Data-driven framework for the techno-economic assessment of sustainable aviation fuel from pyrolysis.
Abstract
The
aviation
sector
plays
a
crucial
role
in
quickly
moving
people
and
goods
around
the
world.
It
also
greatly
helps
economic
growth
social
integration
of
countries.
As
industry
continues
to
experience
rapid
growth,
there
is
tendency
for
an
increase
emissions
associated
with
industry.
Sustainable
fuel
(SAF)
presents
way
reduce
environmental
effects
by
providing
clean-burning,
renewable
substitute
conventional
jet
fuel.
SAF
can
be
produced
from
diverse
processes
feedstocks.
Fast
pyrolysis
(FP)
promising
thermochemical
process
production
due
its
advantages
including
low-cost
feedstocks,
faster
reaction
times,
simpler
technology,
making
it
more
cost-effective
scalable
compared
other
processes.
However,
preliminary
estimation
viability
FP
complex
tedious
requiring
detailed
models
several
assumptions.
Moreover,
relationship
between
feedstock
properties
minimum
selling
price
(MSP)
often
challenging
estimate.
To
address
these
challenges,
present
study
developed
data-driven
framework
MSP
FP.
target
output
feature
MSP.
enhance
model
accuracy
predictions,
synthetic
data
was
created
using
Generative
Adversarial
Networks
(GAN)
Variational
Autoencoders
(VAE),
hyperparameter
optimization
conducted
Grid
Search.
Five
surrogate
were
evaluated:
linear
regression,
gradient
boost
regression
(GBR),
random
forest
(RF),
extreme
(XGBoost),
Elastic
net.
GBR
RF
showed
most
promise
based
on
metrics
like
R²,
RMSE,
MAE
both
original
datasets.
Specifically,
achieved
Train
R²
0.9999
Test
0.9277,
while
had
scores
0.9789
0.9255,
respectively.
use
VAE
notably
enhanced
accuracy.
Additionally,
publicly
available
GUI
has
been
researchers
estimate
Aviation
Fuel
biomass
properties,
plant
capacity,
location.
Research Square (Research Square), Год журнала: 2024, Номер unknown
Опубликована: Июль 12, 2024
Язык: Английский
