Modelling and Simulation in Engineering,
Journal Year:
2025,
Volume and Issue:
2025(1)
Published: Jan. 1, 2025
The
synthesis
problem
of
the
number
array
elements,
element
spacing,
and
formation
is
widely
concerned
in
sparse
optimization.
local
optimum
still
an
urgent
to
be
solved
existing
optimization
algorithms.
A
algorithm
on
improved
sparrow
search
(ISSA)
proposed
this
paper.
Firstly,
a
probabilistic
following
strategy
optimize
(SSA),
it
can
improve
global
capability
algorithm.
Secondly,
adaptive
Cauchy–Gaussian
mutation
are
used
avoid
falling
into
situation,
more
high‐quality
areas
searched
extremum
escape
ability
convergence
performance
Finally,
peak
sidelobe
level
(PSLL)
as
fitness
function
adaptively
position
elements.
Experimental
simulations
show
that
approach
has
good
main
lobe
response
low
response.
In
planar
array,
decreases
by
−1.41
dB
compared
with
genetic
(GA)
0.69
lower
than
SSA.
linear
−1.09
differential
evolution
0.40
arrays
significantly
enhances
accuracy
robustness
antenna
error
estimation.
Polymer Composites,
Journal Year:
2025,
Volume and Issue:
unknown
Published: April 9, 2025
Abstract
Fiber‐reinforced
polymer
(FRP)
tendons
are
preferred
in
civil
engineering
for
their
lightweight
properties,
high
strength,
corrosion
resistance,
and
electrical
insulation.
However,
initial
defects
that
arise
during
material
preparation
can
adversely
affect
the
mechanical
performance
service
life
of
structures.
Local
identification
technology
is
inadequate
FRP
products
with
variable
thickness
cross‐sections,
especially
tendons,
resulting
low
detection
efficiency.
This
article
presents
an
innovative
inverse
problem‐solving
framework
aimed
at
simultaneously
identifying
location
severity
through
frequency
change
rates.
A
convolutional
recurrent
neural
network
(CRNN)
model
was
developed
to
establish
mapping
between
rates
associated
damage
information,
including
severity.
The
CRNN
model's
database
generated
from
finite
element
models
(FEM),
which
were
validated
against
Euler
beam
vibration
theory,
demonstrating
absolute
error
less
than
1%.
trained
using
this
optimized
data
matrix
reconstruction,
refinement,
dilated
convolution,
achieving
a
mean
(Mae)
0.115%
predicting
rate.
significantly
surpassed
CNN
(0.318%),
MLP
(0.274%),
LSTM
(0.334%)
models.
served
as
surrogate
problem,
addressed
Slime
Mold
Algorithm
(SMA)
model.
prediction
SMA
under
0.5%,
notably
better
FEM.
Consequently,
identifies
defects'
offering
valuable
insights
applications
various
products.
Highlights
achieved
MAE
rates,
41.6%
MLP.
Optimized
identified
97.8%
accuracy.
Hammering
method
effectively
excited
first
8
frequencies
tendons.
Experimental
theoretical
errors
FEM
analysis
stayed
below
Modelling and Simulation in Engineering,
Journal Year:
2025,
Volume and Issue:
2025(1)
Published: Jan. 1, 2025
The
synthesis
problem
of
the
number
array
elements,
element
spacing,
and
formation
is
widely
concerned
in
sparse
optimization.
local
optimum
still
an
urgent
to
be
solved
existing
optimization
algorithms.
A
algorithm
on
improved
sparrow
search
(ISSA)
proposed
this
paper.
Firstly,
a
probabilistic
following
strategy
optimize
(SSA),
it
can
improve
global
capability
algorithm.
Secondly,
adaptive
Cauchy–Gaussian
mutation
are
used
avoid
falling
into
situation,
more
high‐quality
areas
searched
extremum
escape
ability
convergence
performance
Finally,
peak
sidelobe
level
(PSLL)
as
fitness
function
adaptively
position
elements.
Experimental
simulations
show
that
approach
has
good
main
lobe
response
low
response.
In
planar
array,
decreases
by
−1.41
dB
compared
with
genetic
(GA)
0.69
lower
than
SSA.
linear
−1.09
differential
evolution
0.40
arrays
significantly
enhances
accuracy
robustness
antenna
error
estimation.
