PeerJ Computer Science,
Год журнала:
2025,
Номер
11, С. e2722 - e2722
Опубликована: Фев. 28, 2025
The
Atom
Search
Optimization
(ASO)
algorithm
is
a
recent
advancement
in
metaheuristic
optimization
inspired
by
principles
of
molecular
dynamics.
It
mathematically
models
and
simulates
the
natural
behavior
atoms,
with
interactions
governed
forces
derived
from
Lennard-Jones
potential
constraint
based
on
bond-length
potentials.
Since
its
inception
2019,
it
has
been
successfully
applied
to
various
challenges
across
diverse
fields
technology
science.
Despite
notable
achievements
rapidly
growing
body
literature
ASO
domain,
comprehensive
study
evaluating
success
implementations
still
lacking.
To
address
this
gap,
article
provides
thorough
review
half
decade
advancements
research,
synthesizing
wide
range
studies
highlight
key
variants,
their
foundational
principles,
significant
achievements.
examines
applications,
including
single-
multi-objective
problems,
introduces
well-structured
taxonomy
guide
future
exploration
ASO-related
research.
reviewed
reveals
that
several
variants
algorithm,
modifications,
hybridizations,
implementations,
have
developed
tackle
complex
problems.
Moreover,
effectively
domains,
such
as
engineering,
healthcare
medical
Internet
Things
communication,
clustering
data
mining,
environmental
modeling,
security,
engineering
emerging
most
prevalent
application
area.
By
addressing
common
researchers
face
selecting
appropriate
algorithms
for
real-world
valuable
insights
into
practical
applications
offers
guidance
designing
tailored
specific
Energy,
Год журнала:
2023,
Номер
275, С. 127430 - 127430
Опубликована: Апрель 8, 2023
Predicting
electricity
demand
data
is
considered
an
essential
task
in
decisions
taking,
and
establishing
new
infrastructure
the
power
generation
network.
To
deliver
a
high-quality
prediction,
this
paper
proposes
hybrid
combination
technique,
based
on
deep
learning
model
of
Convolutional
Neural
Networks
Echo
State
Networks,
named
as
CESN.
Daily
from
four
sites
(Roderick,
Rocklea,
Hemmant
Carpendale),
located
Southeast
Queensland,
Australia,
have
been
used
to
develop
proposed
prediction
model.
The
study
also
analyzes
five
other
machine
learning-based
models
(support
vector
regression,
multilayer
perceptron,
extreme
gradient
boosting,
neural
network,
Light
Gradient
Boosting)
compare
evaluate
outcomes
approach.
results
obtained
experimental
showed
that
able
obtain
highest
performance
compared
existing
developed
for
daily
forecasting.
Based
statistical
approaches
utilized
study,
approach
presents
accuracy
among
models.
algorithm
excellent
accurate
forecasting
method,
which
outperformed
state
art
algorithms
are
currently
problem.
Measurement,
Год журнала:
2022,
Номер
202, С. 111759 - 111759
Опубликована: Авг. 19, 2022
Global
solar
radiation
(GSR)
prediction
plays
an
essential
role
in
planning,
controlling
and
monitoring
power
systems.
However,
its
stochastic
behaviour
is
a
significant
challenge
achieving
satisfactory
results.
This
study
aims
to
design
innovative
hybrid
model
that
integrates
feature
selection
mechanism
using
Slime-Mould
algorithm,
Convolutional-Neural-Network
(CNN),
Long–Short-Term-Memory
Neural
Network
(LSTM)
final
CNN
with
Multilayer-Perceptron
output
(SCLC
algorithm
hereafter).
The
proposed
was
applied
six
farms
Queensland
(Australia)
at
daily
temporal
horizons
different
time
steps.
comprehensive
benchmarking
of
the
obtained
results
those
from
two
Deep-Learning
(CNN-LSTM,
Deep-Neural-Network)
three
Machine-Learning
(Artificial-Neural-Network,
Random-Forest,
Self-Adaptive
Differential-Evolutionary
Extreme-Learning-Machines)
models
highlighted
higher
performance
all
selected
farms.
From
obtained,
this
work
establishes
designed
SCLC
could
have
practical
utility
for
applications
renewable
sustainable
energy
resource
management.
Advances in environmental engineering and green technologies book series,
Год журнала:
2023,
Номер
unknown, С. 1 - 18
Опубликована: Дек. 30, 2023
Earth
observations
have
become
a
developing
trend
over
the
last
decade
because
of
their
ability
to
enable
real-time
tracking
and
forecasting
various
environmental
phenomena,
including
landslides,
drought,
floods,
wildfires.
However,
conventional
approaches
in
observation
relied
on
guide
processing
or
human
interpretation
statistics.
Via
mixing
AI,
statement's
achievement
has
progressed
significantly.
AI
offers
automatic
timely
analysis
significant
volumes
faraway
sensing
satellite
TV
for
computer
facts,
considering
natural
events
approaches.
The
software
Remark
enabled
several
advantages,
improved
accuracy
mapping
classification
gadgets,
detection
hobby
which
include
homes,
roads,
forests,
changes
land
use
cover.
