A
novel
real-time
approximated
MPC
control
policy
based
on
deep
learning
is
proposed
to
address
the
high
computational
burden
of
model
predictive
(MPC)
for
large-scale
systems
and
those
with
fast
dynamics.
This
method
approximates
optimal
solution
distributed
optimization
problems
in
ALADIN-based
parallel
design
framework,
resulting
a
highly
effective
approach
that
outperforms
other
well-known
methods
solving
problem.
The
numerical
case
study
shows
promising
results,
demonstrating
potential
this
implementation.
Processes,
Год журнала:
2024,
Номер
12(6), С. 1140 - 1140
Опубликована: Май 31, 2024
For
mitigating
global
warming,
polymer
electrolyte
membrane
fuel
cells
have
become
promising,
clean,
and
sustainable
alternatives
to
existing
energy
sources.
To
increase
the
density
efficiency
of
(PEMFC),
a
comprehensive
numerical
modeling
approach
that
can
adequately
predict
multiphysics
performance
relative
actual
test
such
as
an
acceptable
depiction
electrochemistry,
mass/species
transfer,
thermal
management,
water
generation/transportation
is
required.
However,
models
suffer
from
reliability
issues
due
their
dependency
on
several
assumptions
made
for
sake
simplification,
well
poor
choices
approximations
in
material
characterization
electrochemical
parameters.
In
this
regard,
data-driven
machine
learning
could
provide
missing
more
appropriate
parameters
conventional
computational
fluid
dynamics
models.
The
purpose
present
overview
explore
state
art
individual
components
PEMFC,
limitations,
how
they
be
significantly
improved
by
hybrid
techniques
integrating
with
approaches.
Furthermore,
detailed
future
direction
proposed
solution
related
PEMFC
its
impact
transportation
sector
discussed.
