International Journal of Energy Research,
Journal Year:
2024,
Volume and Issue:
2024(1)
Published: Jan. 1, 2024
Biomass
is
an
excellent
source
of
green
energy
with
numerous
benefits
such
as
abundant
availability,
net
carbon
zero,
and
renewable
nature.
However,
the
conventional
methods
biomass
combustion
are
polluting
poor
efficiency
processes.
gasification
overcomes
these
challenges
provides
a
sustainable
method
for
supply
greener
fuel
in
form
producer
gas.
The
gas
can
be
employed
gaseous
compression
ignition
engines
dual‐fuel
systems.
process
complex
well
nonlinear
that
highly
dependent
on
ambient
environment,
type
biomass,
composition
medium.
This
makes
modeling
systems
quite
difficult
time‐consuming.
Modern
machine
learning
(ML)
techniques
offer
use
experimental
data
convenient
approach
to
forecasting
In
present
study,
two
modern
efficient
ML
techniques,
random
forest
(RF)
AdaBoost,
were
this
purpose.
outcomes
results
baseline
method,
i.e.,
linear
regression.
RF
could
forecast
hydrogen
yield
R
2
0.978
during
model
training
0.998
test
phase.
AdaBoost
was
close
behind
at
0.948
0.842
mean
squared
error
low
0.17
0.181
testing,
respectively.
case
heating
value
model,
0.971
respectively,
Both
provided
compared
regression,
but
RFt
best
among
all
three.
International Journal of Environmental Research,
Journal Year:
2024,
Volume and Issue:
19(1)
Published: Oct. 24, 2024
Abstract
This
study
aims
to
explore
the
application
of
artificial
intelligence
(AI)
in
resolution
sustainability
challenges,
with
a
specific
focus
on
environmental
studies.
Given
rapidly
evolving
nature
this
field,
there
is
an
urgent
need
for
more
frequent
and
dynamic
reviews
keep
pace
innovative
applications
AI.
Through
systematic
analysis
191
research
articles,
we
classified
AI
techniques
applied
field
sustainability.
Our
review
found
that
65%
studies
supervised
learning
methods,
18%
employed
unsupervised
learning,
17%
utilized
reinforcement
approaches.
The
highlights
neural
networks
(ANN),
are
most
commonly
contexts,
accounting
23%
reviewed
methods.
comprehensive
overview
identifies
key
trends
proposes
new
avenues
address
complex
issue
achieving
Sustainable
Development
Goals
(SDGs).
Graphic
abstract
Journal of Environmental Science and Health Part A,
Journal Year:
2025,
Volume and Issue:
unknown, P. 1 - 16
Published: Feb. 2, 2025
There
are
several
uses
for
biomass-derived
materials
(BDMs)
in
the
irrigation
and
farming
industries.
To
solve
problems
with
material,
process,
supply
chain
design,
BDM
systems
have
started
to
use
machine
learning
(ML),
a
new
technique
approach.
This
study
examined
articles
published
since
2015
understand
better
current
status,
future
possibilities,
capabilities
of
ML
supporting
environmentally
friendly
development
applications.
Previous
applications
were
classified
into
three
categories
according
their
objectives:
material
process
performance
prediction
sustainability
evaluation.
helps
optimize
BDMs
systems,
predict
properties
performance,
reverse
engineering,
data
difficulties
evaluations.
Ensemble
models
cutting-edge
Neural
Networks
operate
satisfactorily
on
these
datasets
easily
generalized.
neural
network
poor
interpretability,
there
not
been
any
studies
assessment
that
consider
geo-temporal
dynamics;
thus,
building
methods
is
currently
practical.
Future
research
should
follow
workflow.
Investigating
potential
system
optimization,
evaluation
sustainable
requires
further
investigation.
Sustainability,
Journal Year:
2025,
Volume and Issue:
17(7), P. 3216 - 3216
Published: April 4, 2025
Pyrolysis
presents
a
promising
solution
for
the
complete
purification
and
recycling
of
waste
salt.
However,
presence
organic
pollutants
in
salts
significantly
hinders
their
practical
application,
owing
to
diverse
sources
strong
resistance
degradation.
This
study
developed
predictive
models
removal
from
salt
using
three
machine
learning
techniques:
Random
Forest
(RF),
Support
Vector
Machine,
Artificial
Neural
Network.
The
were
evaluated
based
on
total
carbon
(TOC)
rate
mass
loss
rate,
with
RF
model
demonstrating
high
accuracy,
achieving
R2
values
0.97
0.99,
respectively.
Feature
engineering
revealed
that
contribution
components
was
minimal,
rendering
them
redundant.
importance
analysis
identified
temperature
as
most
critical
factor
TOC
removal,
while
moisture
content
nitrogen
key
determinants
loss.
Partial
dependence
plots
further
elucidated
individual
interactive
effects
these
variables.
validated
both
literature
data
laboratory
experiments,
user
interface
Python
GUI
library.
provides
novel
insights
into
pyrolysis
process
establishes
foundation
optimizing
its
application.
