Extreme
meteorological
events
and
rapid
urbanization
have
led
to
serious
urban
flooding
problems.
Characterizing
spatial
variations
in
susceptibility
elucidating
its
driving
factors
are
essential
for
preventing
damages
from
pluvial
flooding.
However,
conventional
methods,
limited
by
heterogeneity
the
intricate
mechanisms
of
flooding,
frequently
underestimated
dense
areas.
Therefore,
this
study
proposed
a
comprehensive
cascade
modeling
chain
consisting
XGBoost,
SHapley
Additive
exPlanations
(SHAP),
Partial
Dependence
Plots
(PDP)
combination
with
K-means
recognizing
specific
influence
morphology
patterns
risk
agglomeration
under
different
rainfall
scenarios.
The
XGBoost
model
demonstrated
enhanced
accuracy
robustness
relative
other
three
benchmark
models:
RF,
SVR,
BPDNN.
This
superiority
was
effectively
validated
during
both
training
independent
testing
Shenzhen.
results
indicated
that
3D
characteristics
were
dominant
waterlogging
magnitude,
which
occupied
46.02%
contribution.
Through
PDP
analysis,
multi-staged
trends
highlighted
critical
thresholds
interactions
between
significant
indicators
like
building
congestion
degree
(BCD)
floor
area
ratio
(FAR).
Specifically,
optimal
intervals
BCD
0
0.075
coupled
FAR
values
0.5
1
potential
substantially
mitigate
risks.
These
findings
emphasize
need
strategic
configuration
within
planning
frameworks.
In
terms
spatial-temporal
assessment,
aggregation
effect
high-risk
areas
prone
prolonged
duration
or
high-intensity
scenarios
emerged
old
districts.
approach
present
will
provide
quantitative
insights
into
adaptation
strategies
sustainable
design.
Atmosphere,
Год журнала:
2024,
Номер
15(10), С. 1250 - 1250
Опубликована: Окт. 19, 2024
This
research
investigates
the
application
of
machine
learning
models
to
optimise
renewable
energy
systems
and
contribute
achieving
Net
Zero
emissions
targets.
The
primary
objective
is
evaluate
how
can
improve
forecasting,
grid
management,
storage
optimisation,
thereby
enhancing
reliability
efficiency
sources.
methodology
involved
various
models,
including
Long
Short-Term
Memory
(LSTM),
Random
Forest,
Support
Vector
Machines
(SVMs),
ARIMA,
predict
generation
demand
patterns.
These
were
evaluated
using
metrics
such
as
Mean
Absolute
Error
(MAE)
Root
Squared
(RMSE).
Key
findings
include
a
15%
improvement
in
after
optimisation
10–20%
increase
battery
efficiency.
Forest
achieved
lowest
MAE,
reducing
prediction
error
by
approximately
8.5%.
study
quantified
CO2
emission
reductions
source,
with
wind
power
accounting
for
15,000-ton
annual
reduction,
followed
hydropower
solar
10,000
7500
tons,
respectively.
concludes
that
significantly
enhance
system
performance,
measurable
errors
emissions.
improvements
could
help
close
“ambition
gap”
20%,
supporting
global
efforts
meet
1.5
°C
Paris
Agreement
Heliyon,
Год журнала:
2023,
Номер
9(8), С. e18794 - e18794
Опубликована: Июль 29, 2023
Smart
cities
have
been
introduced
globally.
It
involves
technical
development
and
economic,
social,
environmental
objectives.
In
response
to
the
Fourth
Industrial
Revolution
(Industry
4.0)
global
trends,
Korea
has
prepared
legal
institutional
measures
for
smart
city
composition.
This
study
reviewed
importance
of
key
documents
agreements
in
Daegu
Metropolitan
City
reduce
disaster
risk
vulnerable
context
cities.
25
research
studies
were
critically
systematically
from
perspective
reduction
its
safety
areas,
aims
property
damage
casualties
that
may
occur
because
physical
events
such
as
collapse,
water-related
disasters,
heatwaves
by
up
20%.
mitigation
data
collection,
sharing,
propagation.
The
entire
process
is
handled
on
a
platform
called
Data
hub.
According
government,
solving
social
problems
managing
disasters
city,
it
striving
improve
efficiency
other
However,
limitations
service-oriented
necessary
engage
citizens
participate,
raise
awareness
educate
them
platform.
results
recommend
future
focus
resilience
worldwide.
Land,
Год журнала:
2024,
Номер
13(9), С. 1464 - 1464
Опубликована: Сен. 10, 2024
Extreme
climatic
conditions
cause
a
decrease
in
ecosystem
services,
the
disruption
of
ecological
balance,
and
damage
to
human
populations,
especially
areas
with
socially
vulnerable
groups.
Nature-based
solutions
applying
blue-green
infrastructure
(BGI)
against
these
negative
impacts
climate
change
have
an
important
role
planning
sustainable
cities.
This
study
aims
identify
priority
develop
scenarios
strategies
for
spatial
understand
tradeoffs
approaches
maximize
benefits
services
provided
by
BGI
cities
arid
semi-arid
climates,
using
Phoenix,
Arizona,
swiftly
urbanizing
city
Sonoran
Desert,
as
area.
Using
GIS-based
multi-criteria
decision-making
techniques
Green
Infrastructure
Spatial
Planning
model
integrated
city’s
existing
water
structures,
this
is
conducted
at
US
census
scale.
The
hotspots
are
mapped
from
combined
evaluation
expert
stakeholder-driven
weighting.
In
where
Phoenix
identified,
center
area
high
density
impervious
surfaces
identified
highest
It
revealed
that
social
vulnerability
environmental
risks
(flooding,
heat)
positive
correlation
stormwater
management
urban
heat
island
criteria
should
be
considered
first
planning.
