International Journal of Renewable Energy Development,
Год журнала:
2023,
Номер
12(5), С. 923 - 929
Опубликована: Авг. 16, 2023
Fluid
catalytic
cracking
could
convert
crude
palm
oil
into
valuable
green
fuels
to
substitute
fossil
fuels.
This
study
aimed
predict
the
phenomenon
and
yield
in
industrial
fluid
riser
using
computational
dynamics.
A
three-dimensional
transient
simulation
Eulerian-Lagrangian
with
multiphase
particle-in-cell
is
investigate
reactive
gas-particle
hydrodynamics
four-lump
kinetic
network
model
rare
earth-Y
catalyst
for
behaviors.
The
results
show
that
velocity
profile
increase
middle
of
reactor
because
reaction
process
produces
OLP
Gas
products
has
a
lighter
molecular
weight.
endothermic
causes
temperature
decrease
heat
comes
from
catalyst.
analysis
shows
accurately
predicts
fuel
oil.
As
result,
conversion,
organic
liquid
product
yield,
correspond
70
wt%,
28.8
27.5
respectively.
Compared
experimental
study,
prediction
showed
good
agreement
determined
optimal
dimension.
methodology
are
guidelines
optimizing
FCC
CPO.
Petroleum Science,
Год журнала:
2024,
Номер
21(4), С. 2849 - 2869
Опубликована: Янв. 22, 2024
Since
chemical
processes
are
highly
non-linear
and
multiscale,
it
is
vital
to
deeply
mine
the
multiscale
coupling
relationships
embedded
in
massive
process
data
for
prediction
anomaly
tracing
of
crucial
parameters
production
indicators.
While
integrated
method
adaptive
signal
decomposition
combined
with
time
series
models
could
effectively
predict
variables,
does
have
limitations
capturing
high-frequency
detail
operation
state
when
applied
complex
processes.
In
light
this,
a
novel
Multiscale
Multi-radius
Multi-step
Convolutional
Neural
Network
(MsrtNet)
proposed
mining
spatiotemporal
information.
First,
industrial
from
Fluid
Catalytic
Cracking
(FCC)
using
Complete
Ensemble
Empirical
Mode
Decomposition
Adaptive
Noise
(CEEMDAN)
extract
multi-energy
scale
information
feature
subset.
Then,
convolution
kernels
varying
stride
padding
structures
established
decouple
long-period
encapsulated
within
data.
Finally,
reconciliation
network
trained
reconstruct
results
obtain
final
output.
MsrtNet
initially
assessed
its
capability
untangle
among
variables
Tennessee
Eastman
(TEP).
Subsequently,
performance
evaluated
predicting
product
yield
2.80
×
106
t/a
FCC
unit,
taking
diesel
gasoline
as
examples.
conclusion,
can
achieve
maximum
reduction
11%
error
compared
other
time-series
models.
Furthermore,
robustness
transferability
underscore
promising
potential
broader
applications.
Processes,
Год журнала:
2025,
Номер
13(2), С. 464 - 464
Опубликована: Фев. 8, 2025
Catalyst
loss
is
a
typical
fault
that
impacts
the
long-term
operation
of
fluidized
catalytic
cracking
(FCC)
in
oil
refining
process.
The
FCC
disengager
critical
place
for
separating
catalyst
from
gas.
A
fast
and
precise
fault-cause
judgment
vital
avoiding
failures.
In
this
study,
novel
method
failures
with
quantitative
criteria
was
established
via
tree
analysis
(FTA)
method,
based
on
relationship
model
between
flow
field
signals
faults
investigated
by
computational
fluid
dynamics
(CFD).
FTA
defines
three
intermediate
events:
fragmentation,
process
mechanical
fault.
CFD
results,
it
found
detailed
reason
can
be
inferred
changes
characteristic
parameters
within
disengager.
For
example,
when
rate
may
rapidly
increase
factor
around
200.
Furthermore,
pressure
drop
cyclone
separator
decreases
35%,
which
indicates
dipleg
has
fractured.
new
been
applied
cases
two
industrial
disengagers.
accurately
pinpointed
sudden
reduction
inlet
velocity
blockage
at
as
main
factors
leading
to
faults,
respectively.
results
are
consistent
actual
reasons,
demonstrating
reliability
method.
This
study
could
contribute
providing
theoretical
support
enhancing
accuracy
diagnosis
thereby
ensuring
safe
stable
unit.
Processes,
Год журнала:
2023,
Номер
12(1), С. 61 - 61
Опубликована: Дек. 27, 2023
This
study
reports
a
novel
hybrid
model
for
the
prediction
of
six
critical
process
variables
importance
in
an
industrial-scale
FCC
(fluid
catalytic
cracking)
riser
reactor:
vacuum
gas
oil
(VGO)
conversion,
outlet
temperature,
light
cycle
(LCO),
gasoline,
gases,
and
coke
yields.
The
proposed
is
developed
via
integration
computational
particle-fluid
dynamics
(CPFD)
methodology
with
artificial
intelligence
(AI).
adopted
solves
first
principle
(FPM)
equations
numerically
using
CPFD
Barracuda
Virtual
Reactor
22.0®
software.
Based
on
216
these
simulations,
performance
reactor
unit
was
assessed
VGO
cracking
kinetics
at
CREC-UWO.
dataset
obtained
employed
training
testing
machine
learning
(ML)
algorithm.
algorithm
based
multiple
output
feedforward
neural
network
(FNN)
selected
to
allow
one
establish
correlations
between
feeding
conditions
its
outcoming
parameters,
0.83
averaged
regression
coefficient
overall
RMSE
1.93
being
obtained.
research
underscores
value
integrating
simulations
ML
optimize
industrial
processes
enhance
their
predictive
accuracy,
offering
significant
advancements
operations.
Industrial & Engineering Chemistry Research,
Год журнала:
2023,
Номер
62(51), С. 22005 - 22015
Опубликована: Дек. 12, 2023
An
ever-increasing
complexity
of
the
models
catalytic
cracking
(which
is
rarely
justified
by
available
monitoring
data)
makes
"practical"
modeling
for
industrial
units
challenging.
In
this
work,
we
develop
a
simple
numerical
model
an
riser,
with
phenomenological
parameters
that
are
determined
from
data.
The
based
on
four-lump
reaction
scheme
(with
coke
being
one
lumps),
kinetic
reactor
zeolite-containing
catalyst
used
production
wet
gases.
expression
rate
reflects
reactions
in
gaseous
phase
occur
presence
solid
(catalyst)
phase.
Hydrodynamics
nonisothermal
reactive
gas–solid
mixtures
captured
two-fluid
model.
strong
turbulence
nature
flow
it
possible
to
disregard
theory-based
values
viscous,
diffusion,
and
thermal
conductivity
coefficients
(additionally
reducing
number
needed
empirical
coefficients).
resultant
applied
in-depth
analysis
fields
reactor,
demonstrating
good
agreement
results
more
sophisticated
approaches.
also
calculation
optimal
intakes
water
vapor
catalyst.
