Atmosphere,
Journal Year:
2024,
Volume and Issue:
15(11), P. 1344 - 1344
Published: Nov. 9, 2024
In
recent
years,
the
application
of
machine
learning
methods
has
become
increasingly
common
in
atmospheric
science,
particularly
modeling
and
predicting
processes
that
impact
air
quality.
This
study
focuses
on
hydrogen
production
from
solid
oxide
electrolytic
cells
(SOECs),
a
technology
with
significant
potential
for
reducing
greenhouse
gas
emissions
improving
We
developed
two
models
using
artificial
neural
networks
(ANNs)
support
vector
(SVM)
to
predict
production.
The
input
variables
are
current,
voltage,
communication
delay
time,
real-time
measured
production,
while
output
variable
is
at
next
sampling
time.
Both
address
critical
issue
hysteresis.
Using
50
h
SOEC
system
data,
we
evaluated
effectiveness
ANN
SVM
methods,
incorporating
time
as
an
variable.
results
show
model
superior
terms
prediction
performance.
Specifically,
shows
strong
predictive
performance
ε
=
0.01–0.02
h,
RMSE
2.59
×
10−2,
MAPE
33.34
10−2%,
MAE
1.70
10−2
Nm3/h,
R2
99.76
10−2.
At
0.03
yields
2.74
34.43
1.73
99.73
model,
error
values
2.70
44.01
2.24
99.74
they
2.67
43.44
2.11
99.75
With
this
precision,
positive
implications
pollution
control
strategies
development
cleaner
energy
technologies,
contributing
overall
improvements
quality
reduction
pollutants.
