This
paper
aims
to
design
three
control
solutions
for
nonlinear
processes
with
Shape
Memory
Alloy
(SMA)
wire
actuators
viewed
as
controlled
dead
time.
The
proposed
are
the
PI
controller
Smith
predictor
configuration,
sliding
mode
configuration
and
super
twisting
configuration.
parameters
of
controllers
optimally
tuned
using
a
Grey
Wolf
optimizer
algorithm.
An
accurate
evolved
Takagi-Sugeno-Kang
fuzzy
model
SMA
is
used
validate
structures
by
simulations
their
performance
compared
evaluated.
Romanian Journal of Information Science and Technology,
Journal Year:
2023,
Volume and Issue:
2023(3-4), P. 278 - 288
Published: Sept. 28, 2023
This
paper,
dedicated
to
Acad.
Florin
Gheorghe
Filip,
at
his
15th
anniversary
as
the
Chairman
of
Information
Science
and
Technology
Section
Romanian
Academy,
suggests
an
approach
describe
data
obtained
in
biomonitoring
studies
using
mosses,
on
example
Republic
Moldova.
In
total,
33
moss
samples
were
collected
territory
Moldova,
elemental
composition
which
was
determined
by
neutron
activation
analysis
atomic
absorption
spectrometry.
At
first
stage
work,
a
correlation
carried
out
with
ranking
order
decreasing
total
correlation,
made
it
possible
preliminarily
reduce
number
elements
two
factors.
sorting,
iron,
content
environment
can
be
associated
anthropogenic
activity,
chosen
element
determining
rank.
The
next
work
smoothing
discrete
cosine
transform,
for
codes
rewritten
algorithm
ported
Excel-VBA
environment,
is
most
suitable
preprocessing
graphical
display
experimental
data.
Also,
developed
linearly
independent
(basis)
vectors
matrix
itself
decomposed.
Two
principal
roots
identified,
larger
one
several
thousand
units,
three
lower
roots,
are
less
than
unit
excluded,
since
their
absolute
values
differ
or
more
orders
magnitude
downward
from
eigenvalues
influencing
components.
paper
75th
anniversary.
Data in Brief,
Journal Year:
2023,
Volume and Issue:
49, P. 109309 - 109309
Published: June 11, 2023
In
this
article,
a
benchmark
for
real-world
bin
packing
problems
is
proposed.
This
dataset
consists
of
12
instances
varying
levels
complexity
regarding
size
(with
the
number
packages
ranging
from
38
to
53)
and
user-defined
requirements.
fact,
several
real-world-oriented
restrictions
were
taken
into
account
build
these
instances:
PeerJ Computer Science,
Journal Year:
2023,
Volume and Issue:
9, P. e1613 - e1613
Published: Oct. 9, 2023
In
the
field
of
e-commerce
warehousing,
maximizing
utilization
packing
bins
is
a
fundamental
goal
for
all
major
logistics
enterprises.
However,
determining
appropriate
size
poses
practical
challenge
many
companies.
Given
limited
research
on
open-size
3D
bin
problem
as
well
high
complexity
and
lengthy
computation
time
existing
models,
this
study
focuses
optimizing
multiple-bin
sizes
within
context.
Building
upon
research,
we
propose
hybrid
integer
programming
model,
denoted
three
dimensional
multiple
option
rectangular
(3D-MODRPP),
to
address
problem.
Additionally,
leverage
well-established
hardware
software
technologies
system
capable
accommodating
types
with
open
dimensions.
To
reduce
model
number
constraints,
introduce
novel
assumption
method
0–1
variables
in
overlap
rotation
constraints.
By
applying
approach,
significantly
streamline
computational
associated
calculations.
Furthermore,
refine
dataset
by
developing
customized
version
based
classical
Three-Dimensional
One-Size
Dependent
Rectangular
Packing
Problem
(3D-ODRPP)
dataset,
leading
improved
outcomes.
Through
comprehensive
analysis
results,
our
exhibits
remarkable
advancements
addressing
strong
heterogeneous
problem,
weak
actual
sizes.
Specifically,
it
reduces
increases
space
utilization.
The
designed
paves
way
applications
proposed
providing
researchers
broader
prospects
directions
expand
scope
investigation
packing.
Consequently,
contributes
solving
complex
problems,
reducing
waste,
enhancing
transportation
efficiency.
Romanian Journal of Information Science and Technology,
Journal Year:
2024,
Volume and Issue:
27(3-4), P. 310 - 322
Published: Sept. 30, 2024
Based
on
the
assumption
that
radionuclide
composition
of
gas-aerosol
emissions
from
nuclear
power
plants
(NPPs)
into
environment
is
determined
by
fuel
in
reactor
(but
not
identical
to
it),
an
analysis
probability
occurrence
various
isotopes
possible
conducted
this
paper.
It
confirmed
activity
fuel,
and
consequently
emissions,
are
uranium
fission
fragments,
proving
decay
rate
radionuclides
proportional
appearance
with
certain
mass
numbers.
also
shown
there
no
functional
relationship
between
yield
their
atomic
mass,
as
variability
(fluctuations)
content
VVER-440
NPP
reaches
20
orders
magnitude,
which
prevents
application
computational
methods
for
processing
results.
This
paper
proves
that,
order
determine
observable
activity,
applying
a
Zipf-Mandelbrot
type
distribution
“rank
-
size”
coordinates,
it
carry
out
mathematical
modeling
based
appropriate
obtained
Because
of
its
complexity
and
prominence
in
the
logistic
field,
Pallet
Loading
Problem
(PLP)
has
caught
attention
academics
practitioners
as
one
most
challenging
cutting
packing
optimization
problems.
To
solve
PLP,
several
state-of-the-art
Operations
Research
techniques
have
been
used
existing
literature.
This
study
evaluates
capacity
Deep
Reinforcement
Learning
(DRL),
trendiest
machine
learning.
Actually,
DRL
is
a
combination
(RL)
approach
with
Neural
Network
(DNN).
A
specification
phase
training
process
DNN
input/output
nodes
embedded
to
this
NP-hard
combinatorial
problem
provided.
We
UML
standard
formally
specify
complex
sequential
interaction
between
different
components
architecture
during
process.
Thus,
was
tailored
PLP
without
taking
into
account
practical
stability
constraints.
The
proposed
implemented,
computational
experimentation
conducted
on
16
Benchmark
instances
from
Numerical
results
demonstrate
that
effectively
solves
an
average
volume
utilization
nearly
99%.
ensure
package
stability,
realistic
full
support
conditions
were
incorporated
loading
algorithm,
successfully
resolving
corresponding
PLP.
Despite
time-consuming
nature
process,
consistently
achieves
competitive
results,
total
approximately
98%.
Research Square (Research Square),
Journal Year:
2024,
Volume and Issue:
unknown
Published: April 8, 2024
Abstract
Research
focused
on
the
conjunction
between
quantum
computing
and
routing
problems
has
been
very
prolific
in
recent
years.
Most
of
works
revolve
around
classical
such
as
Traveling
Salesman
Problem
or
Vehicle
Routing
Problem.
Even
though
working
these
is
valuable,
it
also
undeniable
that
their
academic-oriented
nature
falls
short
real-world
requirements.
The
main
objective
this
research
to
present
a
solving
method
for
realistic
instances,
avoiding
problem
relaxations
technical
shortcuts.
Instead,
quantum-classical
hybrid
solver
developed,
coined
Q4RPD,
considers
set
real
constraints
heterogeneous
fleet
vehicles,
priority
deliveries,
capacities
characterized
by
two
values:
weight
dimensions
packages.
\Q4RPD
resorts
Leap
Constrained
Quadratic
Model
Hybrid
Solver
D-Wave.
To
demonstrate
application
an
experimentation
composed
six
different
instances
conducted,
aiming
serve
illustrative
examples.
arXiv (Cornell University),
Journal Year:
2023,
Volume and Issue:
unknown
Published: Jan. 1, 2023
The
Bin
Packing
Problem
is
a
classic
problem
with
wide
industrial
applicability.
In
fact,
the
efficient
packing
of
items
into
bins
one
toughest
challenges
in
many
logistic
corporations
and
critical
issue
for
reducing
storage
costs
or
improving
vehicle
space
allocation.
this
work,
we
resort
to
our
previously
published
quantum-classical
framework
known
as
Q4RealBPP,
elaborate
on
solving
real-world
oriented
instances
Problem.
With
purpose,
paper
gravitates
following
characteristics:
i)
existence
heterogeneous
bins,
ii)
extension
solve
not
only
three-dimensional,
but
also
one-
two-dimensional
problem,
iii)
requirements
item-bin
associations,
iv)
delivery
priorities.
All
these
features
have
been
tested
paper,
well
ability
Q4RealBPP
instances.
