Energies,
Journal Year:
2024,
Volume and Issue:
17(11), P. 2487 - 2487
Published: May 22, 2024
Battery
state
of
health
(SOH),
which
is
a
crucial
parameter
the
battery
management
system,
reflects
rate
performance
degradation
and
aging
level
lithium-ion
batteries
(LIBs)
during
operation.
However,
traditional
machine
learning
models
face
challenges
in
accurately
diagnosing
SOH
complex
application
scenarios.
Hence,
we
developed
deep
framework
for
estimation
without
prior
knowledge
capacity.
Our
incorporates
series
neural
networks
(DNNs)
that
utilize
direct
current
internal
resistance
(DCIR)
feature
to
estimate
SOH.
The
correlation
DCIR
with
fade
capacity
quantified
as
strong
under
various
conditions
using
Pearson
coefficients.
We
K-fold
cross-validation
method
select
hyperparameters
DNN
optimal
hyperparameter
compared
significant
advantages
reliable
prediction
accuracies.
proposed
algorithm
subjected
robustness
validation,
experimental
results
demonstrate
model
achieves
precision,
mean
absolute
error
(MAE)
less
than
0.768%
root
square
(RMSE)
1.185%,
even
when
LIBs
are
varying
study
highlights
superiority
reliability
combining
DNNs
features
estimation.
Reliability Engineering & System Safety,
Journal Year:
2023,
Volume and Issue:
241, P. 109603 - 109603
Published: Aug. 29, 2023
Predictive
health
assessment
is
of
vital
importance
for
smarter
battery
management
to
ensure
optimal
and
safe
operations
thus
make
the
most
use
life.
This
paper
proposes
a
general
framework
aging
prognostics
in
order
provide
predictions
knee,
lifetime,
state
degradation,
rate
variations,
as
well
health.
Early
information
used
predict
knee
slope
other
life-related
via
deep
multi-task
learning,
where
convolutional-long-short-term
memory-bayesian
neural
network
proposed.
The
structure
also
online
degradation
detection
accelerating
aging.
two
probabilistic
predicted
boundaries
identify
regions
assessment.
To
avoid
wrong
premature
alarms,
empirical
model
data
preprocessing
together
with
learning.
A
cloud-edge
considered
fine-tuning
adopted
performance
improvement
during
cycling.
proposed
flexible
adjustment
different
practical
requirements
can
be
extrapolated
batteries
aged
under
conditions.
results
indicate
that
early
are
improved
using
method
compared
multiple
single
feature-based
benchmarks,
integration
algorithm
improved.
sequence
prediction
reliable
lengths
root
mean
square
errors
less
than
1.41%,
guide
predictive
management.
IISE Transactions,
Journal Year:
2024,
Volume and Issue:
unknown, P. 1 - 17
Published: Aug. 7, 2024
Traditionally,
Gaussian
assumption,
implied
by
the
Wiener
process,
is
widely
admitted
for
modeling
degradation
processes.
However,
when
data
exhibit
heavy
tails,
this
assumption
not
suitable.
To
overcome
limitation,
article
proposes
a
novel
class
of
tail-weighted
multivariate
model,
which
built
upon
Student-t
process.
The
model
able
to
account
both
between-unit
variability
and
process
dependency,
while
allows
adjusting
tail
heaviness
through
tuning
parameter
degree
freedom.
For
reliability
assessment,
we
derive
system
function
present
an
efficient
Monte
Carlo
method
its
evaluation.
Further,
introduce
expectation-maximization
algorithm
estimation
design
bootstrap
interval
estimation.
Comprehensive
simulation
studies
are
conducted
validate
effectiveness
inference
method.
Finally,
proposed
methodology
applied
analyze
two
real-world
datasets.
Sustainable Energy Technologies and Assessments,
Journal Year:
2023,
Volume and Issue:
60, P. 103457 - 103457
Published: Sept. 17, 2023
Lithium-ion
battery
has
presented
a
rapid
growth
as
the
power
source
of
electric
vehicles
(EVs).
The
state
health
(SOH)
estimation
plays
an
important
role
in
ensuring
safe
operation
system.
Currently,
model-based
and
data-driven
methods
have
been
comprehensively
reviewed
by
considering
strengths
drawbacks.
However,
these
approaches
present
high
complexity
due
to
complex
test,
modelling
processing
algorithm.
Developing
SOH
based
on
simple
test
can
help
suppress
cost
improve
efficiency.
there
is
no
work
review
development
for
EV
batteries
traditional
classification
needs
be
updated
align
with
current
research.
This
paper
reviews
discusses
state-of-the-art
techniques
over
past
decade.
Particularly,
it
gives
reclassifications
working
principles
techniques.
Moreover,
their
advantages
disadvantages
when
applied
practice
are
discussed
incorporating
experimental
studies.
Eventually,
this
meaningful
suggestions
both
practical
applications
future
methods.
It
considered
that
suggest
valuable
guidance
academic
investigation
engineering
applications.
IEEE Access,
Journal Year:
2024,
Volume and Issue:
12, P. 43984 - 43999
Published: Jan. 1, 2024
This
paper
presents
a
comprehensive
survey
of
machine
learning,
deep
and
digital
twin
technology
methods
for
predicting
managing
the
battery
state
health
in
electric
vehicles.
Battery
estimation
is
essential
optimizing
usage,
performance,
safety,
cost-effectiveness
Estimating
complex
undertaking
due
to
its
dependency
on
multiple
factors.
These
factors
include
characteristics
such
as
type,
chemistry,
size,
temperature,
current,
voltage,
impedance,
cycle
number,
driving
pattern.
There
are
drawbacks
traditional
methods,
experimental
model-based
approaches,
terms
accuracy,
complexity,
expense,
viability
real-time
applications.
By
employing
variety
algorithms
discover
nonlinear
dynamic
link
between
parameters
health,
data-driven
techniques
like
technologies
can
get
beyond
these
restrictions.
Data-driven
also
incorporate
physics
domain
knowledge
improve
explainability
interpretability
results.
reviews
latest
advancements
challenges
using
management
The
discusses
future
directions
opportunities
further
research
development
this
field.
scope
spans
publications
from
year
2021
2023.
