Nature Communications,
Journal Year:
2024,
Volume and Issue:
15(1)
Published: Aug. 27, 2024
The
fracture
behaviour
of
artificial
metamaterials
often
leads
to
catastrophic
failures
with
limited
resistance
crack
propagation.
In
contrast,
natural
materials
such
as
bones
and
ceramics
possess
microstructures
that
give
rise
spatially
controllable
path
toughened
material
advances.
This
study
presents
an
approach
is
inspired
by
nature's
strengthening
mechanisms
develop
a
systematic
design
method
enabling
damage-programmable
engineerable
microfibers
in
the
cells
can
program
micro-scale
behaviour.
Machine
learning
applied
provide
effective
engine
accelerate
generation
offer
advanced
toughening
functionality
bowing,
deflection,
shielding
seen
materials;
are
optimised
for
given
programming
path.
paper
shows
features
effectively
enable
crack-resisting
on
basis
tip
interactions,
shielding,
bridging
synergistic
combinations
these
mechanisms,
increasing
up
1,235%
absorbed
energy
comparison
conventional
metamaterials.
proposed
have
broad
implications
damage-tolerant
materials,
lightweight
engineering
systems
where
significant
resistances
or
highly
programmable
damages
high
performances
sought
after.
Advanced Engineering Materials,
Journal Year:
2024,
Volume and Issue:
26(20)
Published: Aug. 28, 2024
Honeycombs
are
widely
used
in
engineering
protection,
while
the
gap
between
peak
and
mean
stresses
remains
to
be
narrowed,
interaction
effects
among
walls
weak.
To
break
these
limits,
gradient
curved‐walled
honeycombs
have
been
proposed
recently.
However,
their
in‐plane
crashworthiness
has
never
studied,
which
restricts
actual
applications
complex
load
environment.
For
this
purpose,
article
adopts
experiments,
finite‐element
simulations,
theoretical
analysis
reveal
crash
performance
of
honeycombs.
Quasistatic
dynamic
experiments
carried
out
for
3D‐printed
honeycomb
specimens
made
316L
stainless
steel,
numerical
simulations
conducted
by
ABAQUS/Explicit.
Compared
traditional
straight‐
honeycombs,
display
stabler
deformation
mode
more
efficient
mechanical
response.
Their
EA
SEA
respectively
25.5%
6.4%
larger
than
straight‐walled
energy
absorption
force
efficiencies
nearly
2.4
5.3
times
larger,
respectively.
On
basis,
plastic
hinge
model
with
high
accuracy
established,
derive
analytical
solutions
force–displacement
relations.
This
work
extends
comprehensive
properties
application
prospects
sets
an
example
develop
effective
simplification
desirable
on
complex‐shaped
structures.
Mathematics,
Journal Year:
2025,
Volume and Issue:
13(2), P. 264 - 264
Published: Jan. 15, 2025
In
tunnel
engineering,
joint
shear
slip
caused
by
external
disturbances
is
a
key
factor
contributing
to
landslides,
instability
of
surrounding
rock
masses,
and
related
hazards.
Therefore,
accurately
characterizing
the
macromechanical
properties
joints
essential
for
ensuring
engineering
safety.
Given
significant
influence
morphology
on
mechanical
behavior,
this
study
employs
frequency
spectrum
fractal
dimension
(D)
domain
amplitude
integral
(Rq)
as
quantitative
descriptors
morphology.
Using
Fourier
transform
techniques,
reconstruction
method
developed
model
with
arbitrary
shape
characteristics.
The
numerical
calibrated
through
3D
printing
direct
tests.
Systematic
parameter
analysis
validates
selected
indices
effective
Furthermore,
multiple
machine
learning
algorithms
are
employed
construct
robust
predictive
model.
Machine
learning,
recognized
rapidly
advancing
field,
plays
pivotal
role
in
data-driven
applications
due
its
powerful
analytical
capabilities.
study,
six
algorithms—Random
Forest
(RF),
Support
Vector
Regression
(SVR),
BP
Neural
Network,
GA-BP
Genetic
Programming
(GP),
ANN-based
MCD—are
evaluated
using
300
samples.
performance
each
algorithm
assessed
comparative
their
accuracy
based
correlation
coefficients.
results
demonstrate
that
all
achieve
satisfactory
performance.
Notably,
Random
(RF)
excels
rapid
accurate
predictions
when
handling
similar
training
data,
while
MCD
consistently
delivers
stable
precise
across
diverse
datasets.
