Optical Engineering,
Journal Year:
2024,
Volume and Issue:
63(12)
Published: Dec. 10, 2024
When
utilizing
the
tunable
diode
laser
absorption
spectroscopy
technique
for
gas
concentration
measurement,
second
harmonic
signal
is
significantly
affected
by
noise,
leading
to
a
decrease
in
measurement
accuracy.
We
proposed
noise
reduction
algorithm
address
this
issue.
It
combines
multi-resolution
singular
value
decomposition
(MRSVD)
and
an
improved
complete
ensemble
empirical
mode
with
adaptive
(ICEEMDAN)
optimized
crested
porcupine
optimizer
(ICPO)
algorithm.
First,
MRSVD
utilized
denoise
signal,
followed
optimizing
ICEEMDAN
parameters
using
ICPO
Subsequently,
decomposed
reconstructed
based
on
correlation
coefficients
of
intrinsic
functions
obtain
final
denoised
signal.
In
experiment,
achieved
signal-to-noise
ratio
116.8707
dB,
linearity
between
maximum
from
0.9893
0.9984.
Moreover,
average
relative
error
decreased
3.12%.
The
exhibits
superior
denoising
efficiency
compared
traditional
methods
proves
effective
open
optical
range
signals.
Photoacoustics,
Journal Year:
2025,
Volume and Issue:
42, P. 100692 - 100692
Published: Feb. 1, 2025
An
innovative
and
miniature
photoacoustic
spectroscopy
(PAS)
gas
sensor
based
on
a
multivariate-coupled
amplification
cell
(MVCA-PAC)
with
total
length
of
100
mm
was
developed
to
achieve
ultra-sensitive
trace
CH4
detection.
Acoustic
pressure
distribution
simulations
reveal
that
at
the
first-order
resonance
frequency,
MVCA-PAC
achieves
maximum
acoustic
approximately
3.9
times
higher
than
conventional
cell.
The
absorption
optical
path
reached
2068
through
22
reflections,
resulting
in
2-fold
increase
amplitude
signals
compared
traditional
an
equivalent
path.
Furthermore,
single-pass
cell,
2-f
signal
intensity
increased
by
factor
4.5.
Allan
variance
analysis
indicated
detection
limit
0.572
ppm
for
averaging
time
300
s.
To
further
improve
measurement
precision
designed
sensor,
Chebyshev
rational
fractional-order
filtering
(CRFOF)
algorithm
introduced
PAS
processing
first
time.
Post-processing
results
demonstrated
15.4-fold
improvement
precision,
achieving
0.578
ppm.
Finally,
continuous
monitoring
atmospheric
over
48-hour
period
validated
reliability
feasibility
sensor.
Review of Scientific Instruments,
Journal Year:
2025,
Volume and Issue:
96(1)
Published: Jan. 1, 2025
In
response
to
the
problem
of
noise
interference
in
knock
detection
signal
received
by
pickup
ceramic
sheet
non-destructive
testing,
a
removal
method
is
proposed
based
on
improved
secretary
bird
optimization
algorithm
(ISBOA)
optimized
variational
mode
decomposition
(VMD)
combined
with
wavelet
thresholding.
First,
using
Newton–Raphson
search
rule
and
smooth
exploitation
variation
strategy.
Second,
ISBOA
used
select
key
parameters
VMD.
Third,
subjected
VMD
obtain
intrinsic
functions
(IMFs),
permutation
entropy
each
IMF
component
calculated
divide
it
into
effective
or
component.
Finally,
denoised
thresholding,
processed
IMFs
components
are
reconstructed
signal.
The
denoising
simulated
5
dB
signal-to-noise
ratio
shows
that
11.59
root
mean
square
error
reduced
73.6%,
which
most
significant
effect
compared
other
similar
algorithms.
addition,
tests
signals
pieces
different
types
defects
also
show
has
wide
applicability
an
excellent
effect.
Optics Express,
Journal Year:
2024,
Volume and Issue:
32(19), P. 33698 - 33698
Published: Aug. 27, 2024
To
mitigate
the
susceptibility
of
absorption
lines
to
noise
interference
during
gas
measurement
using
off-axis
integrated
cavity
output
spectroscopy
(OA-ICOS)
technique,
thereby
enhancing
accuracy,
this
paper
introduces
a
novel
denoising
method.
This
method
synergistically
integrates
Grey
Wolf
Optimization
(GWO)
and
Cuckoo
Search
(CS)
with
Singular
Value
Decomposition
(SVD)
Sequential
Variational
Mode
(SVMD).
Initially,
optimal
solution
for
quadratic
penalty
coefficient
in
SVMD
is
ascertained
via
iterative
optimization
CSGWO
algorithm.
Subsequently,
circulant
matrix
established
extract
singular
values
each
modal
component.
A
threshold
set
discriminate
between
useful
signals,
corresponding
are
nullified.
The
signal
components
then
reconstructed,
yielding
final
processed
signal.
proposed
algorithm
was
applied
both
simulated
experimental
target
signals
compared
common
filtering
algorithms
such
as
WT,
VMD-WTD,
S-G.
In
processing
results,
signal-to-noise
ratio
(SNR)
spectrum
improved
from
21.4
39.95,
correlation
increased
0.99715
0.99946.
Results
indicate
that
exhibits
superior
identification
suppression
capabilities
other
algorithms.
After
CSGWO-SVMD-SVD
algorithm,
accuracy
stability
data
detected
based
on
integration
technology
have
been
greatly
improved.
Scientific Journal of Technology,
Journal Year:
2024,
Volume and Issue:
6(4), P. 17 - 24
Published: April 22, 2024
To
address
the
challenge
of
extracting
bearing
fault
features,
this
study
proposes
a
new
rolling
feature
extraction
method
based
on
Sparrow
Search
Algorithm
(SSA)
to
optimize
Variational
Mode
Decomposition
(VMD)
and
Multipoint
Optimal
Minimum
Entropy
Deconvolution
with
Convolution
Adjustment
(MOMEDA).
Firstly,
SSA
is
employed
identify
optimal
parameters
in
VMD,
followed
by
utilization
correlation
coefficients
kurtosis
filter
relevant
Intrinsic
Function
(IMF)
components.
Subsequently,
MOMEDA
applied
denoise
reconstructed
signal,
mitigating
interference
caused
pulse
signals.
Finally,
envelope
spectrum
analysis
conducted
denoised
signal.
Experimental
results
demonstrate
efficacy
proposed
features
noise
interference.
