Intelligent Pressure Monitoring Method of BP Neural Network Optimized by Genetic Algorithm: A Case Study of X Well Area in Yinggehai Basin
Processes,
Год журнала:
2024,
Номер
12(11), С. 2439 - 2439
Опубликована: Ноя. 5, 2024
While
drilling
formation
pressure
monitoring
is
an
important
basis
for
ensuring
safety
and
oil
gas
discovery,
the
calculation
of
existing
methods
complicated
accuracy
difficult
to
improve.
Taking
actual
well
data
area
X
in
Yinggehai
Basin
as
object,
correlation
analysis
was
first
carried
out
select
standardize
features,
relevant
effective
parameters
were
extracted.
Two
kinds
neural
networks,
back-propagation
network
BP
GA-BP
optimized
by
genetic
algorithm,
used
establish
artificial
intelligence
models
based
on
10
measuring
logging
data,
respectively.
The
application
effect
model
evaluated
results
while
drilling.
show
that
91.25%,
92.89%.
latter
has
a
better
pore
pressure.
In
areas
with
high
degree
control,
introduction
technology
advantages
simplicity,
speed
precision,
can
provide
reference
other
Язык: Английский
Optimization of Rubber Sheet Rolling Machine Parameters using a Taguchi-based TOPSIS Linear Programming Model
Surasit Phokha,
Chailai Sasen,
Pariwat Nasawat
и другие.
Engineering Technology & Applied Science Research,
Год журнала:
2025,
Номер
15(1), С. 20508 - 20516
Опубликована: Фев. 2, 2025
The
Multi-Response
Optimization
(MRO)
problem
is
a
critical
aspect
of
the
engineering
design,
particularly
in
improving
process
efficiency
and
product
quality.
This
study
focuses
on
optimizing
parameters
for
rubber
sheet
rolling
machine,
vital
component
Thailand's
natural
industry.
objective
to
enhance
its
operational
consistency
by
addressing
key
criteria,
such
as
production
time
thickness.
A
novel
approach
integrating
Taguchi
method
Technique
Order
Preference
Similarity
Ideal
Solution
Linear
Programming
(TOPSIS-LP)
model
proposed.
systematically
designs
experiments,
while
(TOPSIS)
consolidates
multiple
performance
indicators
into
single
optimal
solution.
Optimal
roller
gaps
4.5
mm,
3.0
2.0
0.1
mm
first,
second,
third,
fourth
pairs,
were,
respectively,
identified.
results
demonstrated
reduction
thickness
2.06
(5.94%
improvement)
9.71
seconds
per
(1.33%
compared
original
settings.
qualitative
analysis
confirmed
robustness
reliability
optimized
parameters,
achieving
consistent
across
various
evaluation
methods.
presents
significant
advancement
MRO
problem,
offering
robust
framework
applicable
similar
challenges
industrial
findings
provide
foundation
future
automation
optimization
efforts,
driving
sustainable
improvements
manufacturing
Язык: Английский
Optimization of the Production of ε-poly-L-lysine by Streptomyces albus under ethanol stress using Back propagation Neural Network and Genetic Algorithm (BP-GA)
Research Square (Research Square),
Год журнала:
2025,
Номер
unknown
Опубликована: Март 26, 2025
Abstract
To
optimize
the
process
of
ε-poly-L-lysine
(ε-PL)
synthesis
by
Streptomyces
albulus
under
ethanol
stress,
this
study
established
a
Backpropagation
Neural
Network
(BP-NN)
model
based
on
results
single-factor
experiments
and
Plackett-Burman
(PB)
test
design,
using
response
surface
as
data
samples.
The
was
then
optimized
globally
Genetic
Algorithm
(GA).
showed
that
optimal
for
synthesizing
ε-PL
stress
an
addition
1.4554%,
time
0.00
hours,
pH
6.9274,
with
predicted
yield
2.2743
g/L,
actual
obtained
from
experiment
2.277304
g/L.
high
consistency
between
value
verified
reliability
optimization
effect
parameters.
This
provides
example
applying
BP-GA
in
microbial
metabolic
engineering.
Язык: Английский
Experimental Study on the Pelleting and Coating Performance of Red Clover Seeds
Coatings,
Год журнала:
2024,
Номер
14(11), С. 1443 - 1443
Опубликована: Ноя. 13, 2024
This
study
aimed
to
optimize
the
pelleting
and
coating
process
for
red
clover
seeds,
addressing
issue
of
low
success
rates.
Through
theoretical
analysis
experimental
research,
pan
fill
rate,
powder
supply
quantity,
time
were
identified
as
key
factors
influencing
rate.
Single-factor
experiments
conducted
investigate
effects
these
parameters
on
quality
seed
coating.
Based
results,
orthogonal
trials
carried
out,
response
surface
was
employed
reveal
influence
patterns
interactions
each
factor.
The
research
results
indicate
that
affecting
ranked
in
order
importance,
are
time,
quantity.
mathematical
model
optimization,
optimal
combination
determined
be
rate
35.9%,
quantity
160.2
g,
6.9
s.
Under
conditions,
a
94.3%
achieved
validation
experiments.
provides
foundation
practical
guidance
optimizing
which
is
significant
improving
promoting
cultivation.
Язык: Английский
The Effectiveness of Implementing Blended Learning Model in College English Speaking Teaching Supported by Information Technology
Applied Mathematics and Nonlinear Sciences,
Год журнала:
2024,
Номер
9(1)
Опубликована: Янв. 1, 2024
Abstract
The
blended
learning
approach
has
been
explored
many
times
in
theory
and
seems
to
be
argued
about
its
superiority,
but
the
effect
practice
still
lacks
effective
data
prove
it.
This
paper
takes
full
account
of
contingencies
that
exist
teaching
practice,
focuses
on
English-speaking
as
object
study,
improves
control
design.
Integrate
concept
clarify
overall
mechanism
spoken
English.
Ensure
experimental
group
have
comparable
levels
English
through
random
selection,
try
our
best
exclude
interference
other
factors
experiment.
On
hand,
we
build
an
objective
model
quality
by
combining
BP
neural
network
genetic
algorithm.
was
fully
verified
integrating
students’
performance
evaluation
results.
results
show
before
after
teaching,
students
different
degrees
improvement
satisfaction,
motivation,
attitude.
model’s
also
validates
effectiveness
mode
Язык: Английский