Machine Learning-Based Impact of Rotational Speed on Mixing, Mass Transfer, and Flow Parameter Prediction in Solid–Liquid Stirred Tanks
Processes,
Год журнала:
2025,
Номер
13(5), С. 1423 - 1423
Опубликована: Май 7, 2025
To
address
the
issues
of
low
solid–liquid
mixing
and
mass
transfer
efficiency
difficult
real-time
regulation
in
resource
utilization
non-ferrous
metal
smelting
slag,
this
study
constructs
a
research
framework
integrating
Computational
Fluid
Dynamics-Discrete
Element
Method
(CFD-DEM)
coupling
models
machine
learning.
The
systematically
investigates
particle
motion
characteristics
laws
stirred
tanks
enables
an
intelligent
prediction
key
parameters.
Through
CFD-DEM
two-way
simulation,
quantifies
dispersion
using
relative
standard
deviation
(RSD)
calculates
coefficient
(k)
based
on
Hughmark
model,
revealing
effects
size
impeller
speed
efficiency.
For
parameter
prediction,
data
are
used
to
train
multi-model
library,
with
model
performance
evaluated
through
comparative
experiments.
results
show
that
increasing
rotational
shortens
time,
reduces
RSD
values
by
25–40%,
increases
force,
decreases
stability
during
circulation
phase.
Different
learning
(ML)
algorithms
exhibit
varying
performances
time-series
coefficients.
Notably,
GA-BP
achieves
fitting
degree
R
0.99
both
predictions,
meeting
requirements
for
structural
optimization
tanks.
This
provides
theoretical
support
technical
pathways
control
tanks,
offering
engineering
application
value
fields
such
as
hydrometallurgy
solid
waste
utilization.
Язык: Английский
Numerical simulation of particle distribution characteristics and spatiotemporal disorder under the influence of various modified pitched blade turbine impellers
Physics of Fluids,
Год журнала:
2025,
Номер
37(6)
Опубликована: Июнь 1, 2025
In
the
process
of
solid–liquid
mixing,
pitched
blade
turbine
(PBT)
is
easy
to
cause
a
stable
structure
particle
flow,
forming
pseudo-uniform
state,
and
reducing
mixing
efficiency.
this
study,
morphology
PBT
impeller
was
improved,
three
new
impellers
were
designed:
cut
diversion
1
(CDI-1),
2
(CDI-2),
serrated
impeller.
The
discrete
element
method–volume
fluid
coupling
method
used
numerically
simulate
distribution
flow
field
characteristics
under
action
An
evaluation
average
deviation
range
(DRav)
proposed
quantify
disorder
spatial
temporal
distribution.
results
show
that
suspension
ratio
increased
by
13.00%,
9.00%,
18.00%,
respectively,
values
DRav
are
0.001,
0.0013,
0.015
m,
which
effectively
breaks
steady-state
in
stirring.
Язык: Английский