Sensors,
Journal Year:
2025,
Volume and Issue:
25(5), P. 1542 - 1542
Published: March 2, 2025
To
address
the
challenges
posed
by
complex
backgrounds
and
low
occurrence
in
photovoltaic
cell
images
captured
industrial
sensors,
we
propose
a
novel
defect
detection
method:
MRA-YOLOv8.
First,
multi-branch
coordinate
attention
network
(MBCANet)
is
introduced
into
backbone.
The
(CANet)
incorporated
to
mitigate
noise
impact
of
background
information
on
task,
multiple
branches
are
employed
enhance
model’s
feature
extraction
capability.
Second,
integrate
multi-path
module,
ResBlock,
neck.
This
module
provides
finer-grained
multi-scale
features,
improving
from
enhancing
robustness.
Finally,
implement
alpha-minimum
point
distance-based
IoU
(AMPDIoU)
head.
loss
function
enhances
accuracy
robustness
small
object
integrating
minimum
(MPDIoU)
Alpha-IoU
methods.
results
demonstrate
that
MRA-YOLOv8
outperforms
other
mainstream
methods
performance.
On
electroluminescence
anomaly
(PVEL-AD)
dataset,
proposed
method
achieves
mAP50
91.7%,
representing
an
improvement
3.1%
over
YOLOv8
16.1%
transformer
(DETR).
SPDI
our
69.3%,
showing
2.1%
6.6%
DETR.
also
exhibits
great
deployment
potential.
It
can
be
effectively
integrated
with
drone-based
inspection
systems,
allowing
for
efficient
accurate
PV
plant
inspections.
Moreover,
tackle
issue
data
imbalance,
generating
synthetic
via
generative
adversarial
networks
(GANs),
which
supplement
limited
samples
improve
generalization
ability.
Renewable and Sustainable Energy Reviews,
Journal Year:
2022,
Volume and Issue:
168, P. 112772 - 112772
Published: July 14, 2022
Irradiance-to-power
conversion
is
an
essential
step
of
state-of-the-art
photovoltaic
(PV)
power
forecasting,
regardless
the
source
and
post-processing
irradiance
forecasts.
The
two
distinct
approaches
for
mapping
forecasts
to
PV
are
physical
data-driven,
which
can
also
be
hybridized.
contribution
this
paper
twofold;
first,
it
proposes
a
concept
identifies
best
implementation
hybrid
machine
learning
irradiance-to-power
method.
Second,
head-to-head
comparison
physical,
methods
performed
operational
day-ahead
forecasting
14
plants
in
Hungary
based
on
numerical
weather
prediction
(NWP).
To
respect
rule
consistency
but
still
obtain
as
complete
picture
possible,
directives
set,
namely
minimizing
mean
absolute
error
(MAE)
root
square
(RMSE),
separate
sets
optimized
both
directives.
results
reveal
that
years
training
data,
method
involves
most
physically-calculated
predictors
reduce
MAE
by
5.2%
10.4%
compared,
respectively,
model
chains
without
any
considerations.
important
modeling
steps
separation
transposition
modeling,
rest
simulation
left
models
significant
increase
errors.
optimization
found
even
case
modeling;
therefore,
should
become
standard
procedure
practical
applications.
Finally,
only
beneficial
at
least
one
year
while
initial
period
operation
plant,
advised
stay
with
modeling.
guidelines
recommendations
help
researchers
practitioners
design
optimize
their
accuracy
Solar Energy,
Journal Year:
2023,
Volume and Issue:
252, P. 72 - 80
Published: Feb. 1, 2023
Probabilistic
solar
forecasts
may
take
the
form
of
predictive
probability
distributions,
ensembles,
quantiles,
or
interval
forecasts.
State-of-the-art
approaches
build
on
input
from
numerical
weather
prediction
(NWP)
models
and
post-processing
with
statistical
machine
learning
methods.
We
propose
a
probabilistic
benchmark
based
deterministic
forecast
clear-sky
irradiance,
introduce
new
methods
for
that
merge
techniques
modern
neural
networks,
discuss
spatio-temporal
scenario
forecasts,
illustrate
assessment
ability
via
proper
scoring
rules
calibration
checks.
expect
future
forecasting
efforts
to
be
increasingly
probabilistic,
encourage
continuing
close
interaction
operational
prediction,
where
innovations
sophisticated
networks
supplement
challenge
traditional
approaches.
Energies,
Journal Year:
2023,
Volume and Issue:
16(18), P. 6724 - 6724
Published: Sept. 20, 2023
Sibuyan
Island
is
experiencing
a
significant
increase
in
electricity
demand
due
to
population
growth,
urbanization,
and
industrial
development.
The
island
plans
use
solar
energy,
recognizing
its
abundance
renewable
nature;
thus,
this
study
was
conducted
visualize
the
spatial
distribution
of
exploration
suitability
using
geographic
information
system
(GIS).
Various
criteria,
including
climatology,
location,
geography,
meteorology,
disaster
susceptibility,
were
considered
assessment.
Parameters
affected
by
government
policies,
such
as
protected
areas,
proximity
rivers,
roads
faults,
ancestral
domains,
proclaimed
watersheds,
also
considered.
weighted,
levels
highlighted
AHP.
revealed
that
about
5.88%
(2674.06
km2)
categorized
highly
suitable
for
farm,
34.99%
(15,908.21
suitable,
2.49%
(1129.95
moderately
majority,
56.64%
(25,754.47
km2),
not
projects.
A
power
map
developed
reference
local
governments
residents
establishing
PV
systems
their
respective
sites,
thus
maximizing
full
potential
land.
directs
future
studies
Island,
supporting
ongoing
efforts
maximize
energy
utilization.
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.
