International Journal of Advanced Computer Science and Applications,
Journal Year:
2023,
Volume and Issue:
14(9)
Published: Jan. 1, 2023
End-Edge-Cloud
Computing
(EECC)
has
been
applied
in
many
fields,
due
to
the
increased
popularity
of
smart
devices.
But
cooperation
end
devices,
edge
and
cloud
resources
is
still
challenge
for
improving
service
quality
resource
efficiency
EECC.
In
this
paper,
we
focus
on
task
offloading
address
challenge.
We
formulate
problem
as
mixed
integer
nonlinear
programming,
solve
it
by
Genetic
Algorithm
(GA).
GA-based
algorithm,
each
chromosome
code
a
solution,
evolution
iteratively
search
global
best
solution.
To
improve
performance
offloading,
integrate
two
improvement
schemes
into
which
are
replacement
rescheduling,
respectively.
The
replace
every
individual
its
better
offspring
after
crossing,
substitutes
selection
operator
population
evolution.
rescheduling
rejected
available
resources,
given
solution
from
chromosome.
Extensive
experiments
conducted,
results
show
that
our
proposed
algorithm
can
upto
32%
user
satisfaction,
12%
efficiency,
35.3%
processing
compared
with
nine
classical
up-to-date
algorithms.
Research Square (Research Square),
Journal Year:
2024,
Volume and Issue:
unknown
Published: Feb. 2, 2024
Abstract
Digital
twins
have
attracted
more
and
attention
in
the
past
few
years.
To
put
digital
into
practice,
a
large
number
of
modeling
approaches
been
proposed,
vast
amounts
data
collected,
their
accuracy
has
improving.
However,
current
research
paid
insufficient
to
multi-scale
features
shop
floor,
which
hinders
effective
application
twin
floor.
address
problem
how
achieve
multi-level
multi-dimensional
fusion
models
with
production
process
data,
this
paper
first
proposes
structured
framework
for
sorting
out
all
collected
real-time;
then
supporting
real-time
from
unit
level
system
level.
The
method
judges
parsed
received
streams
through
full-factor
semanticization
framework,
at
same
time
fuses
constructed
model
multiple
dimensions
layers,
forming
as
blood
skeleton.
Finally,
micro-assembly-based
environment
is
selected
case
study
verify
correctness
feasibility
proposed
grooming
method.
Journal of Physics Conference Series,
Journal Year:
2024,
Volume and Issue:
2781(1), P. 012004 - 012004
Published: June 1, 2024
Abstract
As
a
huge
and
complex
system,
the
power
grid
involves
multiple
levels
various
interconnected
components,
making
it
difficult
to
monitor
operational
risks
in
real-time.
Therefore,
this
study
proposes
real-time
risk
monitoring
method
operation
based
on
cloud-edge
collaboration
technology.
Through
cloud
edge
technology,
data
processing
results
are
designed
an
extreme
gradient
boosting
clustering
(XGBoost)
algorithm
is
used
complete
clustering.
The
level
of
calculated
completed.
experimental
indicate
that
application
process
research
has
shorter
delay
higher
accuracy.
