A Comprehensive Review of Multiple Physical and Data-Driven Model Fusion Methods for Accurate Lithium-Ion Battery Inner State Factor Estimation
Junjie Tao,
No information about this author
Shunli Wang,
No information about this author
Wen Cao
No information about this author
et al.
Batteries,
Journal Year:
2024,
Volume and Issue:
10(12), P. 442 - 442
Published: Dec. 13, 2024
With
the
rapid
global
growth
in
demand
for
renewable
energy,
traditional
energy
structure
is
accelerating
its
transition
to
low-carbon,
clean
energy.
Lithium-ion
batteries,
due
their
high
density,
long
cycle
life,
and
efficiency,
have
become
a
core
technology
driving
this
transformation.
In
lithium-ion
battery
storage
systems,
precise
state
estimation,
such
as
of
charge,
health,
power,
crucial
ensuring
system
safety,
extending
lifespan,
improving
efficiency.
Although
physics-based
estimation
techniques
matured,
challenges
remain
regarding
accuracy
robustness
complex
environments.
advancement
hardware
computational
capabilities,
data-driven
algorithms
are
increasingly
applied
management,
multi-model
fusion
approaches
emerged
research
hotspot.
This
paper
reviews
application
between
models
critically
analyzes
advantages,
limitations,
applicability
models,
evaluates
effectiveness
robustness.
Furthermore,
discusses
future
directions
improvement
model
adaptability,
performance
under
operating
conditions,
aiming
provide
theoretical
support
practical
guidance
developing
management
technologies.
Language: Английский
User Behavior Analysis and Prediction Model Construction in Higher Education Management Information Systems
Shanshan Yu,
No information about this author
Lihua Zhong,
No information about this author
Yuwang Liu
No information about this author
et al.
Applied Mathematics and Nonlinear Sciences,
Journal Year:
2024,
Volume and Issue:
9(1)
Published: Jan. 1, 2024
Abstract
The
education
management
system
is
an
important
tool
for
universities
to
manage
academic
information
and
of
staff
students.
article
constructs
a
user
behavior
analysis
model
based
on
machine
learning
prediction
LR-XGBoost
analyze
predict
in
the
educational
colleges
universities.
Specifically,
after
collection
data
university
system,
descriptive
statistics
cluster
are
performed.
Then,
relationship
between
dropout/leaving
explored
by
model,
performance
tested
comparing
accuracy
with
other
models.
Cluster
3
users
had
significantly
higher
means
12
behaviors
college
than
those
clusters
2
1.
Among
behaviors,
only
closing
web
pages
p-value
greater
0.05
did
not
pass
z-test.
best
95.51%
94.53%
behavioral
anomalous
prediction,
respectively,
F1
values
92.34%
95.33%,
respectively.
It
has
clear
overall
advantage
predicting
next
1-10
days.
Language: Английский
Pattern Identification of Community Engagement Behaviors in a Big Data Environment and Its Impact on Community Health Development
Jingyu Luo,
No information about this author
Zheng Li,
No information about this author
Jianwei Liu
No information about this author
et al.
Applied Mathematics and Nonlinear Sciences,
Journal Year:
2024,
Volume and Issue:
9(1)
Published: Jan. 1, 2024
Abstract
Improving
residents’
community
participation
capacity
in
the
big
data
environment
helps
to
realize
common
action
and
resource
sharing
among
subjects
promotes
healthy
development
of
community.
This
paper
takes
theory
planned
behavior
ladder
civic
as
guide,
designs
distributes
questionnaires
related
patterns
obtain
research
data,
identifies
using
K-Means
clustering
algorithm
introduces
PROMETHEE
Ⅱ
method
measure
patterns.
The
impact
model
health
based
on
structural
equations
was
constructed
analyze
degree
development.
There
are
six
categories
behavioral
patterns,
overall
net
flow
complete
is
highest
at
0.047.
coefficient
equation
0.149,
test
result
has
a
significant
1%
level.
Different
behaviors
can
promote
help
improve
governance.
Language: Английский