Chinese Journal of Aeronautics,
Journal Year:
2024,
Volume and Issue:
unknown
Published: March 1, 2024
Swarm
Intelligence
(SI)
is
a
collective
behavior
that
emerges
from
interaction
between
individuals
in
group.
Typical
SI
includes
fish
schooling,
ant
foraging,
bird
migration,
and
so
on.
A
great
deal
of
models
have
been
introduced
to
characterize
the
mechanism
SI.
This
article
reviews
several
typical
classifies
them
into
four
categories:
self-driven
particle
models,
with
Boids
model
as
primary
example;
pheromone
communication
including
colony
which
serves
foundation
for
optimization;
leadership
decision
utilizing
hierarchical
dynamics
pigeon
flock
prime
instance;
empirical
research
employ
topological
rule
starling
classic
model.
On
this
basis,
each
type
elaborated
upon
terms
its
overview,
applications,
evaluation.
More
specifically,
multi-agent
swarm
control,
path
optimization
obstacle
avoidance,
formation
consensus
trajectory
tracking
dense
crowd
social
networks
analysis
are
surveyed
application
category,
respectively.
Furthermore,
more
precise
effective
modeling
techniques
described.
Limitations
potential
directions
further
exploration
study
presented.
Nondestructive Testing And Evaluation,
Journal Year:
2022,
Volume and Issue:
38(2), P. 275 - 296
Published: Sept. 5, 2022
The
occurrence
of
multiple
faults
is
a
practical
problem
in
the
bearings
rotating
machines,
and
early
diagnosis
such
issues
an
intelligent
manner
vital
era
industry
4.0.
present
work
investigated
various
combinations
bearing
faults,
including
dual
fault
conditions.
Two
prevalent
methods
were
employed:
vibration
monitoring
using
time-frequency
scalograms
extracted
through
Continuous
Wavelet
Transform
(CWT)
non-invasive
Infrared
Thermography
(IRT).
A
2-D
Convolutional
Neural
Network
(CNN)
was
used
to
classify
conditions
automated
feature
extraction.
proposed
methodology
validated
at
two
constant
speed
19
Hz
29
continuously
accelerated
decelerated
range
-
Hz.
Adequate
accuracy
achieved
both
case
vibration-based
diagnosis,
with
99.39
%
99.97
%.
Meanwhile,
IRT-based
100
classification
for
all
These
results
signify
robustness
reliability
varying
