Water Environment Research,
Год журнала:
2024,
Номер
96(8)
Опубликована: Авг. 1, 2024
Abstract
In
this
study,
we
employed
the
response
surface
method
(RSM)
and
long
short‐term
memory
(LSTM)
model
to
optimize
operational
parameters
predict
chemical
oxygen
demand
(COD)
removal
in
electrocoagulation‐catalytic
ozonation
process
(ECOP)
for
pharmaceutical
wastewater
treatment.
Through
RSM
simulation,
quantified
effects
of
reaction
time,
ozone
dose,
current
density,
catalyst
packed
rate
on
COD
removal.
Then,
optimal
conditions
achieving
a
efficiency
exceeding
50%
were
identified.
After
evaluating
ECOP
performance
under
optimized
conditions,
LSTM
predicted
(56.4%),
close
real
results
(54.6%)
with
0.2%
error.
outperformed
predictive
capacity
initial
concentration
effluent
discharge
standards,
intelligent
adjustment
operating
becomes
feasible,
facilitating
precise
control
based
model.
This
strategy
holds
promise
enhancing
treatment
scenarios.
Practitioner
Points
study
utilized
optimization.
(56.4%)
closely
matched
experimental
(54.6%),
minimal
error
0.2%.
demonstrated
superior
capacity,
enabling
parameter
adjustments
enhanced
control.
Intelligent
improving
Water,
Год журнала:
2025,
Номер
17(3), С. 310 - 310
Опубликована: Янв. 23, 2025
When
confronted
with
different
influent
conditions,
WWTPs
often
lack
targeted
and
effective
operational
control
strategies.
For
the
three
typical
scenarios
of
low
C/N,
water
temperature
high
temperature,
441
carbon
source
dosage
DO
concentration
coordination
strategies
were
designed
under
premise
ensuring
effluent
quality
meets
standard.
The
purpose
was
to
provide
clear
guidance
for
efficient
operation
in
scenarios.
To
determine
optimal
strategy,
prediction
model
based
on
LSTM
GRU
constructed
testing.
results
showed
that:
(1)
LSTM-GRU
is
better
than
SVR
RF
predicting
COD
TN;
(2)
In
C/N
scenario,
should
be
controlled
between
0.23
t/h
0.26
t/h,
ranging
from
2.0
mg/L
2.6
mg/L;
(3)
0.25
0.27
2.8
(4)
0.20
2.5
mg/L.
Water Research X,
Год журнала:
2024,
Номер
26, С. 100291 - 100291
Опубликована: Дек. 3, 2024
Sudden
shocking
load
events
featuring
significant
increases
in
inflow
quantities
or
concentrations
of
wastewater
treatment
plants
(WWTPs),
are
a
major
threat
to
the
attainment
treated
effluents
discharge
quality
standards.
To
aid
real-time
decision-making
for
stable
WWTP
operations,
this
study
developed
probabilistic
deep
learning
model
that
comprises
encoder-decoder
long
short-term
memory
(LSTM)
networks
with
added
capacity
producing
probability
predictions,
enhance
robustness
effluent
prediction
under
such
events.
The
LSTM
(P-ED-LSTM)
was
tested
an
actual
WWTP,
where
bihourly
total
nitrogen
performed
and
compared
classical
models,
including
LSTM,
gated
recurrent
unit
(GRU)
Transformer.
It
found
events,
P-ED-LSTM
could
achieve
49.7%
improvement
accuracy
predictions
concentration
GRU,
A
higher
quantile
data
from
output,
indicated
value
more
approximate
real
quality.
also
exhibited
predictive
power
next
multiple
time
steps
scenarios.
captured
approximately
90%
over-limit
discharges
up
6
hours
ahead,
significantly
outperforming
other
models.
Therefore,
model,
its
robust
adaptability
fluctuations,
has
potential
broader
applications
across
WWTPs
different
processes,
as
well
providing
strategies
system
regulation
emergency
conditions.
