Biomimetics,
Год журнала:
2023,
Номер
8(4), С. 377 - 377
Опубликована: Авг. 18, 2023
With
the
rapid
development
of
geometric
modeling
industry
and
computer
technology,
design
shape
optimization
complex
curve
shapes
have
now
become
a
very
important
research
topic
in
CAGD.
In
this
paper,
Hybrid
Artificial
Hummingbird
Algorithm
(HAHA)
is
used
to
optimize
composite
shape-adjustable
generalized
cubic
Ball
(CSGC–Ball,
for
short)
curves.
Firstly,
algorithm
(AHA),
as
newly
proposed
meta-heuristic
algorithm,
has
advantages
simple
structure
easy
implementation
can
quickly
find
global
optimal
solution.
However,
there
are
still
limitations,
such
low
convergence
accuracy
tendency
fall
into
local
optimization.
Therefore,
paper
proposes
HAHA
based
on
original
AHA,
combined
with
elite
opposition-based
learning
strategy,
PSO,
Cauchy
mutation,
increase
population
diversity
avoid
falling
optimization,
thus
improve
rate
AHA.
Twenty-five
benchmark
test
functions
CEC
2022
suite
evaluate
overall
performance
HAHA,
experimental
results
statistically
analyzed
using
Friedman
Wilkerson
rank
sum
tests.
The
show
that,
compared
other
advanced
algorithms,
good
competitiveness
practicality.
Secondly,
order
better
realize
curves
engineering,
CSGC–Ball
parameters
constructed
SGC–Ball
basis
functions.
By
changing
parameters,
whole
or
be
adjusted
more
flexibly.
Finally,
make
ideal
shape,
curve-shape
model
established
minimum
energy
value,
solve
model.
Two
representative
numerical
examples
comprehensively
verify
effectiveness
superiority
solving
problems.
International Journal of Computational Intelligence Systems,
Год журнала:
2024,
Номер
17(1)
Опубликована: Янв. 3, 2024
Abstract
According
to
Moore’s
law,
computer
processing
hardware
technology
performance
is
doubled
every
year.
To
make
effective
use
of
this
technological
development,
the
algorithmic
solutions
have
be
developed
at
same
speed.
Consequently,
it
necessary
design
parallel
algorithms
implemented
on
machines.
This
helps
exploit
multi-core
environment
by
executing
multiple
instructions
simultaneously
processors.
Traveling
Salesman
(TSP)
a
challenging
non-deterministic-hard
optimization
problem
that
has
exponential
running
time
using
brute-force
methods.
TSP
concerned
with
finding
shortest
path
starting
point
and
returning
after
visiting
list
points,
provided
these
points
are
visited
only
once.
Meta-heuristic
been
used
tackle
find
near-optimal
in
reasonable
time.
paper
proposes
River
Formation
Dynamics
Optimization
Algorithm
(RFD)
solve
problem.
The
parallelization
technique
depends
dividing
population
into
different
processors
Map-Reduce
framework
Apache
Spark.
experiments
accomplished
three
phases.
first
phase
compares
speedup,
time,
efficiency
RFD
1
(sequential
RFD),
4,
8,
16
cores.
second
proposed
water-based
algorithms,
namely
Water
Flow
algorithm,
Intelligent
Drops,
Cycle
Algorithm.
achieve
fairness,
all
system
specifications
values
for
shared
parameters.
third
reported
results
metaheuristic
were
literature.
demonstrate
algorithm
best
majority
instances,
achieving
lowest
times
across
core
counts.
Our
findings
highlight
importance
selecting
most
suitable
count
based
characteristics
optimal
optimization.
Journal of Computational Design and Engineering,
Год журнала:
2024,
Номер
11(2), С. 37 - 69
Опубликована: Янв. 17, 2024
Abstract
In
this
study,
an
improved
version
of
aquila
optimizer
(AO)
known
as
EHAOMPA
has
been
developed
by
using
the
marine
predators
algorithm
(MPA).
MPA
is
a
recent
and
well-behaved
with
unique
memory
saving
fish
aggregating
devices
mechanism.
At
same
time,
it
suffers
from
various
defects
such
inadequate
global
search,
sluggish
convergence,
stagnation
local
optima.
However,
AO
contented
robust
exploration
capability,
fast
convergence
speed,
high
search
efficiency.
Thus,
proposed
aims
to
complement
shortcomings
while
bringing
new
features.
Specifically,
representative-based
hunting
technique
incorporated
into
stage
enhance
population
diversity.
random
opposition-based
learning
introduced
exploitation
prevent
sticking
This
study
tests
performance
EHAOMPA’s
on
23
standard
mathematical
benchmark
functions,
29
complex
test
functions
CEC2017
suite,
six
constrained
industrial
engineering
design
problems,
convolutional
neural
network
hyperparameter
(CNN-hyperparameter)
optimization
for
Corona
Virus
Disease
19
(COVID-19)
computed
tomography-image
detection
problem.
compared
four
existing
types,
achieving
best
both
numerical
practical
issues.
Compared
other
methods,
function
results
demonstrate
that
exhibits
more
potent
higher
rate,
increased
accuracy,
ability
avoid
The
excellent
experimental
in
problems
indicate
great
potential
solving
real-world
problems.
combination
multiple
strategies
can
effectively
improve
algorithm.
source
code
publicly
available
at
https://github.com/WangShuang92/EHAOMPA.
Cluster Computing,
Год журнала:
2024,
Номер
27(6), С. 7775 - 7802
Опубликована: Апрель 1, 2024
Abstract
An
efficient
variant
of
the
recent
sea
horse
optimizer
(SHO)
called
SHO-OBL
is
presented,
which
incorporates
opposition-based
learning
(OBL)
approach
into
predation
behavior
SHO
and
uses
greedy
selection
(GS)
technique
at
end
each
optimization
cycle.
This
enhancement
was
created
to
avoid
being
trapped
by
local
optima
improve
quality
variety
solutions
obtained.
However,
can
occasionally
be
vulnerable
stagnation
in
optima,
a
problem
concern
given
low
diversity
horses.
In
this
paper,
an
suggested
for
tackling
genuine
global
systems.
To
investigate
validity
SHO-OBL,
it
compared
with
nine
robust
optimizers,
including
differential
evolution
(DE),
grey
wolf
(GWO),
moth-flame
algorithm
(MFO),
sine
cosine
(SCA),
fitness
dependent
(FDO),
Harris
hawks
(HHO),
chimp
(ChOA),
Fox
(FOX),
basic
ten
unconstrained
test
routines
belonging
IEEE
congress
on
evolutionary
computation
2020
(CEC’20).
Furthermore,
three
different
design
engineering
issues,
welded
beam,
tension/compression
spring,
pressure
vessel,
are
solved
using
proposed
its
applicability.
addition,
one
most
successful
approaches
data
transmission
wireless
sensor
network
that
little
energy
clustering.
assist
process
choosing
optimal
power-aware
cluster
heads
based
predefined
objective
function
takes
account
residual
power
node,
as
well
sum
powers
surrounding
nodes.
Similarly,
performance
competitors.
Thorough
simulations
demonstrate
outperforms
terms
power,
lifespan,
extended
stability
duration.
Biomimetics,
Год журнала:
2023,
Номер
8(4), С. 377 - 377
Опубликована: Авг. 18, 2023
With
the
rapid
development
of
geometric
modeling
industry
and
computer
technology,
design
shape
optimization
complex
curve
shapes
have
now
become
a
very
important
research
topic
in
CAGD.
In
this
paper,
Hybrid
Artificial
Hummingbird
Algorithm
(HAHA)
is
used
to
optimize
composite
shape-adjustable
generalized
cubic
Ball
(CSGC–Ball,
for
short)
curves.
Firstly,
algorithm
(AHA),
as
newly
proposed
meta-heuristic
algorithm,
has
advantages
simple
structure
easy
implementation
can
quickly
find
global
optimal
solution.
However,
there
are
still
limitations,
such
low
convergence
accuracy
tendency
fall
into
local
optimization.
Therefore,
paper
proposes
HAHA
based
on
original
AHA,
combined
with
elite
opposition-based
learning
strategy,
PSO,
Cauchy
mutation,
increase
population
diversity
avoid
falling
optimization,
thus
improve
rate
AHA.
Twenty-five
benchmark
test
functions
CEC
2022
suite
evaluate
overall
performance
HAHA,
experimental
results
statistically
analyzed
using
Friedman
Wilkerson
rank
sum
tests.
The
show
that,
compared
other
advanced
algorithms,
good
competitiveness
practicality.
Secondly,
order
better
realize
curves
engineering,
CSGC–Ball
parameters
constructed
SGC–Ball
basis
functions.
By
changing
parameters,
whole
or
be
adjusted
more
flexibly.
Finally,
make
ideal
shape,
curve-shape
model
established
minimum
energy
value,
solve
model.
Two
representative
numerical
examples
comprehensively
verify
effectiveness
superiority
solving
problems.
