Prospects and Challenges of Reinforcement Learning- Based HVAC Control
Journal of Building Engineering,
Год журнала:
2024,
Номер
unknown, С. 111080 - 111080
Опубликована: Окт. 1, 2024
Язык: Английский
Multi-Objective Day-Ahead Scheduling for Air Conditioning Load Considering Dynamic Carbon Emission Factor
Kun Zhang,
Zhengxun Guo,
Ji Wang
и другие.
Electronics,
Год журнала:
2025,
Номер
14(8), С. 1550 - 1550
Опубликована: Апрель 11, 2025
The
traditional
optimal
scheduling
of
air
conditioning
load
has
traditionally
focused
on
improving
user
comfort
and
reducing
electricity
costs.
However,
research
carbon
emissions
generated
during
the
operation
is
still
in
developmental
stage.
Currently
used
average
emission
factors
carbon-emission
studies
face
challenges
such
as
delayed
data
updates
difficulty
reflecting
spatiotemporal
variations.
These
issues
contribute
to
inaccurate
quantification
emissions,
creating
a
challenging
situation,
which
does
not
meet
development
needs
green
power
systems
under
“dual
goals”.
Therefore,
this
paper
proposes
multi-objective
method
for
cooling-dominant
considering
dynamic
factor
(CEF),
conjunction
with
real-time
from
grid
model,
generates
electric
each
moment
throughout
day.
Firstly,
while
comfort,
CEF-based
cost
are
fused
into
user’s
comprehensive
cost,
optimization
model
day-ahead
loads
established.
In
addition,
above
solved
by
NSGA-II
algorithm
obtain
Pareto
front
composed
non-dominated
solutions.
Then,
best
compromise
solution
objectively
selected
through
gray
target
decision
(GTD)
provide
scientific
decision-making
loads.
Finally,
four
users
different
room
temperature
requirements
designed
verify
effectiveness
proposed
strategy.
simulation
results
illustrate
that
compared
single-objective
simple
methods,
strategy
possesses
stronger
trade-off
ability,
can
greatly
reduce
ensuring
comfort.
Язык: Английский
Optimal Load Distribution Control for Airport Terminal Chiller Units Based on Deep Reinforcement Learning
Journal of Building Engineering,
Год журнала:
2024,
Номер
unknown, С. 110787 - 110787
Опубликована: Сен. 1, 2024
Язык: Английский
Innovative Energy Solutions: Evaluating Reinforcement Learning Algorithms for Battery Storage Optimization in Residential Settings
Process Safety and Environmental Protection,
Год журнала:
2024,
Номер
unknown
Опубликована: Окт. 1, 2024
Язык: Английский
Algorithmic Complexity Research on Foreign-related Tourism English Education and Talent Cultivation Models in the Context of Free Trade Port
Applied Mathematics and Nonlinear Sciences,
Год журнала:
2024,
Номер
9(1)
Опубликована: Янв. 1, 2024
Abstract
In
today’s
booming
development
of
free
trade
ports,
China,
as
a
big
country
inbound
and
outbound
tourism,
has
put
forward
higher
requirements
for
the
talent
cultivation
mode
foreign-related
tourism
English.
This
paper
takes
characteristics
English
entry
point
establishes
practical
teaching
system
four-in-one
The
multi-objective
optimization
model
cultivating
foreign
talents
is
proposed,
with
objective
function
enhancing
quality
talents,
corresponding
constraints
are
set.
solved
using
DQN
model,
strategy
gradient
algorithm-based
training
proposed.
test
conditions
set,
simulation
experiments
carried
out,
compared
based
on
optimal
scheme.
loss
obtained
by
after
1700
iterations
fluctuates
between
[0.015,0.019],
when
period
set
to
2
years,
cumulative
return
can
reach
623.87%,
maximum
retracement
only
-
5.457%.
program’s
average
score
14.1
points
than
ordinary
version.
Foreign-related
needs
be
fully
integrated
actual
demand
talent,
faculty
policy
guidelines
help
enhance
training.
Язык: Английский