Research Square (Research Square),
Journal Year:
2023,
Volume and Issue:
unknown
Published: Dec. 12, 2023
Abstract
This
paper
addresses
the
robot
machining
requirements
for
large
aerospace
structural
components
and
provides
a
method
rapid
workpiece
positioning
in
systems
that
combines
ease
of
visual
measurement-based
with
precision
contact-based
positioning.
In
order
to
enhance
calibration
system,
this
introduces
utilizes
ruby
probe
as
tool
perform
sphere-to-sphere
contact
Tool
Center
Point
(TCP).
A
model
is
established,
converting
problem
into
non-linear
least
squares
optimization
problem.
To
address
challenges
multi-dimensional
non-convex
continuous
optimization,
designs
combined
LM-D
algorithm
incorporates
Levenberg-Marquardt
(L-M)
DIRECT
algorithm,
engaging
mutual
iterative
processes
obtain
global
optimum.
approach
ensuring
efficiency
while
maximizing
potential
optimum
solution.
an
convergence
termination
criterion
TCP
which
used
determine
whether
converges
globally.
also
contributes
improving
algorithm's
efficiency.
Experimental
tests
were
conducted
on
typical
industrial
robots,
results
illustrate
superior
performance
terms
both
high
iteration
compared
traditional
methods.
research
offers
promising
efficient
solution
industrial.
Results in Engineering,
Journal Year:
2023,
Volume and Issue:
21, P. 101727 - 101727
Published: Dec. 30, 2023
The
use
of
solar
photovoltaic
panels
as
source
power
for
Brushless
Direct
Current
(BLDC)
motors
requires
a
DC-DC
Converter
circuit.
One
application
energy
is
motors.
main
problem
the
voltage
fluctuation
and
low
DC
generated
by
panel.
This
research
aims
to
improve
performance
Boost
circuit
minimize
fluctuations.
methodology
encompasses
mathematical
modeling
in
form
transfer
functions
optimizing
using
Proportional
Integral
Derivative
(PID)
controller
Firefly
algorithm.
Simulation
testing
results
indicate
an
improvement
transient
response
driver
BLDC
motor.
evidenced
increase
rise
time
from
499
s
820
s,
decrease
settling
3.33
e+03
2.07e+03
reduction
overshoot
0
%
previously
11.4
%.
utilization
firefly
algorithm
optimization
significantly
enhances
system
efficiency,
demonstrated
faster
achievement
stability
without
excessive
oscillation
required
settle.
Overall,
this
study
shows
that
effective
developing
circuits,
improving
efficiency
reducing
eliminating
overshoot.
These
findings
provide
empirical
evidence
effectiveness
artificial
intelligence
algorithms
enhancing
operational
conversion
systems.
Applied Sciences,
Journal Year:
2025,
Volume and Issue:
15(3), P. 1033 - 1033
Published: Jan. 21, 2025
This
paper
addresses
the
machining
requirements
for
large
aerospace
structural
components
using
robotic
systems
and
proposes
a
method
rapid
workpiece
positioning
that
combines
simplicity
of
vision-based
with
precision
contact-based
methods.
To
enhance
accuracy
robot
calibration,
novel
approach
utilizing
ruby
probe
sphere-to-sphere
contact
calibration
Tool
Center
Point
(TCP)
is
introduced.
A
model
formulated,
transforming
process
into
nonlinear
least
squares
(NLS)
optimization
problem.
tackle
challenges
NLS
optimization,
hybrid
LM-D
algorithm
developed,
integrating
Levenberg–Marquardt
(L-M)
DIviding
RECTangle
(DIRECT)
algorithms
in
an
iterative
to
achieve
global
optimum.
ensures
computational
efficiency
while
maximizing
likelihood
finding
globally
optimal
solution.
An
convergence
termination
criterion
TCP
established
determine
convergence,
further
enhancing
algorithm’s
efficiency.
Experimental
validation
was
performed
on
industrial
robots,
demonstrating
proposed
superior
performance
iteration
compared
traditional
research
provides
effective
practical
solution
applications.
IEEE Access,
Journal Year:
2024,
Volume and Issue:
12, P. 103271 - 103298
Published: Jan. 1, 2024
A
recently
created
optimization
algorithm
named
the
Dynamic
Hunting
Leadership
(DHL)
was
inspired
by
leadership
tactics
used
in
hunting
operations.
The
foundation
of
DHL
is
idea
that
successful
can
significantly
increase
endeavors.
Although
has
shown
to
be
relatively
simple
and
tackling
a
variety
practical
issues,
it
been
discovered
suffers
with
efficiently
balancing
global
exploration
local
search
phase,
particularly
high-dimensional
numerical
problems
engineering
applications.
Furthermore,
due
drawbacks,
vulnerable
becoming
stuck
optimal.
present
study
aims
tackle
aforementioned
challenges
introducing
modified
variant
DHL,
referred
as
mDHL,
utilizes
Levy
Flight
technique
localized
development
strategy
augment
each
hunter's
capacity
track
their
prey
attain
superior
optimal
outcomes.
Moreover,
escape
operator
quasi-opposition
learning
are
synergistically
incorporated
foster
hunters'
techniques.
These
knowledge
sharing
between
leaders
hunters,
resulting
harmonious
blend
capabilities.
mDHL
outperform
existing
optimizers
across
20
function
test
suites
varying
dimensions
from
30
200
CEC
2019
functions.
In
addition,
successfully
applied
solve
four
design
cases,
demonstrating
its
practicality.
experimental
findings
indicate
substantial
improvement
over
conventional
emphasizing
potential
competitive
efficient
for
addressing
challenges.
Scientific Reports,
Journal Year:
2024,
Volume and Issue:
14(1)
Published: Oct. 19, 2024
The
Whale
Optimization
Algorithm
(WOA)
is
regarded
as
a
classic
metaheuristic
algorithm,
yet
it
suffers
from
limited
population
diversity,
imbalance
between
exploitation
and
exploration,
low
solution
accuracy.
In
this
paper,
we
propose
the
Spiral-Enhanced
(SEWOA),
which
incorporates
nonlinear
time-varying
self-adaptive
perturbation
strategy
an
Archimedean
spiral
structure
into
original
WOA.
enhances
diversity
of
space,
aiding
algorithm
in
escaping
local
optima.
optimization
dynamic
improves
algorithm's
search
capability
effectiveness
proposed
validated
multiple
perspectives
using
CEC2014
test
functions,
CEC2017
23
benchmark
functions.
experimental
results
demonstrate
that
enhanced
significantly
balances
global
search,
Additionally,
SEWOA
exhibits
excellent
performance
solving
three
engineering
design
problems,
showcasing
its
value
wide
range
potential
applications.
Symmetry,
Journal Year:
2024,
Volume and Issue:
16(3), P. 324 - 324
Published: March 7, 2024
The
pathfinder
algorithm
(PFA)
starts
with
a
random
search
for
the
initial
population,
which
is
then
partitioned
into
only
phase
and
follower
phase.
This
approach
often
results
in
issues
like
poor
solution
accuracy,
slow
convergence,
susceptibility
to
local
optima
PFA.
To
address
these
challenges,
multi-strategy
fusion
proposed
symmetry-enhanced,
improved
algorithm-based
engineering
optimization
problems
(IPFA)
function
problems.
First,
elite
opposition-based
learning
mechanism
incorporated
improve
population
diversity
quality,
enhance
accuracy
of
algorithm;
second,
convergence
speed
algorithm,
escape
energy
factor
embedded
prey-hunting
GWO
replaces
PFA,
increases
improves
efficiency
lastly,
solve
problem
easily
falling
optimum,
optimal
individual
position
perturbed
using
dimension-by-dimension
mutation
method
t-distribution,
helps
jump
out
optimum
rapidly
advance
toward
other
regions.
IPFA
used
testing
on
16
classical
benchmark
test
functions
29
complex
CEC2017
sets.
final
PFA
pressure
vessels
are
5984.8222
5948.3597,
respectively.
tension
springs
0.012719
0.012699,
respectively,
comparable
original
algorithms.
A
comparison
between
algorithms
shows
that
significantly
enhanced
terms
lower
cost
further
verifies
robustness
algorithm.
Gazi Journal of Engineering Sciences,
Journal Year:
2024,
Volume and Issue:
10(2), P. 405 - 418
Published: Aug. 31, 2024
Early
diagnosis
and
prediction
of
psoriasis
is
crucial
to
control
disease
progression,
alleviate
symptoms
reduce
the
risk
complications.Diagnosis
in
early
stages
helps
determine
appropriate
treatment
plan
improve
patient's
quality
life.The
aim
this
study
enable
psoriasis.For
purpose,
a
hybrid
architecture
was
created
using
stacked
auto-encoder,
softmax
classifier
Firefly
Optimization
Algorithm.With
architecture,
architectural
parameters
autocoder
structure
aimed
be
for
all
hyperparameters
within
were
optimized.The
model
implemented
on
"Dermatology"
dataset
UCI
data
warehouse.In
addition,
machine
learning
methods
such
as
K-Nearest
Neighbor
algorithm,
Support
Vector
Machine
Decision
Trees,
which
are
frequently
used
literature,
also
applied
same
dataset.The
findings
obtained
from
experimental
studies
presented
controversial
manner.The
show
that
proposed
achieves
better
results
than
other
methods.At
time,
optimized
with
can
an
alternative
method
patient
decision
support
systems.
