Energies,
Год журнала:
2024,
Номер
17(17), С. 4379 - 4379
Опубликована: Сен. 1, 2024
The
accurate
prediction
of
carbon
dioxide
(CO2)
emissions
in
the
building
industry
can
provide
data
support
and
theoretical
insights
for
sustainable
development.
This
study
proposes
a
hybrid
model
predicting
CO2
that
combines
multi-strategy
improved
particle
swarm
optimization
(MSPSO)
algorithm
with
long
short-term
memory
(LSTM)
model.
Firstly,
(PSO)
is
enhanced
by
combining
tent
chaotic
mapping,
mutation
least-fit
particles,
random
perturbation
strategy.
Subsequently,
performance
MSPSO
evaluated
using
set
23
internationally
recognized
test
functions.
Finally,
predictive
MSPSO-LSTM
assessed
from
Yangtze
River
Delta
region
as
case
study.
results
indicate
coefficient
determination
(R2)
reaches
0.9677,
which
more
than
10%
higher
BP,
LSTM,
CNN
non-hybrid
models
demonstrates
significant
advantages
over
PSO-LSTM,
GWO-LSTM,
WOA-LSTM
models.
Additionally,
mean
square
error
(MSE)
2445.6866
Mt,
absolute
(MAE)
4.1010
both
significantly
lower
those
Overall,
high
accuracy
industry,
offering
robust
development
industry.
This
paper
addresses
the
challenges
in
forecasting
electrical
energy
current
era
of
renewable
integration.
It
reviews
advanced
adaptive
methodologies
while
also
analyzing
evolution
research
this
field
through
bibliometric
analysis.
The
review
highlights
key
contributions
and
limitations
models
with
an
emphasis
on
traditional
methods.
analysis
reveals
that
Long
Short-Term
Memory
(LSTM)
networks,
optimization
techniques,
deep
learning
have
potential
to
model
dynamic
nature
consumption,
but
they
higher
computational
demands
data
requirements.
aims
offer
a
balanced
view
advancements
methods,
guiding
researchers,
policymakers,
industry
experts.
advocates
for
collaborative
innovation
enhance
accuracy
support
development
resilient,
sustainable
systems.
Sustainability,
Год журнала:
2023,
Номер
15(18), С. 13934 - 13934
Опубликована: Сен. 20, 2023
Global
warming
is
a
major
environmental
issue
facing
humanity,
and
the
resulting
climate
change
has
severely
affected
environment
daily
lives
of
people.
China
attaches
great
importance
to
actively
responds
issues.
In
order
achieve
“dual
carbon”
goal,
it
necessary
clearly
define
emission
reduction
path
scientifically
predict
future
carbon
emissions,
which
basis
for
setting
targets.
To
ensure
accuracy
data,
this
study
applies
coefficient
method
calculate
emissions
from
energy
consumption
in
30
provinces,
regions,
cities
1997
2021.
Considering
spatial
correlation
between
different
regions
China,
we
propose
new
machine
learning
prediction
model
that
incorporates
weighting,
namely,
an
LSTM-CNN
combination
with
weighting.
The
weighting
explains
combined
used
analyze
2022
2035
under
scenarios.
results
show
four
convolutional
layers
performs
best.
Compared
other
models,
best
predictive
performance,
MAE
8.0169,
RMSE
11.1505,
R2
0.9661
on
test
set.
Based
scenario
predictions,
found
most
can
peaking
before
2030.
Some
need
adjust
their
development
rates
based
specific
circumstances
as
early
possible.
This
provides
research
direction
deep
time
series
forecasting
proposes
forecasting.
