Thermal Science,
Год журнала:
2023,
Номер
28(2 Part A), С. 863 - 875
Опубликована: Авг. 2, 2023
In
the
field
of
heat
transfer
in
permanent
magnet
synchronous
motors
(PMSM)
for
electric
vehicles,
boundary
element
method
(BEM)
has
been
applied
first
time
to
calculate
steady-state
temperature
PMSM
with
a
spiral
water-cooled
system.
this
investigation,
boundary-integration
equation
problem
is
derived
on
basis
thermodynamic
theory,
and
system
constant
coefficient
differential
equations
obtained
by
discretizing
its
boundaries,
while
results
from
BEM
are
compared
finite
(FEM)
results.
Furthermore,
distribution
characteristics
FEM
were
verified
twice
using
prototype
test
platform.
The
show
that
maximum
relative
error
between
calculation
1.97%,
does
not
exceed
3%,
which
finally
verifies
validity
accuracy
solving
problems
PMSM.
Oilseed
rape
is
an
important
oilseed
crop
planted
worldwide.
Maturity
classification
plays
a
crucial
role
in
enhancing
yield
and
expediting
breeding
research.
Conventional
methods
of
maturity
are
laborious
destructive
nature.
In
this
study,
nondestructive
model
was
established
on
the
basis
hyperspectral
imaging
combined
with
machine
learning
algorithms.
Initially,
images
were
captured
for
3
distinct
ripeness
stages
rapeseed,
raw
spectral
data
extracted
from
images.
The
underwent
preprocessing
using
5
pretreatment
methods,
namely,
Savitzky–Golay,
first
derivative,
second
derivative
(D2nd),
standard
normal
variate,
detrend,
as
well
various
combinations
these
methods.
Subsequently,
feature
wavelengths
processed
spectra
competitive
adaptive
reweighted
sampling,
successive
projection
algorithm
(SPA),
iterative
spatial
shrinkage
interval
variables
(IVISSA),
their
combination
algorithms,
respectively.
models
constructed
following
algorithms:
extreme
machine,
k
-nearest
neighbor,
random
forest,
partial
least-squares
discriminant
analysis,
support
vector
(SVM)
applied
separately
to
full
wavelength
wavelengths.
A
comparative
analysis
conducted
evaluate
performance
diverse
selection
models,
results
showed
that
based
preprocessing-feature
selection-machine
could
effectively
predict
rapeseed.
D2nd-IVISSA-SPA-SVM
exhibited
highest
modeling
performance,
attaining
accuracy
rate
97.86%.
findings
suggest
rapeseed
can
be
rapidly
nondestructively
ascertained
through
imaging.
Energies,
Год журнала:
2024,
Номер
17(10), С. 2307 - 2307
Опубликована: Май 10, 2024
This
study
systematically
explores
and
compares
the
performance
of
various
artificial-intelligence
(AI)-based
models
to
predict
electrical
thermal
efficiency
photovoltaic–thermal
systems
(PVTs)
cooled
by
nanofluids.
Employing
extreme
gradient
boosting
(XGB),
extra
tree
regression
(ETR),
k-nearest-neighbor
(KNN)
models,
their
accuracy
is
quantitatively
evaluated,
effectiveness
measured.
The
results
demonstrate
that
both
XGB
ETR
consistently
outperform
KNN
in
accurately
predicting
efficiency.
Specifically,
model
achieves
remarkable
correlation
coefficient
(R2)
values
approximately
0.99999,
signifying
its
superior
predictive
capabilities.
Notably,
exhibits
a
slightly
compared
estimating
Furthermore,
when
efficiency,
excellence,
with
showing
slight
edge
based
on
R2
values.
Validation
against
new
data
points
reveals
outstanding
performance,
attaining
0.99997
for
0.99995
These
quantitative
findings
underscore
reliability
PVT
study’s
implications
are
significant
system
designers
industry
professionals,
as
incorporation
AI-based
offers
improved
accuracy,
faster
prediction
times,
ability
handle
large
datasets.
presented
this
contribute
optimization,
evaluation,
decision-making
field.
Additionally,
robust
validation
enhances
credibility
these
advancing
overall
understanding
applicability
AI
systems.
