Integrated decision-making with adaptive feature weighting adversarial network for multi-target domain compound fault diagnosis of machinery
Advanced Engineering Informatics,
Год журнала:
2024,
Номер
62, С. 102730 - 102730
Опубликована: Июль 24, 2024
Язык: Английский
A rolling bearing fault diagnosis method under insufficient samples condition based on MSLSTM transfer learning
Journal of Vibroengineering,
Год журнала:
2025,
Номер
27(1), С. 93 - 107
Опубликована: Янв. 22, 2025
It
usually
affects
the
accuracy
and
reliability
of
deep
learning
based
intelligent
diagnosis
methods
under
condition
insufficient
samples.
Existing
for
handling
samples
often
have
problems
such
as
requiring
rich
expert
experience
or
consuming
a
lot
time.
To
solve
above
problems,
rolling
bearing
fault
method
on
multi-scale
long-term
short-term
memory
network
(MSLSTM)
transfer
is
proposed,
which
mainly
consists
an
improved
named
MSLSTM
learning.
By
introducing
convolution
operation
into
traditional
LSTM
to
improve
its
drawback
that
only
extracts
single
type
feature
information,
leads
poor
diagnostic
performance
in
noisy
environments.
Besides,
pooling
layer
global
average
are
replaced
with
avoid
problem
information
loss.
Subsequently,
combined
learning,
fine
tunes
model
parameters
using
small
amount
target
domain
data.
Feasibility
proposed
verified
through
two
kinds
experiments.
The
has
stronger
extraction
ability
training
efficiency
compared
other
models.
Язык: Английский
A Perspective on Artificial Intelligence for Process Manufacturing
Engineering,
Год журнала:
2025,
Номер
unknown
Опубликована: Фев. 1, 2025
Язык: Английский
Spectroscopic Advances in Real Time Monitoring of Pharmaceutical Bioprocesses: A Review of Vibrational and Fluorescence Techniques
Spectroscopy Journal,
Год журнала:
2025,
Номер
3(2), С. 12 - 12
Опубликована: Апрель 1, 2025
The
pharmaceutical
industry
has
witnessed
exponential
growth
in
production
volumes,
driven
by
factors
such
as
an
aging
global
population
and
the
COVID-19
pandemic.
To
meet
demand
for
high
product
quality
alongside
increased
productivity,
there
is
a
growing
emphasis
on
developing
innovative
Fermentation
Analytical
Technology
(FAT)
Process
(PAT)
tools
real-time
performance
monitoring,
modeling,
measurement,
control.
Building
our
earlier
work
involving
in-line
monitoring
of
Bordetella
pertussis
fermentations
using
fluorescence
spectroscopy,
this
review
explores
compares
applications
vibrational
spectroscopy
bioprocess
monitoring.
We
examine
recent
technological
advancements
ongoing
challenges
field.
Various
spectroscopic
techniques
are
evaluated
terms
cost-effectiveness
practical
applicability,
with
particular
focus
promising,
low-cost
solution
effective
Язык: Английский
Editorial overview: Digital design of pharmaceutical manufacturing processes
Current Opinion in Chemical Engineering,
Год журнала:
2025,
Номер
48, С. 101108 - 101108
Опубликована: Март 4, 2025
Язык: Английский
Unsupervised Hybrid Models Integrating Deep Autoencoders and Process Controllers’ Models for Enhanced Process Monitoring and Fault Detection
Industrial & Engineering Chemistry Research,
Год журнала:
2024,
Номер
63(33), С. 14748 - 14760
Опубликована: Авг. 12, 2024
This
paper
introduces
a
novel
hybrid
process
monitoring
model
that
integrates
long
short-term
memory
autoencoders
with
controllers'
models.
The
parameters
of
the
are
optimized
by
minimizing
loss
function,
which
combines
mean
square
error
(MSE)
between
controlled
variables
and
their
reconstructions
from
LSTM-AE
model,
along
MSE
manipulated
obtained
numerically
implemented
exactly
priori
known
controller
equations.
effectiveness
proposed
method
is
evaluated
on
benchmark
an
industrial-scale
penicillin
as
batch
case
study
Tennessee
Eastman
plant
under
decentralized
control
strategy
continuous
study.
A
comparative
analysis
equivalent
nonhybrid
does
not
utilize
equations,
highlights
superiority
in
fault
detection.
These
improvements
result
use
network
fewer
parameters,
thus
making
it
less
susceptible
to
overfitting.
Язык: Английский