The
progression
of
the
Internet
Vehicles
(IoVs)
has
cultivated
a
comprehensive
information
environment,
laying
groundwork
for
vehicle-environment
adaptive
control.
Region-based
traffic
condition
prediction,
one
methods
macroscopic
state
analyses,
provides
insights
into
overall
trends
entire
road
network,
invaluable
urban
applications
and
fostering
cooperation.
In
this
study,
an
energy
management
strategy
based
on
region-based
grade
prediction
(TGP-EMS)
is
proposed
plug-in
hybrid
electric
vehicles
(PHEVs)
IoVs,
strengthening
adaptability
via
accurately
obtaining
future
conditions.
Firstly,
data
collected
from
volunteering
are
processed
to
generate
representative
graphs
conditions
separated
several
grades
with
distinct
attributes.
Secondly,
differentiable
pooling
integrated
hierarchical
deep
learning
framework
establish
model
termed
Graph
Pool.
Thirdly,
optimal
explicit
solving
method
instantaneous
optimization
algorithm
successfully
applied
management.
Moreover,
robustness
introduced
optimized
beetle
antennae
search
(BAS)
algorithm.
Simulation
results,
accompanied
by
hardware-in-the-loop
(HIL)
tests,
suggest
that
Pool
effectively
captures
spatial
features
across
ensuring
accuracy
consistency
in
predicting
TGP-EMS
adeptly
adjusts
power
distribution
these
predictions,
showing
improvement
roughly
13.5%
compared
conventional
rule-based
strategies.
Energies,
Год журнала:
2024,
Номер
17(17), С. 4364 - 4364
Опубликована: Авг. 31, 2024
Electric
vehicles
(EVs)
are
pivotal
in
addressing
the
escalating
environmental
crisis.
While
EV
drivetrains
excel
compared
to
those
of
with
internal
combustion
engines
(ICEs),
their
energy
storage
systems
hampered
by
limited
range,
lifespan,
and
lengthy
charging
times.
Hybrid
(HESSs)
present
a
viable
current
solution
these
issues.
This
review
thoroughly
explores
state
art
emerging
field
multisource
EVs
that
utilize
HESSs,
incorporating
any
combination
batteries
(BTs),
supercapacitors
(SCs),
flywheels
(FWs),
fuel
cells
(FCs),
and/or
transmotors.
In
addition,
paper
systematically
categorizes
evaluates
different
hybrid
configurations,
detailing
potential
topologies
respective
advantages
limitations.
Moreover,
examines
diverse
control
algorithms
used
manage
complex
systems,
focusing
on
effectiveness
operational
efficiency.
By
identifying
research
gaps
technological
challenges,
this
study
aims
delineate
future
directions
could
enhance
deployment
optimization
EVs,
thereby
critical
challenges
such
as
density,
system
reliability,
cost-effectiveness.
The
progression
of
the
Internet
Vehicles
(IoVs)
has
cultivated
a
comprehensive
information
environment,
laying
groundwork
for
vehicle-environment
adaptive
control.
Region-based
traffic
condition
prediction,
one
methods
macroscopic
state
analyses,
provides
insights
into
overall
trends
entire
road
network,
invaluable
urban
applications
and
fostering
cooperation.
In
this
study,
an
energy
management
strategy
based
on
region-based
grade
prediction
(TGP-EMS)
is
proposed
plug-in
hybrid
electric
vehicles
(PHEVs)
IoVs,
strengthening
adaptability
via
accurately
obtaining
future
conditions.
Firstly,
data
collected
from
volunteering
are
processed
to
generate
representative
graphs
conditions
separated
several
grades
with
distinct
attributes.
Secondly,
differentiable
pooling
integrated
hierarchical
deep
learning
framework
establish
model
termed
Graph
Pool.
Thirdly,
optimal
explicit
solving
method
instantaneous
optimization
algorithm
successfully
applied
management.
Moreover,
robustness
introduced
optimized
beetle
antennae
search
(BAS)
algorithm.
Simulation
results,
accompanied
by
hardware-in-the-loop
(HIL)
tests,
suggest
that
Pool
effectively
captures
spatial
features
across
ensuring
accuracy
consistency
in
predicting
TGP-EMS
adeptly
adjusts
power
distribution
these
predictions,
showing
improvement
roughly
13.5%
compared
conventional
rule-based
strategies.
