Energy storage materials,
Год журнала:
2023,
Номер
63, С. 102927 - 102927
Опубликована: Авг. 17, 2023
The
lithium-ion
battery
(LIB)
field
is
moving
towards
the
direction
of
investigating
spatially
resolved
physical
phenomena
in
3D
porous
microstructure
electrodes.
These
pore-scale
simulations
give
new
insights
into
local
dynamics
lithiation/de-lithiation
and
charge
transport.
Nevertheless,
computational
time
these
limits
integration
models
optimization
workflows
cycling
conditions
or
electrode
manufacturing
processes.
Machine
learning
present
a
way
assessing
real-time
performance
materials.
While
several
successful
techniques
for
replicating
with
machine
have
been
proposed,
this
case
study
presents
more
demanding
problem,
due
to
necessity
understanding
behavior
heterogeneous
data,
as
it
evolves
time:
poses
both
scientific
technical
challenge.
To
end,
we
propose
an
autoregressive
multiscale
convolutional
neural
network
model
predict
relevant
quantities
at
solid
phase:
lithium
concentration
(in
active
material)
potential
material
carbon
binder).
are
ultimately
used
reconstruct
discharge
curve.
images
microstructures
input
network,
trained
dataset
finite
element
method
cathode
side
ion
batteries.
We
proof-of-concept
applicability
networks
time-dependent
physics
problems.
exhibits
very
high
accuracy
(with
errors
lower
than
2%)
forecasting
unseen
cathodes.
Nature Communications,
Год журнала:
2023,
Номер
14(1)
Опубликована: Фев. 14, 2023
Abstract
Proton
exchange
membrane
fuel
cells,
consuming
hydrogen
and
oxygen
to
generate
clean
electricity
water,
suffer
acute
liquid
water
challenges.
Accurate
modelling
is
inherently
challenging
due
the
multi-phase,
multi-component,
reactive
dynamics
within
multi-scale,
multi-layered
porous
media.
In
addition,
currently
inadequate
imaging
capabilities
are
limiting
simulations
small
areas
(<1
mm
2
)
or
simplified
architectures.
Herein,
an
advancement
in
achieved
using
X-ray
micro-computed
tomography,
deep
learned
super-resolution,
multi-label
segmentation,
direct
multi-phase
simulation.
The
resulting
image
most
resolved
domain
(16
with
700
nm
voxel
resolution)
largest
flow
simulation
of
a
cell.
This
generalisable
approach
unveils
multi-scale
clustering
transport
mechanisms
over
large
dry
flooded
gas
diffusion
layer
fields,
paving
way
for
next
generation
proton
cells
optimised
structures
wettabilities.
npj Materials Sustainability,
Год журнала:
2025,
Номер
3(1)
Опубликована: Янв. 7, 2025
Exploring
sustainable
alternative
constituents
is
a
key
pathway
to
carbon-neutralization
of
concrete,
but
often
limited
insufficient
understandings
how
they
interact
with
conventional
concrete
components
at
microscale.
In
this
paper
we
reviewed
the
most
cutting-edge
microprobes
used
for
such
purposes,
from
both
laboratory
setup
synchrotron
radiation-based
techniques.
We
also
provided
practical
guidelines
on
sample
preparation
and
result
analysis,
which
could
benefit
researchers
who
plan
adopt
these
methods
Environmental Earth Sciences,
Год журнала:
2022,
Номер
81(3)
Опубликована: Янв. 25, 2022
Abstract
Image
segmentation
remains
the
most
critical
step
in
Digital
Rock
Physics
(DRP)
workflows,
affecting
analysis
of
physical
rock
properties.
Conventional
techniques
struggle
with
numerous
image
artifacts
and
user
bias,
which
lead
to
considerable
uncertainty.
This
study
evaluates
advantages
using
random
forest
(RF)
algorithm
for
fractured
rocks.
The
quality
is
discussed
compared
two
conventional
processing
methods
(thresholding-based
watershed
algorithm)
an
encoder–decoder
network
form
convolutional
neural
networks
(CNNs).
segmented
images
RF
method
were
used
as
ground
truth
CNN
training.
samples
are
acquired
by
X-ray
computed
tomography
scanning
(XCT).
skeletonized
3D
calculated,
providing
information
about
mean
mechanical
aperture
roughness.
porosity,
permeability,
flow
fields,
preferred
paths
analyzed
DRP
approach.
Moreover,
breakthrough
curves
obtained
from
tracer
injection
experiments
evaluate
each
method.
results
show
that
overestimate
fracture
aperture.
Both
machine
learning
approaches
promising
handle
all
complexities
without
any
prior
CT-image
filtering.
However,
implementation
has
superior
inherent
over
CNN.
resource-saving
(e.g.,
quickly
trained),
does
not
need
extensive
training
dataset,
can
provide
uncertainty
a
measure
evaluating
quality.
variation
properties
highlights
importance
choosing
appropriate
ADVANCES IN GEO-ENERGY RESEARCH,
Год журнала:
2023,
Номер
8(1), С. 5 - 18
Опубликована: Фев. 2, 2023
Digital
rock
technology
is
becoming
essential
in
reservoir
engineering
and
petrophysics.
Three-dimensional
digital
reconstruction,
image
resolution
enhancement,
segmentation,
parameters
prediction
are
all
crucial
steps
enabling
the
overall
analysis
of
rocks
to
overcome
shortcomings
limitations
traditional
methods.
Artificial
intelligence
technology,
which
has
started
play
a
significant
role
many
different
fields,
may
provide
new
direction
for
development
technology.
This
work
presents
systematic
review
deep
learning
methods
that
being
applied
tasks
within
analysis,
including
reconstruction
rocks,
high-resolution
acquisition,
grayscale
parameter
prediction.
The
results
these
applications
prove
state-of-the-art
can
help
advance
approach
scientific
knowledge
field
rocks.
also
discusses
future
research
developments
on
application
Cited
as:
Li,
X.,
B.,
Liu,
F.,
T.,
Nie,
X.
Advances
Geo-Energy
Research,
2023,
8(1):
5-18.
https://doi.org/10.46690/ager.2023.04.02
Energy & Fuels,
Год журнала:
2023,
Номер
37(4), С. 2475 - 2497
Опубликована: Янв. 25, 2023
The
complex
and
multiscale
nature
of
shale
gas
transport
imposes
new
challenges
to
the
already
well-developed
techniques
for
conventional
reservoirs,
especially
digital
core
analysis.
Multiscale
complicated
pore
systems
distinctive
properties
limit
most
reconstruction
methods
not
applicable.
High-precision
imaging
experiments
play
a
key
role
in
characterization
structures
mineral
components.
While
exhilarating
breakthroughs
physical
experimental
hybrid
superposition
have
made
significant
achievements
rock
reconstruction,
rapidly
evolving
deep
learning
also
present
promising
option.
Benefiting
from
techniques,
pore-scale
flow
can
be
directly
simulated
based
on
or
indirectly
modeled
using
network
model.
It
is
precise
realistic
investigate
at
scale
considering
desorption,
surface
diffusion,
slippage
nanopores.
In
this
paper,
we
reviewed
recent
advances
off-mentioned
processes
presented
hand
research
field.