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 Thermoplastic Composite Materials,
Год журнала:
2025,
Номер
unknown
Опубликована: Май 17, 2025
This
study
explores
the
impact
of
CO
2
laser
cutting
parameters
on
surface
roughness
and
kerf
width
3D-printed
Carbon
Fiber
reinforced
Polylactic
Acid
(PLA-CF)
composites
while
developing
phenomenological
models
using
hybrid
artificial
intelligence
techniques.
PLA-CF
possess
certain
mechanical
properties
quality.
The
values
were
measured
under
different
conditions
(such
as
plate
thickness,
power,
speed)
predicted
multiple
linear
regression,
particle
swarm
optimization-based
adaptive
neuro
fuzzy
inference
system,
ant
colony
system
models.
Experimental
results
showed
that
are
influenced
significantly
by
parameters,
showing
importance
accurately
selecting
parameters.
most
dominant
factor
entered
model
speed:
speed
was
increased,
decreased,
but
higher
levels
power
resulted
in
width.
Thickness
provided
a
non-linear
input:
decreased
from
to
2.5
mm,
then
increased
4
mm.
least
(0.809
mm)
obtained
at
90
W
9
mm/s
speed,
with
mm
thickness.
Surface
minimum
(1.878
µm)
thickness
3
speed.
Among
models,
gave
best
accuracy,
achieving
lowest
mean
squared
error
highest
correlation
coefficient,
whereas
performed
better
than
regression
not
optimization.
These
results,
therefore,
validate
applicability
for
predicting
quality
during
cutting.