Dynamics of Fourier's and Fick's laws on the convectively heated oscillatory sheet under Arrhenius kinetics: The finite-difference technique
Journal of Computational Science,
Год журнала:
2024,
Номер
82, С. 102428 - 102428
Опубликована: Авг. 30, 2024
Язык: Английский
Hybrid photovoltaic/thermal performance prediction based on machine learning algorithms with hyper-parameter tuning
International Journal of Sustainable Energy,
Год журнала:
2024,
Номер
43(1)
Опубликована: Июнь 28, 2024
A
hybrid
Photovoltaic/Thermal(PV/T)
approach
is
proposed
in
this
study
based
on
extensive
research
and
a
comparative
analysis
of
several
hyperparameter
tuning
methods.
The
models
analyzed
are
Linear
Regression
(LR),
Random
Forest
(RF),
XGBoost
Regression,
AdaBoost
Edge
Support
Vector
(SVR),
elastic
net,
lasso
(L)
models.
Grid
search
optimisation
was
used
to
maximise
all
the
model's
hyperparameters.
detailed
presented
as
well
strategies
for
tweaking
positive
negative
suggested
PV/T
evaluated
two
ways.
First,
cumulative
yield
solar
still
obtained.
Second,
support
vector
regression,
followed
by
function
provide
maximum
accuracy
PV
output.
findings
show
that
RF
SVR
achieved
uttermost
precision
both
before
after
use
approach,
with
r2
scores
0.9952,
0.9935,
Root
Mean
Squared
Error
values
0.2583
0.5087
while
utilising
grid
optimisation.
Язык: Английский
Heat and mass transfer dynamics in curved Ш-Chip: ISPH simulations and ANN analysis
Case Studies in Thermal Engineering,
Год журнала:
2024,
Номер
60, С. 104796 - 104796
Опубликована: Июль 11, 2024
This
research
investigates
heat
and
mass
transfer
behavior
in
nano-enhanced
phase
change
materials
(NEPCM)
within
reactive
systems
using
incompressible
smoothed
particle
hydrodynamics
(ISPH)
coupled
with
artificial
neural
network
(ANN)
predictions.
It
introduces
a
novel
model
comprising
six
vertical
rods
one
curved
rod,
forming
unique
Ш-Chip
configuration.
The
complexity
of
the
NEPCM
demonstrates
its
relevance
fluid
dynamics
analyses
applicable
across
diverse
fields
including
food
processing,
electronics
cooling,
chip
manufacturing,
exchangers,
solar
energy
systems.
study
explores
influence
various
dimensionless
parameters
such
as
Frank-Kamenetskii
number
(Fk),
Hartmann
(Ha),
Soret
Dufour
numbers
(Sr&Du),
Lewis
(Le),
Rayleigh
(Ra),
fractional
order
parameter
(α)
on
phenomena.
Heat/mass
sources
from
embedded
filled
are
considered
model.
An
employing
multilayer
perceptron
(MLP)
structure
is
utilized
to
accurately
predict
average
Nusselt
(Nu‾)
Sherwood
(Sh‾)
numbers,
demonstrating
applicability
analyses.
Key
insights
highlight
role
Fk
Ra
enhancing
convection
flow
nanofluid
velocity
Ш-Chip.
Furthermore,
variations
Ha
lead
observable
reductions
due
intensified
Lorentz
forces,
while
increased
Le
values
result
decreased
velocities
thermal
diffusion
becomes
dominant.
aids
transition
unsteady
steady
state.
configuration
Ш-Chip,
combined
incorporation
heat/mass
particles,
offers
promising
applications
underscores
crucial
understanding
pertinent
ensure
effective
development
tailored
specific
needs.
Additionally,
it
emphasizes
practical
value
ANN
models
predictive
modeling
tackle
complexities
encountered
engineering
applications.
Язык: Английский
Applications of neural networking in Eyring-Powell nanofluid dynamics on a rotating surface in a porous medium
Alexandria Engineering Journal,
Год журнала:
2024,
Номер
108, С. 568 - 582
Опубликована: Авг. 6, 2024
One
of
the
fundamental
aspects
solving
difficult
and
nonlinear
mathematical
ideas
is
use
Artificial
Neural
Networks
due
to
their
exceptional
efficiency
in
handling
such
problems.
In
many
complex
fields
as
computational
fluid
system,
biological
computation,
biotechnology,
a
distinct
computing
structure
provided
by
Networks,
which
extremely
valuable.
The
main
purpose
this
article
dig
out
abilities
Levenberg-Marquardt
technique
using
back-propagation
artificial
neural
networks
regarding
mechanics
heat
transport
assessment
nanoparticles.
This
interdisciplinary
field
explores
mass
transfer
through
objects
fluids,
impacts
on
temperature
well
concentration
distributions.
With
help
modelling
numerical
solution
methodologies,
researchers
can
simulate
analyze
these
processes.
present
analysis
communicates
Eyring-Powell
flow
caused
rotating
disk
placed
horizontal
direction.
over
non-linear
partial
differential
equations
modeled.
After
converting
ordinary
ones,
they
are
tackled
numerically
shooting
technique.
algorithm
used
with
reference
datasets,
having
70
%
training,
15
testing,
validation.
method
validated
mean
squared
error,
error
histogram
comprehensive
regression
analysis.
These
figures
show
accuracy
proposed
for
Flow
features
velocity,
profiles
exemplified
quantitatively
have
been
graphically
discussed.
Velocity
decreases
porosity
increases
parameter
while
thermophoresis
Brownian
motion
parameters.
Consistency
shown
getting
minimum
absolute
approaching
zero,
showing
strength
approach.
Язык: Английский
Implementation of stacking regressor model on the flow induced by TiO2‐H2O and Ti6Al4V‐H2O nanofluid with waste discharge concentration
ZAMM ‐ Journal of Applied Mathematics and Mechanics / Zeitschrift für Angewandte Mathematik und Mechanik,
Год журнала:
2024,
Номер
unknown
Опубликована: Окт. 5, 2024
Abstract
The
present
investigation
examines
the
circulation
of
and
based
nanofluids
while
considering
concentration
waste
discharge.
An
innovative
stacking
regressor
model
is
used
to
increase
prediction
accuracy.
Using
Shooting
Runge
Kutta
Fehlberg's
fourth
fifth‐order
schemes,
governing
equations
are
converted
into
ordinary
differential
using
similarity
transformation
then
numerically
solved.
findings
represented
graphically,
model's
correctness
assessed
Gaussian
Process
Regression,
Categorical
Boost,
Extreme
Gradient
Boosting,
Random
Forest,
with
linear
regression
acting
as
a
meta‐model.
closely
related
testing
training
data
show
consistency
stability.
Magnetic
field
inclination
angle
will
decline
velocity,
space,
temperature‐dependent
internal
heat
generation
factors
enhance
temperature.
Raising
pollutant
external
source
parameter
raises
concentration.
In
all
cases,
shows
better
performance
than
nanofluid.
work's
application
ranges
from
fluid
dynamics
management.
By
offering
precise
forecasts
nanofluid
concentration,
proposed
may
aid
in
designing
optimizing
discharge
systems.
Язык: Английский
A comprehensive numerical study on heat transfer and friction characteristics of offset-strip fins
Applied Thermal Engineering,
Год журнала:
2024,
Номер
256, С. 124083 - 124083
Опубликована: Авг. 5, 2024
Offset-strip
fins
are
among
the
most
used
geometries
in
compact
heat
exchangers.
The
geometric
and
flow
parameters
of
affect
their
transfer
effectiveness
head
losses.
Hence,
accurate
predictive
models
needed
to
guide
design
process.
However,
correlations
available
literature
valid
only
for
a
limited
set
configurations
regimes.
This
work
discusses
derivation
multivariate
response
surfaces
equivalent
Darcy
Colburn
factors
offset-strip
fins.
These
feature
clear
applicability
ranges
extend
over
wide
Reynolds
Prandtl
number
(50≤Re≤12000,
0.71≤Pr≤190).
In
addition,
novel
empirical
model
scaling
exponent
is
proposed.
analysis
carried
out
through
Design
Experiment
approach,
performed
with
computational
techniques.
Each
numerical
experiment
by
CFD
periodic
fin
geometry.
obtained
approximate
results
mean
deviation
±8.4%.
Moreover,
application
complete
exchangers
issued
maximum
deviations
±7.8%
±20%,
respectively,
thus
highlighting
usefulness
proposed
modelling
applications.
Язык: Английский