Research Square (Research Square),
Journal Year:
2023,
Volume and Issue:
unknown
Published: Dec. 27, 2023
Abstract
Rockfall
is
one
of
the
primary
geological
hazards
in
karst
regions.
In
order
to
study
susceptibility
distribution
patterns
rockfall
disasters
areas,
research
areain
Xincheng
County
selected
this
and
data
are
collected
at
172
historical
points
under
different
environments.
Various
factors,
including
aspect,
slope,
elevation,
terrain
relief,
plan
curvature,
profile
landform
type,
roughness,
coefficient
variation,
lithology,
fault
distance,
rainfall,
distance
rivers,
NDVI
(Normalized
Difference
Vegetation
Index),
roads,
employed
construct
four
coupling
models,
e.g.
IV-RF,
IV-CHAID,
IV-MLP
IV-SVM.
Through
comparative
analysis
accuracy
reliability
these
optimal
evaluation
model
determined.
The
results
indicate
corresponding
AUC
(Area
Under
Curve)
values
for
IV-MLP,
IV-SVM,
0.854,
0.86,
0.862,
0.888,
respectively.
For
prediction
variation
identified
as
most
significant
accounting
21%,
18%,
11%,
These
factors
indirectly
promote
water
movement
consequently
influencing
occurrences.
Unmanned
aerial
vehicle
(UAV)
systems
are
widely
used
in
many
forest-related
fields
owing
to
their
cost-intensive
and
precise
surveying
technology.
This
study
classified
erosion
susceptibility
(ES)
a
timber
harvesting
area
using
machine
learning
(ML)
statistical
approaches.
In
dataset
generation
for
the
training
testing
process,
digital
surface
model
(DSM)
of
difference
(DoD)
July–June
was
utilized
as
dependent
variable,
six
terrain
maps
DSM
June
were
independent
variables.
The
ES
threshold
set
at
5
cm
binary
classification
pixels
while
processing
ML
(e.g.,
random
forest
extra
gradient
boost
[XGB])
logistic
regression)
algorithms
development.
overall
accuracy
(OA),
receiver
operating
characteristics,
under
curve
(AUC)
calculated
validation.
Although
AUC
all
models
did
not
appear
acceptable
(AUC
>
0.7),
XGB
showed
best
performance
regarding
time
duration,
OA,
by
2
h,
64%,
0.63,
respectively.
Despite
low
model,
wheel
tracks
edges
operation
road
determined
be
susceptible
areas
map
XGB.
Research Square (Research Square),
Journal Year:
2024,
Volume and Issue:
unknown
Published: July 29, 2024
Abstract
Identifying
the
prone
sites
and
recognizing
influencing
factors
of
rock
failure
remains
a
major
challenge,
especially
for
regions
lacking
historical
database
chronological
evolution
different
potential
frequency
amplitude
this
hazard
in
mountain
zones.
In
context,
present
study
aims
to
delineate
movement
rocky
masses
after
frequent
torrential
rainfall
assess
main
driving
landslide
hazards
Matmata
region
(SE
Tunisia).
The
used
approach
relies
on
field
observations,
remotely
sensed
data,
digital
photogrammetry,
GIS-multi
criteria
assessment.
analysis
kinematics
cliffs
triggering
between
2016
2023
highlights
relative
about
39
m
carbonate
related
impacts
geological
factors,
weathering,
land
use
changes,
hydrogeology,
human
activities
slope
stability
rockfall
occurrences.
hierarchical
influence
these
illustrates
relevant
spatio-temporal
variability
susceptibility
indices.
southern
part
is
characterized
by
highest
degree
vulnerability
due
many
such
as
slope,
lithology.
spatial
distribution
final
index
indicates
varying
degrees
across
area
amplified
during
last
years
given
extreme
events.
map
validated
inventory.
findings
highlight
relevance
explained
high
urban
expansion
infrastructure
development
hilly
areas.
obtained
results
valuable
tool
decision-making
management
mitigation
measures.
Processes,
Journal Year:
2024,
Volume and Issue:
12(12), P. 2705 - 2705
Published: Nov. 30, 2024
With
the
rapid
expansion
of
market
and
increase
in
pig
farming
density,
improving
automation
intelligence
farms
has
become
key.
Despite
continuous
progress
this
field,
there
is
still
a
lack
intelligent
systems
for
cleaning
disinfecting
pigs.
In
paper,
we
conduct
research
from
perspective
product
functional
requirements.
By
conducting
to
obtain
raw
data
on
user
needs,
Analytic
Hierarchy
Process
(AHP)
used
hierarchical
analysis
needs.
The
demand
indicator
system
washing
disinfection
integrated
summarized
at
three
levels:
goals,
criteria,
indicators.
Combined
with
competitor
literature
methods,
obtained
operative
words
design
Using
Quality
Function
Deployment
(QFD)
convert
requirements
into
performance
indicators
design,
quality
house
model
constructed.
We
next
analyzed
terms
functionality,
usage,
safety,
appearance,
completed
conceptual
design.
Finally,
improved
mechanical
structure
mobile
nozzle,
supplemented
control
relied
upon
nozzle
movement,
enhanced
scientific
rational
This
study
provides
new
ideas
development
equipment
farms,
promoting
precision
farming.
Research Square (Research Square),
Journal Year:
2023,
Volume and Issue:
unknown
Published: May 2, 2023
Abstract
Landslides
often
cause
great
losses
to
people,
so
it
is
important
know
the
extent
of
landslide
damage
reduce
impact
disasters.
Although
many
scholars
have
conducted
on
disaster
susceptibility,
there
are
still
some
issues,
such
as
an
unreasonable
negative
sample
selection
strategy
and
absence
subjective
environmental
information
study
area
in
a
single
machine
learning
evaluation
model.
Therefore,
analytic
hierarchy
process
(AHP)
method
weighted
by
improved
Random
Forest
(RF)
model
proposed
for
evaluating
susceptibility
based
optimization.
On
basis
density
analysis
data,
this
employs
specific
factor
(CF)
generate
data.
The
RF
adaptive
boosting
(ADB_RF)
obtain
objective
weights,
which
then
combined
with
weights
obtained
AHP.
Meanwhile,
case
disasters
Chuxiong
Autonomous
Prefecture
Yunnan
Province
China.
results
show:
(1)
can
objectively
reflect
prone
landslides
high
accuracy
effectiveness.
(2)
under
line
CF-combination
reached
96.1%,
indicating
degree
accuracy.
(3)
In
northwest
Prefecture,
greater
number
extremely
high-risk
areas
than
southeast,
possibility
another
high,
needs
be
focused.
research
findings
significant
reference
value
preventing
mitigating
losses.
Research Square (Research Square),
Journal Year:
2023,
Volume and Issue:
unknown
Published: Dec. 27, 2023
Abstract
Rockfall
is
one
of
the
primary
geological
hazards
in
karst
regions.
In
order
to
study
susceptibility
distribution
patterns
rockfall
disasters
areas,
research
areain
Xincheng
County
selected
this
and
data
are
collected
at
172
historical
points
under
different
environments.
Various
factors,
including
aspect,
slope,
elevation,
terrain
relief,
plan
curvature,
profile
landform
type,
roughness,
coefficient
variation,
lithology,
fault
distance,
rainfall,
distance
rivers,
NDVI
(Normalized
Difference
Vegetation
Index),
roads,
employed
construct
four
coupling
models,
e.g.
IV-RF,
IV-CHAID,
IV-MLP
IV-SVM.
Through
comparative
analysis
accuracy
reliability
these
optimal
evaluation
model
determined.
The
results
indicate
corresponding
AUC
(Area
Under
Curve)
values
for
IV-MLP,
IV-SVM,
0.854,
0.86,
0.862,
0.888,
respectively.
For
prediction
variation
identified
as
most
significant
accounting
21%,
18%,
11%,
These
factors
indirectly
promote
water
movement
consequently
influencing
occurrences.
