Fractal and Fractional,
Год журнала:
2024,
Номер
8(12), С. 720 - 720
Опубликована: Дек. 5, 2024
Aiming
at
the
uncertainty
problem
caused
by
time-varying
modeling
parameters
associated
with
ship
speed
in
course
tracking
control
of
underactuated
surface
vessels
(USVs),
this
paper
proposes
a
algorithm
based
on
dynamic
neural
fuzzy
model
(DNFM).
The
DNFM
simultaneously
adjusts
structure
and
during
learning
fully
approximates
inverse
dynamics
ships.
Online
identification
lays
foundation
for
motion
control.
trained
DNFM,
serving
as
an
controller,
is
connected
parallel
fractional-order
PIλDμ
controller
to
be
used
ship’s
course.
Moreover,
weights
can
further
adjusted
tracking.
Taking
actual
data
5446
TEU
large
container
ship,
simulation
experiments
are
conducted,
respectively,
tracking,
under
wind
wave
interferences,
comparison
five
different
controllers.
This
proposed
overcome
influence
parameters,
desired
quickly
effectively.
Tribology Transactions,
Год журнала:
2025,
Номер
unknown, С. 1 - 21
Опубликована: Янв. 30, 2025
1.
Rolling
element
bearings
are
the
critical
parts
of
every
rotating
machinery
and
their
failure
is
one
main
reasons
machine
downtime
even
breakdown.
The
significance
early
detection
cannot
be
overstated,
as
it
plays
a
crucial
role
in
maintaining
proper
functioning
equipment,
enhancing
production
efficiency,
ensuring
safety.
Among
them,
envelope
analysis
most
effective
widely
used
approach,
working
according
to
principle
linear
filtering
process
signals
remove
undesirable
components.
However,
characteristic
frequencies
can
no
longer
evident
or
overwhelmed
due
weak
signal.
We
propose
an
fault
feature
extraction
method
that
combines
correlation
entropy
with
improved
2.5-dimensional
square
spectrum.
This
approach
designed
overcome
relatively
impact
failures,
which
easily
obscured
by
external
background
noise.
Specifically,
matrix
simplified
into
series
entropies,
formula
for
higher-order
spectrum
enhanced
achieve
superior
noise
removal
performance.
Furthermore,
enhance
frequency
(FCF),
Fourier
transform
has
been
refined
transform.
Simulation
results
demonstrate
our
proposed
excels
at
extracting
bearing
characteristics
detecting
inner
outer
ring
FCFs,
thereby
offering
practical
value.
PeerJ Computer Science,
Год журнала:
2025,
Номер
11, С. e2805 - e2805
Опубликована: Апрель 1, 2025
To
overcome
the
mechanical
limitations
of
traditional
inertia
weight
optimization
methods,
this
study
draws
inspiration
from
machine
learning
models
and
proposes
an
strategy
based
on
K-nearest
neighbors
(KNN)
principle
with
dynamic
adjustment
properties.
Unlike
conventional
approaches
that
determine
solely
number
iterations,
proposed
allows
to
more
accurately
reflect
relative
distance
between
individuals
target
value.
Consequently,
it
transforms
discrete
“iteration-weight”
mapping
($t\rightarrow
w$)
into
a
continuous
“distance-weight”
($d\rightarrow
w$),
thereby
enhancing
adaptability
capability
algorithm.
Furthermore,
inspired
by
entropy
method,
introduces
entropy-based
allocation
mechanism
in
crossover
mutation
process
improve
efficiency
high-quality
information
inheritance.
validate
its
effectiveness,
is
incorporated
Seahorse
Optimization
Algorithm
(SHO)
systematically
evaluated
using
31
benchmark
functions
CEC2005
CEC2021
test
suites.
Experimental
results
demonstrate
improved
SHO
algorithm,
integrating
logistic-KNN
crossover-mutation
mechanism,
exhibits
significant
advantages
terms
convergence
speed,
solution
accuracy,
algorithm
stability.
further
investigate
performance
improvements,
conducts
ablation
experiments
analyze
each
modification
separately.
The
confirm
individual
significantly
enhances
overall
Revista Venezolana de Gerencia,
Год журнала:
2025,
Номер
30(110), С. 1047 - 1061
Опубликована: Апрель 4, 2025
Este
estudio
tiene
como
objetivo
analizar
y
predecir
patrones
de
comportamiento
los
usuarios
transporte
interprovincial
en
Ecuador
mediante
técnicas
aprendizaje
automático.
Se
utilizó
un
conjunto
datos
proporcionado
por
la
Unión
Cooperativas
Transporte
Interprovincial
que
abarca
viajes
realizados
entre
2022
2024.
La
metodología
incluyó
implementación
K-means
para
segmentación
PCA
reducción
dimensional.
Inicialmente,
identificó
cuatro
clústeres,
pero
el
solapamiento
grupos
motivó
aplicación
PCA,
mejorando
separación.
Los
resultados
revelaron
grupos:
Ritmo
Diario,
Exploradores
Fin
Semana,
Nómadas
Eventos
Viajeros
Flexibles.
Esta
ofrece
información
clave
optimizar
servicios
mejorar
experiencia
del
usuario
al
ajustar
recursos
a
las
necesidades
cada
grupo.
