ACS Omega,
Journal Year:
2023,
Volume and Issue:
8(15), P. 13593 - 13604
Published: April 4, 2023
The
main
reaction
range
(350-550
°C)
of
oil-based
drilling
cutting
(OBDC)
pyrolysis
was
studied
by
a
thermogravimetric
analyzer
and
vacuum
tube
furnace.
average
activation
energies
calculated
four
model-free
methods
were
185.5
kJ/mol
(FM),
184.16
(FWO),
166.17
(KAS),
176.03
(Starink).
mechanism
predicted
the
Criado
(Z-master
plot)
method.
It
is
found
that
high
heating
rate
helpful
to
predict
mechanism,
but
it
cannot
be
described
single
model.
Under
conditions
target
temperature
higher
than
350
°C,
residence
time
50
min,
laying
thickness
less
20
mm,
lower
15
residual
oil
content
0.3%
recovery
mineral
98.43%.
Solid
phase
products
accounted
for
more
70%,
reached
maximum
17.04%
at
450
then
decreased
15.87%
500
°C.
Aromatic
hydrocarbons,
as
coking
precursors,
are
transformed
from
low
ring
ring.
Recycled
can
reconfigure
mud
(OBM).
research
results
provide
theoretical
basis
process
optimization.
PLoS ONE,
Journal Year:
2023,
Volume and Issue:
18(8), P. e0289318 - e0289318
Published: Aug. 16, 2023
Accurate
prediction
of
wave
overtopping
at
sea
defences
remains
central
to
the
protection
lives,
livelihoods,
and
infrastructural
assets
in
coastal
zones.
In
addressing
increased
risks
rising
levels
more
frequent
storm
surges,
robust
assessment
methods
for
are
increasingly
important.
Methods
predicting
have
typically
relied
on
empirical
relations
based
physical
modelling
numerical
simulation
data.
recent
years,
with
advances
computational
efficiency,
data-driven
techniques
including
advanced
Machine
Learning
(ML)
become
readily
applicable.
However,
methodological
appropriateness
performance
evaluation
ML
vertical
seawalls
has
not
been
extensively
studied.
This
study
examines
predictive
four
techniques,
namely
Random
Forest
(RF),
Gradient
Boosted
Decision
Trees
(GBDT),
Support
Vector
Machines-Regression
(SVR),
Artificial
Neural
Network
(ANN)
discharge
seawalls.
The
models
developed
using
data
from
EurOtop
(2018)
database.
Hyperparameter
tuning
is
performed
curtail
algorithms
intrinsic
features
dataset.
Feature
Transformation
Selection
adopted
reduce
redundancy
overfitting.
Comprehensive
statistical
analysis
shows
superior
RF
method,
followed
turn
by
GBDT,
SVR,
ANN
models,
respectively.
addition
this,
Tree
(DT)
such
as
GBDT
shown
be
computationally
efficient
than
SVR
ANN,
performing
simulations
rapidly
that
other
methods.
approaches
can
a
reliable
effective
method
evaluating
across
wide
range
hydrodynamic
structural
conditions.
Journal of the Science of Food and Agriculture,
Journal Year:
2024,
Volume and Issue:
104(10), P. 6208 - 6220
Published: March 7, 2024
Five
computational
intelligence
approaches,
namely
Gaussian
process
regression
(GPR),
artificial
neural
network
(ANN),
decision
tree
(DT),
ensemble
of
trees
(EoT)
and
support
vector
machine
(SVM),
were
used
to
describe
the
evolution
moisture
during
dehydration
glutinous
rice.
The
hyperparameters
models
optimized
with
three
strategies:
Bayesian
optimization,
grid
search
random
search.
To
understand
parameters
that
facilitate
model
adaptation
process,
global
sensitivity
analysis
(GSA)
was
compute
impact
input
variables
on
output.
