extract
features.
highest
rate
compared
with
them.
Finally,
other
six
methods,
rates
SSS
are
at
least
6%
4.75%
higher,
respectively.
Therefore,
methods
proposed
this
paper
more
effective
application
recognition.
Sensors,
Journal Year:
2022,
Volume and Issue:
22(16), P. 6281 - 6281
Published: Aug. 21, 2022
A
rolling
bearing
fault
diagnosis
method
based
on
whale
gray
wolf
optimization
algorithm-variational
mode
decomposition-support
vector
machine
(WGWOA-VMD-SVM)
was
proposed
to
solve
the
unclear
characterization
of
vibration
signal
due
its
nonlinear
and
nonstationary
characteristics.
algorithm
(WGWOA)
by
combining
(WOA)
(GWO),
decomposed
using
variational
decomposition
(VMD).
Each
eigenvalue
extracted
as
eigenvector
after
VMD,
training
test
sets
model
were
divided
accordingly.
The
support
(SVM)
used
optimized
WGWOA.
validity
this
verified
two
cases
Case
Western
Reserve
University
data
set
laboratory
test.
results
show
that
in
University,
compared
with
existing
VMD-SVM
method,
accuracy
rate
WGWOA-VMD-SVM
five
repeated
tests
reaches
100.00%,
which
preliminarily
verifies
feasibility
algorithm.
In
case,
diagnostic
effect
is
backpropagation
neural
network,
SVM,
VMD-SVM,
WOA-VMD-SVM,
GWO-VMD-SVM,
WGWOA-VMD-SVM.
Test
highest,
a
single
99.75%,
highest
above
six
methods.
WGWOA
plays
good
role
optimizing
VMD
SVM.
without
overlap.
has
better
convergence
performance
than
WOA
GWO,
further
superiority
among
research
can
provide
an
effective
improvement
for
technology.
IEEE Access,
Journal Year:
2023,
Volume and Issue:
11, P. 12887 - 12910
Published: Jan. 1, 2023
The
advent
of
Industry
4.0
has
resulted
in
the
widespread
usage
novel
paradigms
and
digital
technologies
within
industrial
production
manufacturing
systems.
objective
making
operations
monitoring
easier
also
implied
more
effective
data-driven
predictive
maintenance
approaches,
including
those
based
on
machine
learning.
Although
approaches
are
becoming
increasingly
popular,
most
traditional
learning
deep
algorithms
experience
following
three
major
challenges:
1)
lack
training
data
(especially
faulty
data),
2)
incompatible
computation
power,
3)
discrepancy
distribution.
A
new
technique,
such
as
transfer
learning,
can
be
developed
to
overcome
issues
related
for
maintenance.
Motivated
by
recent
big
interest
towards
computer
science
artificial
intelligence,
this
paper
we
provide
a
systematic
literature
review
addressing
research
with
focus
aims
define
context
introducing
specific
taxonomy
relevant
perspectives.
We
discuss
current
advances,
challenges,
open-source
datasets,
future
directions
applications
from
both
theoretical
practical
viewpoints.
