Water,
Год журнала:
2021,
Номер
13(5), С. 644 - 644
Опубликована: Фев. 28, 2021
Every
morning,
water
suppliers
need
to
define
their
pump
schedules
for
the
next
24
h
drinking
production.
Plans
must
be
designed
in
such
a
way
that
is
always
available
and
amount
of
unused
pumped
into
network
reduced.
Therefore,
operators
accurately
estimate
day’s
consumption
profile.
In
real-life
applications
with
standard
profiles,
some
expert
system
or
vector
autoregressive
models
are
used.
Still,
recent
years,
significant
improvements
time
series
prediction
have
been
achieved
through
special
deep
learning
algorithms
called
long
short-term
memory
(LSTM)
networks.
This
paper
investigates
applicability
LSTM
demand
optimal
control
compares
LSTMs
against
other
methods
currently
used
by
suppliers.
It
shown
outperform
since
they
can
easily
integrate
additional
information
like
day
week
national
holidays.
Furthermore,
online-
transfer-learning
capabilities
investigated.
only
couple
days
training
data
achieve
reasonable
results.
As
focus
on
real-world
application
LSTMs,
from
two
different
distribution
plants
benchmarking.
Finally,
it
significantly
operation.
IEEE Transactions on Industrial Informatics,
Год журнала:
2018,
Номер
14(11), С. 5112 - 5122
Опубликована: Фев. 2, 2018
As
energy
intensive
infrastructures,
water
distribution
systems
(WDSs)
are
promising
candidates
for
providing
demand
response
(DR)
and
frequency
regulation
services
in
power
operation.
However,
models
that
tap
the
full
flexibility
of
WDSs
to
provide
while
respecting
operational
constraints
networks
remained
scarce.
This
paper
proposes
a
comprehensive
framework
optimizing
participation
system
operators
(W-DSOs)
DR
markets,
which
captures
joint
variable
pumps
tanks
takes
into
account
underlying
hydraulic
operating
WDSs.
The
proposed
consists
two
optimization
models,
where
first-step
model
optimizes
operation
minimizing
W-DSO's
procurement
cost,
second-step
up
down
offers
by
modifying
tanks,
such
profit
is
maximized.
ensures
availability
taking
interdependence
compatibility
load
reduction
recovery
services.
In
addition,
incorporate
detailed
formulation
associated
constraints,
ensuring
deliverability
systems.
nonlinear
terms
appearing
WDS
linearized
convert
instances
mixed-integer
linear
programming
problems.
implemented
on
15-node
WDS,
using
ancillary
service
prices
California
ISO.
results
reflect
significant
opportunities
W-DSO
markets.
Journal of Global Optimization,
Год журнала:
2018,
Номер
71(1), С. 213 - 235
Опубликована: Март 26, 2018
Bayesian
optimization
has
become
a
widely
used
tool
in
the
and
machine
learning
communities.
It
is
suitable
to
problems
as
simulation/optimization
and/or
with
an
objective
function
computationally
expensive
evaluate.
based
on
surrogate
probabilistic
model
of
whose
mean
variance
are
sequentially
updated
using
observations
"acquisition"
model,
which
sets
next
observation
at
most
"promising"
point.
The
Gaussian
Process
basis
well-known
Kriging
algorithms.
In
this
paper,
authors
consider
pump
scheduling
problem
Water
Distribution
Network
both
ON/OFF
variable
speed
pumps.
global
accounting
for
time
patterns
demand
energy
price
allows
significant
cost
savings.
Nonlinearities,
binary
decisions
case
pumps,
make
challenging,
even
small
Networks.
EPANET
simulator
compute
associated
schedule
verify
that
hydraulic
constraints
not
violated
met.
Two
Optimization
approaches
proposed
where
Random
Forest,
respectively.
Both
tested
different
acquisition
functions
set
test
functions,
benchmark
from
literature
large-scale
real-life
Milan,
Italy.
Water Resources Research,
Год журнала:
2020,
Номер
56(8)
Опубликована: Июль 23, 2020
Abstract
The
optimization
of
water
networks
supports
the
decision‐making
process
by
identifying
optimal
trade‐off
between
costs
and
performance
(e.g.,
resilience
leakage).
A
major
challenge
in
domain
distribution
systems
(WDSs)
is
network
(re)design.
While
complex
nature
WDS
has
already
been
explored
with
analysis
(CNA),
literature
still
lacking
a
CNA
networks.
Based
on
systematic
Pareto‐optimal
solutions
different
WDSs,
several
graph
characteristics
are
identified,
newly
developed
design
approach
for
WDSs
proposed.
results
show
that
obtained
designs
comparable
found
evolutionary
optimization,
but
applicable
large
150,000
pipes)
substantially
reduced
computational
effort
(runtime
reduction
up
to
5
orders
magnitude).
Water,
Год журнала:
2022,
Номер
14(6), С. 851 - 851
Опубликована: Март 9, 2022
Water
distribution
networks
are
vital
hydraulic
infrastructures,
essential
for
providing
consumers
with
sufficient
water
of
appropriate
quality.
The
cost
construction,
operation,
and
maintenance
such
is
extremely
large.
problem
optimization
a
network
governed
by
the
type
size
pipelines
placed
in
network.
This
optimal
diameter
allocation
pipes
has
been
heavily
researched
over
past
few
decades.
study
describes
development
an
algorithm,
‘Smart
Optimization
Program
Distribution
Networks’
(SOP–WDN),
which
applies
genetic
algorithm
to
least-cost
design
networks.
SOP–WDN
demonstrates
application
evolutionary
technique,
i.e.,
linked
simulation
solver
EPANET,
developed
was
applied
three
benchmark
problems
produced
consistently
good
results.
can
be
utilized
as
tool
guiding
engineers
during
rehabilitation
pipelines.
IEEE Transactions on Evolutionary Computation,
Год журнала:
2022,
Номер
27(4), С. 1115 - 1129
Опубликована: Июль 27, 2022
This
article
proposes
a
novel
differential
evolution
algorithm
for
solving
constrained
multimodal
multiobjective
optimization
problems
(CMMOPs),
which
may
have
multiple
feasible
Pareto-optimal
solutions
with
identical
objective
vectors.
In
CMMOPs,
due
to
the
coexistence
of
multimodality
and
constraints,
it
is
difficult
current
algorithms
perform
well
in
both
decision
spaces.
The
proposed
uses
speciation
mechanism
induce
niches
preserving
more
adopts
an
improved
environment
selection
criterion
enhance
diversity.
can
not
only
obtain
but
also
retain
well-distributed
solutions.
Moreover,
set
test
functions
developed.
All
these
characteristics
contain
constraints.
Meanwhile,
this
new
indicator,
comprehensively
considers
feasibility,
convergence,
diversity
solution
set.
effectiveness
method
verified
by
comparing
state-of-the-art
on
real-world
location-selection
problem.
Knowledge-Based Systems,
Год журнала:
2024,
Номер
299, С. 111998 - 111998
Опубликована: Май 29, 2024
Constrained
multi-objective
optimization
problems
(CMOPs)
are
widespread
in
practical
applications
such
as
engineering
design,
resource
allocation,
and
scheduling
optimization.
It
is
high
challenging
for
CMOPs
to
balance
the
convergence
diversity
due
conflicting
objectives
complex
constraints.
Researchers
have
developed
a
variety
of
constrained
algorithms
(CMOAs)
find
set
optimal
solutions,
including
evolutionary
machine
learning-based
methods.
These
exhibit
distinct
advantages
solving
different
categories
CMOPs.
Recently,
(CMOEAs)
emerged
popular
approach,
with
several
literature
reviews
available.
However,
there
lack
comprehensive-view
survey
on
methods
CMOAs,
limiting
researchers
track
cutting-edge
investigations
this
research
direction.
Therefore,
paper
latest
handling
A
new
classification
method
proposed
divide
literature,
containing
classical
mathematical
methods,
learning
Subsequently,
it
modeling
context
applications.
Lastly,
gives
potential
directions
respect
This
able
provide
guidance
inspiration
scholars
studying
