Neural Computing and Applications,
Год журнала:
2024,
Номер
36(27), С. 17145 - 17163
Опубликована: Июнь 6, 2024
Abstract
Sustainable
Development
Goal
7
is
dedicated
to
ensuring
access
clean
and
affordable
energy
that
can
be
utilized
in
various
applications.
Solar
panels
(SP)
are
convert
sunlight
into
electricity,
acting
as
a
renewable
source.
It
important
keep
SP
obtain
the
required
performance,
accumulation
of
snow
dust
on
greatly
affects
amount
electricity
generated.
On
other
hand,
excessive
cleaning
has
some
detrimental
effects
SP,
therefore
should
only
done
when
necessary
not
regular
basis.
Consequently,
it
critical
determine
whether
procedure
by
automatically
detecting
presence
or
while
avoiding
inaccurate
predictions.
Research
efforts
have
been
made
detect
but
most
proposed
methods
do
guarantee
accurate
detection
results.
This
paper
proposes
an
accurate,
reliable,
interpretable
approach
called
Solar-OBNet.
The
Solar-OBNet
dusty
snow-covered
very
efficiently
used
conjunction
with
SP.
based
Bayesian
convolutional
neural
network,
which
enables
express
confidence
its
Two
measurements
estimate
uncertainty
outcomes
Solar-OBNet,
namely
predictive
entropy
standard
deviation.
correct
predictions
showing
low
values
for
also
give
warning
case
erroneous
high
Solar-OBNet’s
efficacy
was
verified
interpreting
results
using
method
Weighted
Gradient-Directed
Class
Activation
Mapping
(Grad-CAM).
achieved
balanced
accuracy
94.07%
average
specificity
95.83%,
outperforming
comparable
methods.
Applied Water Science,
Год журнала:
2024,
Номер
14(11)
Опубликована: Окт. 15, 2024
In
response
to
increasing
flood
risks
driven
by
the
climate
crisis,
urban
areas
require
advanced
forecasting
and
informed
decision-making
support
sustainable
development.
This
study
seeks
improve
reliability
of
reservoir-based
ensure
adequate
lead
time
for
effective
measures.
The
main
objectives
are
predict
hourly
downstream
discharge
at
a
reference
point,
compare
predictions
from
single
reservoir
with
four-hour
against
those
three
reservoirs
seven-hour
time,
evaluate
accuracy
data-driven
approaches.
takes
place
in
Han
River
Basin,
located
Seoul,
South
Korea.
Approaches
include
two
non-deep
learning
(NDL)
(random
forest
(RF),
vector
regression
(SVR))
deep
(DL)
(long
short-term
memory
(LSTM),
gated
recurrent
unit
(GRU)).
Scenario
1
incorporates
data
reservoirs,
while
2
focuses
solely
on
Paldang
reservoir.
Results
show
that
RF
performed
4.03%
(in
R2)
better
than
SVR,
GRU
4.69%
LSTM
1.
2,
none
models
showed
any
outstanding
performance.
Based
these
findings,
we
propose
two-step
approach:
Initial
should
utilize
upstream
long
closer
event,
model
focus
more
accurate
prediction.
work
stands
as
significant
contribution,
making
well-timed
local
administrations
issue
warnings
execute
evacuations
mitigate
damage
casualties
areas.
Remote Sensing,
Год журнала:
2025,
Номер
17(3), С. 375 - 375
Опубликована: Янв. 23, 2025
Flood
susceptibility
provides
scientific
support
for
flood
prevention
planning
and
infrastructure
development
by
identifying
assessing
flood-prone
areas.
The
uncertainty
posed
non-flood
sample
datasets
remains
a
key
challenge
in
mapping.
Therefore,
this
study
proposes
novel
sampling
method
points.
A
model
is
constructed
using
machine
learning
algorithm
to
examine
the
due
point
selection.
influencing
factors
of
are
analyzed
through
interpretable
models.
Compared
generated
random
with
buffer
method,
dataset
spatial
range
identified
frequency
ratio
one-class
vector
achieves
higher
accuracy.
This
significantly
improves
simulation
accuracy
model,
an
increase
24%
ENSEMBLE
model.
(2)
In
constructing
optimal
dataset,
demonstrates
than
other
methods,
AUC
0.95.
(3)
northern
southeastern
regions
Zijiang
River
Basin
have
extremely
high
susceptibility.
Elevation
drainage
density
as
causing
these
areas,
whereas
southwestern
region
exhibits
low
elevation.
(4)
Elevation,
slope,
three
most
important
affecting
Lower
values
elevation
slope
correlate
offers
new
approach
reducing
technical
disaster
mitigation
basin.
VIETNAM JOURNAL OF EARTH SCIENCES,
Год журнала:
2025,
Номер
unknown
Опубликована: Апрель 16, 2025
The
Mekong
Basin
is
the
most
critical
transboundary
river
basin
in
Asia.
This
provides
an
abundant
source
of
fresh
water
essential
for
development
agriculture,
domestic
consumption,
and
industry,
as
well
production
hydroelectricity,
it
also
contributes
to
ensuring
food
security
worldwide.
region
often
subject
floods
that
cause
significant
damage
human
life,
society,
economy.
However,
flood
risk
management
challenges
this
are
increasingly
substantial
due
conflicting
objectives
between
several
countries
data
sharing.
study
integrates
deep
learning
with
optimization
algorithms,
namely
Grasshopper
Optimisation
Algorithm
(GOA),
Adam
Stochastic
Gradient
Descent
(SGD),
open-source
datasets
identify
probably
occurring
basin,
covering
Vietnam
Cambodia.
Various
statistical
indices,
Area
Under
Curve
(AUC),
root
mean
square
error
(RMSE),
absolute
(MAE),
coefficient
determination
(R²),
were
used
evaluate
susceptibility
models.
results
show
proposed
models
performed
AUC
values
above
0.8,
specifying
DNN-Adam
model
achieved
0.98,
outperforming
DNN-GOA
(AUC
=
0.89),
DNN-SGD
0.87),
XGB
0.82.
Regions
very
high
concentrated
Delta
along
River
findings
supporting
decision-makers
or
planners
proposing
appropriate
mitigation
strategies,
planning
policies,
particularly
watershed.