Financial Innovation,
Journal Year:
2024,
Volume and Issue:
10(1)
Published: July 16, 2024
Abstract
Commodity
markets,
such
as
crude
oil
and
precious
metals,
play
a
strategic
role
in
the
economic
development
of
nations,
with
prices
influencing
geopolitical
relations
global
economy.
Moreover,
gold
silver
are
argued
to
hedge
stock
cryptocurrency
markets
during
market
downsides.
Therefore,
accurate
forecasting
metals
is
critical.
Nevertheless,
due
nonlinear
nature,
substantial
fluctuations,
irregular
cycles
predicting
their
challenging
task.
Our
study
contributes
commodity
price
literature
by
implementing
comparing
advanced
deep-learning
models.
We
address
this
gap
including
alongside
our
analysis,
offering
more
comprehensive
understanding
metal
markets.
This
research
expands
existing
knowledge
provides
valuable
insights
into
prices.
In
study,
we
implemented
16
deep-
machine-learning
models
forecast
daily
West
Texas
Intermediate
(WTI),
Brent,
gold,
The
employed
long
short-term
memory
(LSTM),
BiLSTM,
gated
recurrent
unit
(GRU),
bidirectional
units
(BiGRU),
T2V-BiLSTM,
T2V-BiGRU,
convolutional
neural
networks
(CNN),
CNN-BiLSTM,
CNN-BiGRU,
temporal
network
(TCN),
TCN-BiLSTM,
TCN-BiGRU.
compared
performance
baseline
random
forest,
LightGBM,
support
vector
regression,
k-nearest
neighborhood
using
mean
absolute
error
(MAE),
percentage
error,
root
squared
evaluation
criteria.
By
considering
different
sliding
window
lengths,
examine
results
reveal
that
TCN
model
outperforms
others
for
WTI,
silver,
achieving
lowest
MAE
values
1.444,
1.295,
0.346,
respectively.
BiGRU
performs
best
an
15.188
30-day
input
sequence.
Furthermore,
LightGBM
exhibits
comparable
best-performing
overall.
These
findings
critical
investors,
policymakers,
mining
companies,
governmental
agencies
effectively
anticipate
trends,
mitigate
risk,
manage
uncertainty,
make
timely
decisions
strategies
regarding
oil,
Applied Sciences,
Journal Year:
2021,
Volume and Issue:
11(23), P. 11263 - 11263
Published: Nov. 27, 2021
The
Electric
Energy
Consumption
Prediction
(EECP)
is
a
complex
and
important
process
in
an
intelligent
energy
management
system
its
importance
has
been
increasing
rapidly
due
to
technological
developments
human
population
growth.
A
reliable
accurate
model
for
EECP
considered
key
factor
appropriate
policy.
In
recent
periods,
many
artificial
intelligence-based
models
have
developed
perform
different
simulation
functions,
engineering
techniques,
optimal
forecasting
order
predict
future
demands
on
the
basis
of
historical
data.
this
article,
new
metaheuristic
based
Long
Short-Term
Memory
(LSTM)
network
proposed
effective
EECP.
After
collecting
data
sequences
from
Individual
Household
Power
(IHEPC)
dataset
Appliances
Load
(AEP)
dataset,
refinement
accomplished
using
min-max
standard
transformation
methods.
Then,
LSTM
with
Butterfly
Optimization
Algorithm
(BOA)
BOA
used
select
hyperparametric
values
which
precisely
describe
EEC
patterns
discover
time
series
dynamics
domain.
This
extensive
experiment
conducted
IHEPC
AEP
datasets
shows
that
obtains
minimum
error
rate
relative
existing
models.
IEEE Access,
Journal Year:
2022,
Volume and Issue:
10, P. 14184 - 14194
Published: Jan. 1, 2022
Wind
turbines
are
one
of
the
primary
sources
renewable
energy,
which
leads
to
a
sustainable
and
efficient
energy
solution.
It
does
not
release
any
carbon
emissions
pollute
our
planet.
The
wind
farms
monitoring
power
generation
prediction
is
complex
problem
due
unpredictability
speed.
Consequently,
it
limits
decision
management
team
plan
consumption
in
an
effective
way.
Our
proposed
model
solves
this
challenge
by
utilizing
5G-Next
Generation-Radio
Access
Network
(5G-NG-RAN)
assisted
cloud-based
digital
twins’
framework
virtually
monitor
form
predictive
forecast
speed
predict
generated
power.
developed
based
on
Microsoft
Azure
twins
infrastructure
as
5-dimensional
platform.
modeling
deep
learning
approach,
temporal
convolution
network
(TCN)
followed
non-parametric
k-nearest
neighbor
(kNN)
regression.
Predictive
has
two
components.
First,
processes
univariate
time
series
data
its
Secondly,
estimates
for
each
quarter
year
ranges
from
week
whole
month
(i.e.,
medium-term
prediction)
To
evaluate
experiments
performed
onshore
publicly
available
datasets.
obtained
results
confirm
applicability
framework.
Furthermore,
comparative
analysis
with
existing
classical
models
shows
that
designed
approach
better
results.
can
assist
remotely
well
estimate
advance.
IEEE Transactions on Smart Grid,
Journal Year:
2023,
Volume and Issue:
14(5), P. 4073 - 4085
Published: Jan. 16, 2023
This
paper
presents
a
Temporal
Convolutional
Network
(TCN)
based
hybrid
PV
forecasting
framework
for
enhancing
hours-ahead
utility-scale
forecasting.
The
consists
of
two
models:
physics-based
trend
(TF)
model
and
data-driven
fluctuation
(FF)
model.
Three
TCNs
are
integrated
in
the
for:
i)
blending
inputs
from
different
Numerical
Weather
Prediction
sources
TF
to
achieve
superior
performance
on
hourly
profiles,
ii)
capturing
spatial-temporal
correlations
between
detector
sites
target
site
FF
more
accurate
forecast
intra-hour
power
drops,
iii)
reconciling
results
obtain
coherent
with
both
trends
fluctuations
well
preserved.
To
automatically
identify
most
contributive
neighboring
forming
network,
scenario-based
correlation
analysis
method
is
developed,
which
significantly
improves
capability
large
caused
by
cloud
movements.
tested,
validated
using
actual
data
collected
95
farms
North
Carolina.
Simulation
show
that
6
hours
ahead
improved
20%
-
30%
compared
state-of-the-art
methods.
International Journal of Neural Systems,
Journal Year:
2020,
Volume and Issue:
31(03), P. 2130001 - 2130001
Published: Nov. 24, 2020
In
recent
years,
deep
learning
techniques
have
outperformed
traditional
models
in
many
machine
tasks.
Deep
neural
networks
successfully
been
applied
to
address
time
series
forecasting
problems,
which
is
a
very
important
topic
data
mining.
They
proved
be
an
effective
solution
given
their
capacity
automatically
learn
the
temporal
dependencies
present
series.
However,
selecting
most
convenient
type
of
network
and
its
parametrization
complex
task
that
requires
considerable
expertise.
Therefore,
there
need
for
deeper
studies
on
suitability
all
existing
architectures
different
this
work,
we
face
two
main
challenges:
comprehensive
review
latest
works
using
experimental
study
comparing
performance
popular
architectures.
The
comparison
involves
thorough
analysis
seven
types
terms
accuracy
efficiency.
We
evaluate
rankings
distribution
results
obtained
with
proposed
under
architecture
configurations
training
hyperparameters.
datasets
used
comprise
more
than
50,000
divided
into
12
problems.
By
38,000
these
data,
provide
extensive
forecasting.
Among
studied
models,
show
long
short-term
memory
(LSTM)
convolutional
(CNN)
are
best
alternatives,
LSTMs
obtaining
accurate
forecasts.
CNNs
achieve
comparable
less
variability
parameter
configurations,
while
also
being
efficient.
Energies,
Journal Year:
2021,
Volume and Issue:
14(21), P. 6958 - 6958
Published: Oct. 22, 2021
The
accuracy
of
a
predictive
system
is
critical
for
maintenance
and
to
support
the
right
decisions
at
times.
Statistical
models,
such
as
ARIMA
SARIMA,
are
unable
describe
stochastic
nature
data.
Neural
networks,
long
short-term
memory
(LSTM)
gated
recurrent
unit
(GRU),
good
predictors
univariate
multivariate
present
paper
describes
case
study
where
performances
units
compared,
based
on
different
hyperparameters.
In
general,
exhibit
better
performance,
pulp
presses.
final
result
demonstrates
that,
maximize
equipment
availability,
units,
demonstrated
in
paper,
best
options.