IEEE Access,
Journal Year:
2024,
Volume and Issue:
12, P. 31589 - 31604
Published: Jan. 1, 2024
Confronted
with
the
challenges
posed
by
climate
change
and
ongoing
energy
transition,
solar
is
one
of
important
new
sources,
tower
power
plant
has
become
an
innovative
solution
to
promote
clean
development.
The
optimization
heliostat
field
layout
constitutes
a
crucial
aspect
in
enhancing
operational
efficiency
concentrated
plant.
Currently,
garnered
widespread
attention.
In
this
paper,
we
propose
swarm
algorithm
niching
elite
competition
called
NECSO
solve
large-scale
optimization.
First,
aiming
increase
diversity
heterogeneity
within
population,
employ
random
grouping
strategy
partition
population
into
distinct
sub-swarms.
Then,
design
mechanism
harmonize
performance
exploration.
carried
out
any
sub-swarm
enhance
explorability
particles.
occurs
between
elites
which
select
from
each
improve
convergence
Additionally,
develop
mathematical
model
for
layout.
This
employs
currently
advanced
computational
methods,
facilitating
prompt
precise
calculation
optical
To
evaluate
NECSO,
15
practical
cases
varying
scales.
And
then,
conduct
comparative
experiments
eight
mainstream
excellent
algorithms.
results
indicate
that
exhibits
competitive
solving
optimization,
particularly
cases.
Biomimetics,
Journal Year:
2023,
Volume and Issue:
8(3), P. 268 - 268
Published: June 21, 2023
Recently,
swarm
intelligence
algorithms
have
received
much
attention
because
of
their
flexibility
for
solving
complex
problems
in
the
real
world.
a
new
algorithm
called
colony
predation
(CPA)
has
been
proposed,
taking
inspiration
from
predatory
habits
groups
nature.
However,
CPA
suffers
poor
exploratory
ability
and
cannot
always
escape
solutions
known
as
local
optima.
Therefore,
to
improve
global
search
capability
CPA,
an
improved
variant
(OLCPA)
incorporating
orthogonal
learning
strategy
is
proposed
this
paper.
Then,
considering
fact
that
can
go
beyond
optimum
find
solution,
novel
OLCPA-CNN
model
which
uses
OLCPA
tune
parameters
convolutional
neural
network.
To
verify
performance
OLCPA,
comparison
experiments
are
designed
compare
with
other
traditional
metaheuristics
advanced
on
IEEE
CEC
2017
benchmark
functions.
The
experimental
results
show
ranks
first
compared
algorithms.
Additionally,
achieves
high
accuracy
rates
97.7%
97.8%
classifying
MIT-BIH
Arrhythmia
European
ST-T
datasets.
