Journal of Composites Science,
Год журнала:
2025,
Номер
9(4), С. 179 - 179
Опубликована: Апрель 7, 2025
Fiber-reinforced
polymers
(FRPs)
are
effective
for
strengthening
masonry
walls.
Debonding
at
the
polymer–masonry
interface
is
a
major
concern,
requiring
further
investigation
into
behavior.
This
study
utilizes
detailed
micro-modeling
finite
element
(FE)
analysis
to
predict
failure
mechanisms
and
analyze
behavior
of
brick
walls
strengthened
with
externally
bonded
carbon
fiber-reinforced
polymer
(CFRP)
under
in-plane
loading.
The
research
investigates
three
CFRP
configurations
(X,
I,
H).
FE
model
incorporates
nonlinear
components
using
Concrete
Damage
Plasticity
(CDP)
uses
cohesive
approach
unit–mortar
interfaces
bond
joints
between
CFRPs.
results
demonstrate
that
diagonal
reinforcement
enhances
ductility
capacity
wall
systems.
accurately
captures
crack
propagation,
fracture
mechanisms,
shear
strength
both
unreinforced
reinforced
confirms
can
reliably
structural
these
composite
Furthermore,
compares
predicted
strengths
established
design
equations,
highlighting
ACI
440.7R-10
CNR-DT
200/2013
models
as
providing
most
accurate
predictions
when
compared
experimental
results.
Archives of Civil and Mechanical Engineering,
Год журнала:
2024,
Номер
25(1)
Опубликована: Ноя. 12, 2024
Abstract
Conventional
ultra-high
performance
concrete
(UHPC)
has
excellent
development
potential.
However,
a
significant
quantity
of
CO
2
is
produced
throughout
the
cement-making
process,
which
in
contrary
to
current
worldwide
trend
lowering
emissions
and
conserving
energy,
thus
restricting
further
advancement
UHPC.
Considering
climate
change
sustainability
concerns,
cementless,
eco-friendly,
alkali-activated
UHPC
(AA-UHPC)
materials
have
recently
received
considerable
attention.
Following
emergence
advanced
prediction
techniques
aimed
at
reducing
experimental
tools
labor
costs,
this
study
provides
comparative
different
methods
based
on
machine
learning
(ML)
algorithms
propose
an
active
learning-based
ML
model
(AL-Stacked
ML)
for
predicting
compressive
strength
AA-UHPC.
A
data-rich
framework
containing
284
datasets
18
input
parameters
was
collected.
comprehensive
evaluation
significance
features
that
may
affect
AA-UHPC
performed.
Results
confirm
AL-Stacked
ML-3
with
accuracy
98.9%
can
be
used
general
specimens,
been
tested
research.
Active
improve
up
4.1%
enhance
Stacked
models.
In
addition,
graphical
user
interface
(GUI)
introduced
validated
by
tests
facilitate
comparable
prospective
studies
predictions.
