Underground Space,
Год журнала:
2024,
Номер
18, С. 273 - 294
Опубликована: Апрель 26, 2024
This
study
aims
to
predict
the
migration
time
of
toxic
fumes
induced
by
excavation
blasting
in
underground
mines.
To
reduce
numerical
simulation
and
optimize
ventilation
design,
several
back
propagation
neural
network
(BPNN)
models
optimized
honey
badger
algorithm
(HBA)
with
four
chaos
mapping
(CM)
functions
(i.e.,
Chebyshev
(Che)
map,
Circle
(Cir)
Logistic
(Log)
Piecewise
(Pie)
map)
are
developed
time.
125
simulations
computational
fluid
dynamics
(CFD)
method
used
train
test
models.
The
determination
coefficient
(R2),
variance
accounted
for
(VAF),
Willmott's
index
(WI),
root
mean
square
error
(RMSE),
absolute
percentage
(MAPE),
sum
squares
(SSE)
utilized
evaluate
model
performance.
evaluation
results
indicate
that
CirHBA-BPNN
has
achieved
most
satisfactory
performance
reaching
highest
values
R2
(0.9945),
WI
(0.9986),
VAF
(99.4811%),
lowest
RMSE
(15.7600),
MAPE
(0.0343)
SSE
(6209.4),
respectively.
wind
velocity
roadway
(Wv)
is
important
feature
predicting
fumes.
Furthermore,
intrinsic
response
characteristic
optimal
implemented
enhance
interpretability
provide
reference
relationship
between
features
design.
Materials,
Год журнала:
2023,
Номер
16(3), С. 1286 - 1286
Опубликована: Фев. 2, 2023
The
application
of
aseismic
materials
in
foundation
engineering
structures
is
an
inevitable
trend
and
research
hotspot
earthquake
resistance,
especially
tunnel
engineering.
In
this
study,
the
pelican
optimization
algorithm
(POA)
improved
using
Latin
hypercube
sampling
(LHS)
method
Chaotic
mapping
(CM)
to
optimize
random
forest
(RF)
model
for
predicting
performance
a
novel
rubber-concrete
material.
Seventy
uniaxial
compression
tests
seventy
impact
were
conducted
quantify
material
performance,
i.e.,
strength
energy
absorption
properties
four
other
artificial
intelligence
models
generated
compare
predictive
with
proposed
hybrid
RF
models.
evaluation
results
showed
that
LHSPOA-RF
has
best
prediction
among
all
property
concrete
both
training
testing
phases
(R2:
0.9800
0.9108,
VAF:
98.0005%
91.0880%,
RMSE:
0.7057
1.9128,
MAE:
0.4461
0.7364;
R2:
0.9857
0.9065,
98.5909%
91.3652%,
0.5781
1.8814,
0.4233
0.9913).
addition,
sensitive
analysis
indicated
rubber
cement
are
most
important
parameters
properties,
respectively.
Accordingly,
POA-RF
not
only
proven
as
effective
predict
materials,
but
also
provides
new
idea
assessing
performances
field
Applied Sciences,
Год журнала:
2023,
Номер
13(4), С. 2574 - 2574
Опубликована: Фев. 16, 2023
Construction-induced
ground
settlement
is
a
serious
hazard
in
underground
tunnel
construction.
Accurate
prediction
has
great
significance
ensuring
the
surface
building’s
stability
and
human
safety.
To
that
end,
148
sets
of
data
were
collected
from
Singapore
Circle
Line
rail
traffic
project
containing
seven
defining
parameters
to
create
database
for
predicting
settlement.
These
are
depth
(H),
advance
rate
(AR),
EPB
earth
pressure
(EP),
mean
SPTN
value
soil
crown
(Sm),
water
content
layer
(MC),
modulus
elasticity
(E),
grout
used
injecting
into
tail
void
(GP).
Three
hybrid
models
consisting
random
forest
(RF)
three
types
meta-heuristics,
Ant
Lion
Optimizier
(ALO),
Multi-Verse
Optimizer
(MVO),
Grasshopper
Optimization
Algorithm
(GOA),
developed
predict
Furthermore,
absolute
error
(MAE),
percentage
(MAPE),
coefficient
determination
(R2)
root
square
(RMSE)
assess
predictive
performance
constructed
The
evaluation
results
demonstrated
GOA-RF
with
population
size
10
achieved
most
outstanding
capability
indices
MAE
(Training
set:
2.8224;
Test
2.3507),
MAPE
40.5629;
38.5637),
R2
0.9487;
0.9282),
RMSE
4.93;
3.1576).
Finally,
sensitivity
analysis
indicated
MC,
AR,
Sm,
GP
have
significant
impact
on
based
model.
Artificial Intelligence Review,
Год журнала:
2024,
Номер
57(6)
Опубликована: Май 15, 2024
Abstract
In
recent
years,
swarm
intelligence
optimization
algorithms
have
been
proven
to
significant
effects
in
solving
combinatorial
problems.
Introducing
the
concept
of
evolutionary
computing,
which
is
currently
a
hot
research
topic,
into
form
novel
has
proposed
new
direction
for
better
The
longhorn
beetle
whisker
search
algorithm
an
emerging
heuristic
algorithm,
originates
from
simulation
foraging
behavior.
This
simulates
touch
strategy
required
by
beetles
during
foraging,
and
achieves
efficient
complex
problem
spaces
through
bioheuristic
methods.
article
reviews
progress
on
2017
present.
Firstly,
basic
principle
model
structure
were
introduced,
its
differences
connections
with
other
analyzed.
Secondly,
this
paper
summarizes
achievements
scholars
years
improvement
algorithms.
Then,
application
various
fields
was
explored,
including
function
optimization,
engineering
design,
path
planning.
Finally,
proposes
future
directions,
deep
learning
fusion,
processing
multimodal
problems,
etc.
Through
review,
readers
will
comprehensive
understanding
status
prospects
providing
useful
guidance
practical
Nondestructive Testing And Evaluation,
Год журнала:
2024,
Номер
unknown, С. 1 - 24
Опубликована: Фев. 5, 2024
The
dynamic
compressive
strength
(DCS)
of
frozen-thawed
rock
influences
the
stability
mass
in
cold
regions,
especially
when
masses
are
possibly
disturbed
by
loads.
Laboratory
freeze-thaw
weathering
treatment
is
usually
time-consuming,
and
test
destructive.
Therefore,
this
paper
attempts
to
quickly
predict
DCS
sandstones
using
data-driven
methods,
non-destructive
properties,
basic
environmental
parameters.
sparrow
search
algorithm
(SSA),
gorilla
troops
optimiser,
dung
beetle
optimiser
were
chosen
develop
two
hyperparameters
random
forest
(RF).
classic
RF,
back
propagation
neural
network,
support
vector
regression
models
taken
as
control
group.
These
six
developed
DCS.
Their
prediction
results
compared.
Finally,
sensitivity
analysis
was
carried
out
assess
significance
all
input
variables.
indicate
that
SSA
–
RF
model
yields
best
result,
three
optimised
have
better
performance
than
single
machine-learning
models.
Strain
rate,
dry
density,
wave
velocity
found
be
most
important
parameters
prediction,
which
further
indicates
there
also
a
strong
correlation
between
characteristic
impedance
