Load
forecasting
(LF)
models
lay
a
foundation
for
various
smart-grid
applications,
whose
accuracy
is
determined
by
the
input
load
data.
Prior
LF
studies
mainly
make
restrictive
assumptions
on
data,
thus
suffering
from
limited
practicality
in
two
folds:
first,
they
model
patterns
over
time,
ignoring
fact
that
real
data
are
generated
composition
of
multiple
disparate
electrical
endeavours,
leading
to
information
loss
perspective;
and
fail
considering
unexpected
events
which
lead
noises.
To
address
these
issues,
we
propose
novel
multi-type
based
random
forest
density
clustering
(MLF-RFDC),
including
three-fold
ideas:
1)
it
each
endeavour
as
an
independent
matrix;
2)
detects
corrects
noisy
entries
matrix
via
low-rank
structure;
3)
harmonizes
noise-free
matrices
all
types
ensemble
perspective.
Extensive
experiments
taken
ten
benchmark
datasets
three
real-world
datasets,
results
substantiate
superiority
our
approach
11
state-of-the-art
rival
terms
noise
detection,
restoration,
accuracy.
IET Generation Transmission & Distribution,
Journal Year:
2025,
Volume and Issue:
19(1)
Published: Jan. 1, 2025
ABSTRACT
Integrating
renewable
energy
sources
into
smart
grids
increases
supply
and
demand
management
because
are
intermittent
variable.
To
overcome
this
type
of
challenge,
short‐term
load
forecasting
(STLF)
is
essential
for
managing
energy,
demand‐side
flexibility,
the
stability
with
integration.
This
paper
presents
a
new
model
called
BiGRU‐CNN
to
improve
operation
STLF
in
grids.
The
integrates
bidirectional
gated
recurrent
units
(BiGRUs)
temporal
dependencies
convolutional
neural
networks
(CNNs)
extract
spatial
patterns
from
consumption
data.
newly
developed
BiGRU
captures
past
future
contexts
through
processing,
CNN
component
extracts
high‐level
features
enhance
accuracy
prediction.
compared
two
other
hybrid
models,
CNN‐LSTM
CNN‐GRU,
on
real‐world
data
American
electric
power
(AEP)
ISONE
datasets.
Simulation
results
show
that
proposed
outperforms
single‐step
yielding
root
mean
square
error
(RMSE)
121.43
123.57
(ISONE),
absolute
(MAE)
90.95
62.97
percentage
(MAPE)
0.61%
0.41%
(ISONE).
For
multi‐step
forecasting,
yields
RMSE
680.02
581.12
MAE
481.12
411.20
MAPE
3.27%
2.91%
can
generate
accurate
reliable
STLF,
which
useful
massive
energy‐integrated
International Journal of Electrical Power & Energy Systems,
Journal Year:
2024,
Volume and Issue:
161, P. 110166 - 110166
Published: Aug. 14, 2024
Conventional
model-driven
methods
are
hard
to
handle
large-scale
power
flow
with
multivariate
uncertainty,
variable
topology,
and
massive
real-time
repetitive
calculations.
With
the
ability
deal
non-Euclidean
graph-structured
system
data,
graph
deep
learning
shows
great
potential
in
modern
calculation.
However,
general
based
calculation
has
limited
adaptability
because
of
its
sole
mapping
node
information
black-box
attributes.
In
this
paper,
an
edge
attention
network
(EGAT-PFC)
model
is
proposed
improved
for
analysis
complex
scenarios.
First,
dual-model
structure
constructed
realize
a
complete
covering
all
systems.
Second,
learnable
coefficient
mechanism
fusing
features
ensure
global
can
be
completely
considered.
Third,
mechanisms
extended
first-order
neighborhood,
dynamic
normalization,
regularization-based
loss
function
designed
improve
training
performance.
Finally,
visualized
interpretability
developed
show
valuable
vulnerable
nodes
lines
operation.
The
numerical
simulation
verifies
that
EGAT-PFC
high
accuracy,
fast
mapping,
as
well
excellent
topologies.
