Mathematics,
Journal Year:
2023,
Volume and Issue:
11(5), P. 1068 - 1068
Published: Feb. 21, 2023
Recent
years
have
seen
an
increasing
interest
in
developing
robust,
accurate
and
possibly
fast
forecasting
methods
for
both
energy
production
consumption.
Traditional
approaches
based
on
linear
architectures
are
not
able
to
fully
model
the
relationships
between
variables,
particularly
when
dealing
with
many
features.
We
propose
a
Gradient-Boosting–Machine-based
framework
forecast
demand
of
mixed
customers
dispatching
company,
aggregated
according
their
location
within
seven
Italian
electricity
market
zones.
The
main
challenge
is
provide
precise
one-day-ahead
predictions,
despite
most
recent
data
being
two
months
old.
This
requires
exogenous
regressors,
e.g.,
as
historical
features
part
air
temperature,
be
incorporated
scheme
tailored
specific
case.
Numerical
simulations
conducted,
resulting
MAPE
5–15%
zone.
Gradient
Boosting
performs
significantly
better
compared
classical
statistical
models
time
series,
such
ARMA,
unable
capture
holidays.
Energy Reports,
Journal Year:
2022,
Volume and Issue:
8, P. 13189 - 13198
Published: Oct. 17, 2022
In
this
paper,
a
comprehensive
review
is
presented
for
mid-term
load
forecasting.
The
basic
loads
and
effective
factors
are
studied,
then
several
classifications
forecasting
approaches.
main
advantages
drawbacks
of
the
approaches
analyzed.
neuro-fuzzy-based
investigated
in
more
detail,
their
limitations
studied.
Finally,
some
aspects
use
neuro-fuzzy
systems
contributions
that:
(1)
A
such
that
both
classical
methods
new
investigated.
(2)
studied
details,
achievements
discussed.
(3)
Some
models
suggestions
future
practical
applications.
(4)
categories
introduced
better
evaluation
various
methods.
Technologies,
Journal Year:
2023,
Volume and Issue:
11(3), P. 70 - 70
Published: May 26, 2023
In
the
energy-planning
sector,
precise
prediction
of
electrical
load
is
a
critical
matter
for
functional
operation
power
systems
and
efficient
management
markets.
Numerous
forecasting
platforms
have
been
proposed
in
literature
to
tackle
this
issue.
This
paper
introduces
an
effective
framework,
coded
Python,
that
can
forecast
future
based
on
hourly
or
daily
inputs.
The
framework
utilizes
recurrent
neural
network
model,
consisting
two
simpleRNN
layers
dense
layer,
adopts
Adam
optimizer
tanh
loss
function
during
training
process.
Depending
size
input
dataset,
system
handle
both
short-term
medium-term
load-forecasting
categories.
was
extensively
tested
using
multiple
datasets,
results
were
found
be
highly
promising.
All
variations
able
capture
underlying
patterns
achieved
small
test
error
terms
root
mean
square
absolute
error.
Notably,
outperformed
more
complex
networks,
with
0.033,
indicating
high
degree
accuracy
predicting
load,
due
its
ability
data
trends.
Journal of Cleaner Production,
Journal Year:
2023,
Volume and Issue:
411, P. 137357 - 137357
Published: April 29, 2023
The
growth
of
solar
photovoltaic
(PV)
waste
in
the
coming
years
requires
implementation
effective
management
options.
Australia,
with
one
highest
rates
rooftop
PV,
is
still
developing
policy
options
to
manage
these
panels
when
they
reach
their
end-of-life.
This
study
evaluates
environmental
impacts
three
for
mono
and
multi
crystalline
silicon
(c-Si)
panel
modules.
impact
transport
distance
from
transfer
stations
recycling
centre
also
assessed.
life
cycle
assessment
revealed
that,
-1
E+06
kgCO2eq
-2
are
associated
mandatory
product
stewardship
scenarios
under
global
warming
potential
c-Si
modules,
respectively.
However,
non-existence
a
will
produce
1
E+05
both
effects
collecting
most
were
not
(−365.00
kg
CO2-eq,
−698.40
−1032.00
CO2-eq)
compared
keeping
them
away
landfills
fully
(-2
them.
It
was
highlighted
regarding
distances
scenario
serving
over
107
kgCO2eq.
research
model
serves
as
first
conceptual
methodological
framework
(LCA)
related
analysis.
Since
incredibly
significant
PV
processes,
it
recommended
further
reduce
impacts,
other
forms
low-impact
modes
transportation
should
be
explored.
Energy Reports,
Journal Year:
2024,
Volume and Issue:
11, P. 5831 - 5844
Published: May 29, 2024
Electric
energy
demand
forecasting
is
vital
in
contemporary
power
systems,
especially
amidst
market
deregulation
trends
and
the
increasing
influence
of
industrial
customers
on
dynamics.
However,
existing
models
encounter
challenges
such
as
slow
convergence
high
complexity.
Addressing
these
issues,
this
study
proposes
a
hybrid
model
that
combines
Adaptive
Neuro-Fuzzy
Inference
System
(ANFIS)
with
Gene
Expression
Programming
(GEP)
to
enhance
predictions
electrical
consumption.
Validated
using
real-time
monthly
load
data
from
an
user
Uganda,
outperforms
individual
ANFIS
GEP
models,
demonstrating
reduced
errors
minimal
computation
time.
The
application
presents
promising
results,
showcasing
exceptional
predictive
capabilities
offering
potential
improvements
efficiency
precision
for
consumption
evolving
