Energies,
Journal Year:
2023,
Volume and Issue:
17(1), P. 97 - 97
Published: Dec. 23, 2023
Solar
forecasting
is
becoming
increasingly
important
due
to
the
exponential
growth
in
total
global
solar
capacity
each
year.
More
photovoltaic
(PV)
penetration
grid
poses
problems
for
stability
inherent
intermittent
and
variable
nature
of
PV
power
production.
Therefore,
quantities
becomes
operators
market
participants.
This
review
presents
most
recent
relevant
studies
focusing
on
short-term
irradiance
Recent
research
has
turned
machine
learning
address
this
challenge.
The
paper
provides
a
discussion
about
building
model,
including
evaluation
measures
method
selection
through
analysed
literature.
Given
that
data-driven,
focus
been
placed
data
sources
referenced
Open-access
have
compiled
explored.
main
contribution
establishment
benchmark
assessing
performance
models.
utilizes
mentioned
open-source
datasets,
offering
standardized
platform
future
research.
It
serves
crucial
purpose
streamlining
investigations
facilitating
direct
comparisons
among
different
methodologies
field
forecasting.
Sustainability,
Journal Year:
2023,
Volume and Issue:
15(4), P. 3666 - 3666
Published: Feb. 16, 2023
This
paper
presents
atechnical
and
economic
analysis
of
the
proposed
solar
PV/diesel
generator
smart
hybrid
power
plant
for
a
part
SRM
IST,
Delhi-NCR
campus.
The
was
performed
using
five
battery
storage
technologies:
lead-acid,
lithium-ion,
vanadium
flow,
zinc
bromide
nickel-iron.
also
used
HOMER
Pro
software.
conducted
to
assess
performance
parameters
such
as
initial
cost,
simple
payback
period,
return
on
investment,
energy
produced,
renewable
penetration
emission
air
pollutants.
optimal
solution
obtained
SPP(200
kW)/DG(82
kW)/ZB(2000
kWh),
with
cycle
charging
dispatch
strategy.
cost
this
configuration
is
estimated
be
USD163,445,
operating
USD534
per
year.
net
present
USD170,348,
has
been
USD0.090
kWh.
It
that
solution,
diesel
may
consume
only
110
L/year
diesel,
which
minimum
all
configurations.
Sensitivity
between
size
PV
array
battery,
along
variations
in
battery’s
nominal
capacity
fraction.
Advances in Atmospheric Sciences,
Journal Year:
2024,
Volume and Issue:
41(6), P. 1023 - 1067
Published: March 1, 2024
Abstract
Owing
to
the
persisting
hype
in
pushing
toward
global
carbon
neutrality,
study
scope
of
atmospheric
science
is
rapidly
expanding.
Among
numerous
trending
topics,
energy
meteorology
has
been
attracting
most
attention
hitherto.
One
essential
skill
solar
meteorologists
power
curve
modeling,
which
seeks
map
irradiance
and
auxiliary
weather
variables
power,
by
statistical
and/or
physical
means.
In
this
regard,
tutorial
review
aims
deliver
a
complete
overview
those
fundamental
scientific
engineering
principles
pertaining
curve.
Solar
curves
can
be
modeled
two
primary
ways,
one
regression
other
model
chain.
Both
classes
modeling
approaches,
alongside
their
hybridization
probabilistic
extensions,
allow
accuracy
improvement
uncertainty
quantification,
are
scrutinized
contrasted
thoroughly
review.
Journal of Forecasting,
Journal Year:
2024,
Volume and Issue:
43(5), P. 1173 - 1198
Published: Jan. 24, 2024
Abstract
The
creation
of
features
makes
the
difference
in
improving
photovoltaic
forecast
(PVF)
for
on‐grid,
hybrid
and
off‐grid
PV
systems.
importance
PVF
is
tremendous,
it
can
be
essential
optimizing
home
appliances
to
maximize
Renewable
Energy
Sources
(RES)
usage
or
create
performant
bids
electricity
market.
Several
use
cases
are
considered
from
connectivity
point
view.
Therefore,
this
paper,
we
propose
a
Weather
Prediction
Error
(WPE)‐based
method
that
uses
Stacking
Regressor
(SR)
various
systems
coexist
emerging
Communities
(EC)
landscape.
novelty
research
conduct
consists
proposing
several
determining
coefficients
adjust
based
on
WPE.
results
four
types
size
view
investigated.
Compared
with
individual
Machine
Learning
(ML)
models,
R
2
increases
more
than
3%
SR
6%
after
applying
adjustment
coefficients.
Nevertheless,
major
improvement
recorded
inverter
large
industrial
power
plant,
demonstrating
proposed
model
suitable
these
other
metrics
improved
as
well,
especially
Mean
Average
(MAE)
decreases
between
10%
23%.
A
significant
decrease
case
on‐grid
PV,
130
kW
123
using
107
adjustments.
This
represents
around
18%
initial
MAE.
ratio
daily
deviations
also
SR.
For
all
systems,
stabilizes
shorter
interval,
values
0.78
1.33
obtained
ML
models
0.83
1.24
After
final
adjustments,
interval
becomes
shorter,
having
0.90
1.10.
