Case Studies in Construction Materials,
Journal Year:
2024,
Volume and Issue:
21, P. e03416 - e03416
Published: June 14, 2024
Performance
assessment
of
existing
building
structures,
especially
concrete
compressive
strength
assessment,
is
a
crucial
aspect
engineering
construction
for
most
industrialized
countries.
Non-destructive
testing
(NDT)
techniques
are
commonly
employed
to
assess
the
structures.
However,
methods
predicting
using
NDT
and
machine
learning
do
not
take
into
account
mix
proportion
design.
This
study
proposes
an
effective
method
predict
by
combining
tests
with
different
designs
curing
ages.
Specifically,
support
vector
regression
(SVR)
back
propagation
neural
network
(BPNN)
models
established.
Furthermore,
various
evaluation
indexes
utilized
model
performance.
To
construct
validate
prediction
models,
total
180
datasets
containing
specimens
ages
collected
from
research
literature.
The
results
show
that
coefficients
determination
(R2)
SVR
BPNN
test
set
86.0
%
86.7
without
considering
R2
higher
than
95
when
effects
design
age.
ranged
between
92
97
%.
All
better
those
model.
Consequently,
can
be
accurately
evaluate
during
structural
performance
buildings.
Crystals,
Journal Year:
2020,
Volume and Issue:
10(9), P. 741 - 741
Published: Aug. 22, 2020
The
complication
linked
with
the
prediction
of
ultimate
capacity
concrete-filled
steel
tubes
(CFST)
short
circular
columns
reveals
a
need
for
conducting
an
in-depth
structural
behavioral
analyses
this
member
subjected
to
axial-load
only.
distinguishing
feature
gene
expression
programming
(GEP)
has
been
utilized
establishing
model
axial
behavior
long
CFST.
proposed
equation
correlates
CFST
depth,
thickness,
yield
strength
steel,
compressive
concrete
and
length
CFST,
without
any
expensive
laborious
experiments.
A
comprehensive
column
under
load
was
obtained
from
extensive
literature
build
models,
subsequently
implemented
verification
purposes.
This
consists
database
is
comprised
227
data
samples.
External
validations
were
carried
out
using
several
statistical
criteria
recommended
by
researchers.
developed
GEP
demonstrated
superior
performance
available
design
methods
AS5100.6,
EC4,
AISC,
BS,
DBJ
AIJ
codes.
equations
can
be
reliably
used
pre-design
purposes—or
may
as
fast
check
deterministic
solutions.
Materials,
Journal Year:
2020,
Volume and Issue:
13(17), P. 3902 - 3902
Published: Sept. 3, 2020
When
designing
flat
slabs
made
of
steel
fiber-reinforced
concrete
(SFRC),
it
is
very
important
to
predict
their
punching
shear
capacity
accurately.
The
use
machine
learning
seems
be
a
great
way
improve
the
accuracy
empirical
equations
currently
used
in
this
field.
Accordingly,
study
utilized
tree
predictive
models
(i.e.,
random
forest
(RF),
(RT),
and
classification
regression
trees
(CART))
as
well
novel
feature
selection
(FS)
technique
introduce
new
model
capable
estimating
SFRC
slabs.
Furthermore,
automatically
create
structure
models,
current
employed
sequential
algorithm
FS
model.
In
order
perform
training
stage
for
proposed
dataset
consisting
140
samples
with
six
influential
components
depth
slab,
effective
length
column,
compressive
strength
concrete,
reinforcement
ratio,
fiber
volume)
were
collected
from
relevant
literature.
Afterward,
trained
verified
using
above-mentioned
database.
To
evaluate
both
testing
datasets,
various
statistical
indices,
including
coefficient
determination
(R2)
root
mean
square
error
(RMSE),
utilized.
results
obtained
experiments
indicated
that
FS-RT
outperformed
FS-RF
FS-CART
terms
prediction
accuracy.
range
R2
RMSE
values
0.9476–0.9831
14.4965–24.9310,
respectively;
regard,
hybrid
demonstrated
best
performance.
It
was
concluded
three
techniques
paper,
i.e.,
FS-RT,
FS-RF,
FS-CART,
could
applied
predicting
Mechanics of Advanced Materials and Structures,
Journal Year:
2020,
Volume and Issue:
29(12), P. 1782 - 1797
Published: Nov. 3, 2020
In
this
paper,
a
surrogate
Machine-Learning
(ML)
model
based
on
Gaussian
Process
Regression
(GPR)
was
developed
to
predict
the
axial
load
of
square
concrete-filled
steel
tubular
(CFST)
columns
under
compression.
For
purpose,
an
experimental
database
extracted
from
available
literature
and
used
for
development
training
GPR
model.
The
model’s
performance
is
superior
that
existing
models
in
relation
CFST
columns.
practical
application,
Graphical
User
Interface
(GUI)
researchers,
engineers
support
teaching
interpretation
behavior
Computer Modeling in Engineering & Sciences,
Journal Year:
2020,
Volume and Issue:
125(2), P. 815 - 828
Published: Jan. 1, 2020
The
modeling
and
risk
assessment
of
a
pandemic
phenomenon
such
as
COVID-19
is
an
important
complicated
issue
in
epidemiology,
attempt
great
interest
for
public
health
decision-making.
To
this
end,
the
present
study,
based
on
recent
heuristic
algorithm
proposed
by
authors,
time
evolution
investigated
six
different
countries/states,
namely
New
York,
California,
USA,
Iran,
Sweden
UK.
number
COVID-19-related
deaths
used
to
develop
model
it
believed
that
predicted
daily
each
country/state
includes
information
about
quality
system
area,
age
distribution
population,
geographical
environmental
factors
well
other
conditions.
Based
derived
epidemic
curves,
new
3D-epidemic
surface
assess
at
any
its
evolution.
This
research
highlights
potential
tool
which
can
assist
COVID-19.
Mapping
development
through
revealing
dynamic
nature
differences
similarities
among
districts.
Journal of Rock Mechanics and Geotechnical Engineering,
Journal Year:
2022,
Volume and Issue:
14(5), P. 1588 - 1608
Published: Jan. 28, 2022
The
study
proposes
an
improved
Harris
hawks
optimization
(IHHO)
algorithm
by
integrating
the
standard
(HHO)
and
mutation-based
search
mechanism
for
developing
a
high-performance
machine
learning
solution
predicting
soil
compression
index.
HHO
is
newly
introduced
meta-heuristic
(MOA)
used
to
solve
continuous
problems.
Compared
original
HHO,
proposed
IHHO
can
evade
trapping
in
local
optima,
which
turn
raises
capabilities
enhances
relying
on
mutation.
Subsequently,
novel
meta-heuristic-based
soft
computing
technique
called
ELM-IHHO
was
established
extreme
(ELM)
estimate
A
sum
of
688
consolidation
test
data
collected
this
purpose
from
ongoing
dedicated
freight
corridor
railway
project.
To
evaluate
generalization
capability
model,
detailed
comparison
between
other
well-established
MOAs,
such
as
particle
swarm
optimization,
genetic
algorithm,
biogeography-based
integrated
with
ELM,
performed.
Based
outcomes,
model
exhibits
superior
performance
over
MOAs
Archives of Computational Methods in Engineering,
Journal Year:
2022,
Volume and Issue:
30(3), P. 1979 - 2012
Published: Dec. 20, 2022
Abstract
The
article
reviewed
the
four
major
Bioinspired
intelligent
algorithms
for
agricultural
applications,
namely
ecological,
swarm-intelligence-based,
ecology-based,
and
multi-objective
algorithms.
key
emphasis
was
placed
on
variants
of
swarm
intelligence
algorithms,
artificial
bee
colony
(ABC),
genetic
algorithm,
flower
pollination
algorithm
(FPA),
particle
swarm,
ant
colony,
firefly
fish
Krill
herd
because
they
had
been
widely
employed
in
sector.
There
a
broad
consensus
among
scholars
that
certain
BIAs'
were
more
effective
than
others.
For
example,
Ant
Colony
Optimization
Algorithm
best
suited
farm
machinery
path
optimization
pest
detection,
other
applications.
On
contrary,
useful
determining
plant
evapotranspiration
rates,
which
predicted
water
requirements
irrigation
process.
Despite
promising
adoption
hyper-heuristic
agriculture
remained
low.
No
universal
could
perform
multiple
functions
farms;
different
designed
to
specific
functions.
Secondary
concerns
relate
data
integrity
cyber
security,
considering
history
cyber-attacks
smart
farms.
concerns,
benefits
associated
with
BIAs
outweighed
risks.
average,
farmers
can
save
647–1866
L
fuel
is
equivalent
US$734-851,
use
GPS-guided
systems.
accuracy
mitigated
risk
errors
applying
pesticides,
fertilizers,
irrigation,
crop
monitoring
better
yields.
