Numerical Heat Transfer Part B Fundamentals,
Год журнала:
2024,
Номер
unknown, С. 1 - 19
Опубликована: Июль 17, 2024
Nanotechnology
has
recently
led
to
new
possibilities
for
enhancing
heat
transfer
in
exchangers.
The
remarkable
thermal
characteristics
of
graphene
oxide
(GO)-based
nanofluids
with
nanoscale
additions
have
garnered
significant
attention
particular.
When
and
intricate
flow
patterns
are
involved,
traditional
analytical
models
frequently
fail
appropriately
forecast
the
efficiency
Analyzed
phenomenon
laminar
a
exchanger
that
swirling
fluid
dynamics,
pressure
drop,
predicted
condensation
coefficient
(HTC)
LHT.
functionalized
radiated
GO
was
chosen
as
nanomaterials
present
Pre-processing
Data
Methods
managing
outliers
by
machine
learning
(ML)
models,
like
CLAHE
algorithm.
logarithmic
mean
temperature
difference
(LMTD)
can
be
used
determine
driving
power
within
exchanger.
Dynamic
Smagorinsky
Model
(DSLM)
Wall-Adapting
Local
Eddy-viscosity
(WALE)
turbulence
primarily
designed
capturing
turbulent
behavior
flows.
Kern
technique
Hagen–Poiseuille
equation
drop
pumping
needed
shell
tube
through
microtube
based
on
Levenberg–Marquardt
Momentum
Algorithm
predict
Nusselt
number
prediction
sensitivity
HTC
an
LHT-trained
ML
model
is
evaluate
performance.
methods
may
effectively
maximize
performance
setting
nanofluid
accuracy
score
99%,
demonstrating
its
exceptional
predictive
capabilities
results
great
potential
improve
energy
efficiency,
save
operating
costs,
advance
sustainable
practices
various
industrial
applications.
Journal of Materials Research and Technology,
Год журнала:
2023,
Номер
27, С. 7442 - 7456
Опубликована: Ноя. 1, 2023
In
this
study,
the
friction
stir
technique
is
proposed
to
process
aluminum
nanocomposites
reinforced
with
alumina
nanoparticles.
The
effects
of
different
processing
parameters,
including
spindle
speed
(900–1800
rpm),
feed
(10–20
mm/min),
and
number
passes
(1–3)
on
mechanical
dynamic
properties
processed
samples
were
investigated.
investigated
ultimate
tensile
strength,
yield
natural
frequency,
damping
ratio.
An
advanced
machine
learning
approach
composed
a
long
short-term
memory
model
optimized
by
special
relativity
search
algorithm
was
developed
predict
conditions.
adequacy
tested
compared
three
other
models;
predicted
in
good
agreement
measured
properties.
outperformed
models
found
be
powerful
prediction
tool
for
predicting
conditions
obtain
high-quality
nanocomposite
samples.
succeeded
ratio
R2
0.912,
0.952,
0.951,
0.987,
respectively.
obtained
results
showed
that
samples'
loss
factor
increase
passes,
while
shear
modulus,
complex
modulus
decrease
passes.
Thus,
can
used
improve
materials.
The International Journal of Advanced Manufacturing Technology,
Год журнала:
2023,
Номер
130(1-2), С. 527 - 539
Опубликована: Ноя. 30, 2023
Abstract
New
developments
in
manufacturing
processes
impose
the
need
for
experimental
studies
concerning
determination
of
beneficial
process-related
parameter
settings
and
optimization
objectives
related
to
quality
efficiency.
This
work
aims
improve
cutting
geometry,
surface
texture,
arithmetic
roughness
average
case
post-processing
filament
material
extrusion
3D-printed
acrylonitrile
styrene
acrylate
(ASA)
thin
plates
by
a
low-power
CO
2
laser
apparatus.
was
selected
owing
its
unique
properties
thin-walled
customized
constructions.
Three
parameters,
namely
focal
distance,
plate
thickness,
speed,
were
examined
with
reference
Box-Behnken
design
experiments
(BBD)
regression
modeling.
Four
responses
considered:
mean
kerf
width,
Wm
(mm);
down
Wd
upper
Wu
Ra
(μm)
cut
surfaces.
Different
models
tested
their
efficiency
terms
predicting
an
emphasis
on
full
quadratic
regression.
The
results
showed
that
distance
6.5
mm
16
mm/s
speed
optimizes
all
metrics
three
thicknesses.
achieved
adequate
correlation
among
independent
parameters
objectives,
proving
they
can
be
used
process
support
practical
applications.
Case Studies in Thermal Engineering,
Год журнала:
2024,
Номер
60, С. 104645 - 104645
Опубликована: Июнь 5, 2024
In
this
study,
the
performance
of
a
hybrid
power/freshwater
generation
system
is
modeled
using
coupled
artificial
neural
network
(ANN)
model
with
pelican
algorithm
(PA).
The
proposed
composed
Stirling
engine
fixed
to
solar
dish,
desalination
unit,
and
thermoelectric
cooler.
used
generate
electricity
required
operate
electrical-powered
components
as
well
preheat
saline
water.
cooler
supply
water
additional
heat
cool
condensation
surface
unit.
in
terms
yield,
generated
power,
efficiency
was
considered
model's
output;
while
irradiance
dish
diameter
were
inputs.
addition
algorithm,
conventional
gradient
descent
optimizer
employed
an
internal
ANN
model.
prediction
accuracy
two
models
compared
based
on
different
measures.
ANN-PA
outperformed
predicting
efficiency.
computed
root
mean
square
errors
(1.982
L,
104.863
W,
1.227%)
(0.019
1.673
0.047%)
for
efficiency,
respectively.
