Energies,
Journal Year:
2023,
Volume and Issue:
16(6), P. 2574 - 2574
Published: March 9, 2023
The
development
of
data-driven
building
energy
consumption
prediction
models
has
gained
more
attention
in
research
due
to
its
relevance
for
planning
and
conservation.
However,
many
studies
have
conducted
the
inappropriate
application
tools
wrong
conditions.
For
example,
employing
a
tool
develop
model
using
small
sample
size,
despite
recognition
producing
good
results
large
data
This
study
delivers
review
63
with
precise
focus
on
evaluating
performance
based
certain
conditions;
i.e.,
properties,
type
considered,
explored.
identifies
gaps
proposes
future
directions
field
prediction.
Based
reviewed,
outcome
evaluation
shows
that
Support
Vector
Machine
(SVM)
produced
better
than
other
majority
studies.
SVM,
Artificial
Neural
Network
(ANN),
Random
Forest
(RF)
performances
statistical
such
as
Linear
Regression
(LR)
Autoregressive
Integrated
Moving
Average
(ARIMA).
it
is
deduced
none
reviewed
are
predominantly
all
It
clear
their
strengths
weaknesses,
tend
elicit
distinctive
different
Hence,
this
provides
proposed
guideline
selection
weaknesses
Sustainability,
Journal Year:
2023,
Volume and Issue:
15(4), P. 2884 - 2884
Published: Feb. 5, 2023
Proper
analysis
of
building
energy
performance
requires
selecting
appropriate
models
for
handling
complicated
calculations.
Machine
learning
has
recently
emerged
as
a
promising
effective
solution
solving
this
problem.
The
present
study
proposes
novel
integrative
machine
model
predicting
two
parameters
residential
buildings,
namely
annual
thermal
demand
(DThE)
and
weighted
average
discomfort
degree-hours
(HDD).
is
feed-forward
neural
network
(FFNN)
that
optimized
via
the
electrostatic
discharge
algorithm
(ESDA)
analyzing
characteristics
finding
their
optimal
contribution
to
DThE
HDD.
According
results,
proposed
an
double-target
can
predict
required
with
superior
accuracy.
Moreover,
further
verify
efficiency
ESDA,
was
compared
three
similar
optimization
techniques,
atom
search
(ASO),
future
(FSA),
satin
bowerbird
(SBO).
Considering
Pearson
correlation
indices
0.995
0.997
(for
HDD,
respectively)
obtained
ESDA-FFNN
versus
0.992
0.938
ASO-FFNN,
0.926
0.895
FSA-FFNN,
0.994
SBO-FFNN,
ESDA
provided
higher
accuracy
training.
Subsequently,
by
collecting
weights
biases
FFNN,
formulas
were
developed
easier
computation
HDD
in
new
cases.
It
posited
engineers
experts
could
consider
use
along
investigating
buildings.
Environmental Science & Technology,
Journal Year:
2024,
Volume and Issue:
58(15), P. 6457 - 6474
Published: April 3, 2024
The
circular
economy
(CE)
aims
to
decouple
the
growth
of
from
consumption
finite
resources
through
strategies,
such
as
eliminating
waste,
circulating
materials
in
use,
and
regenerating
natural
systems.
Due
rapid
development
data
science
(DS),
promising
progress
has
been
made
transition
toward
CE
past
decade.
DS
offers
various
methods
achieve
accurate
predictions,
accelerate
product
sustainable
design,
prolong
asset
life,
optimize
infrastructure
needed
circulate
materials,
provide
evidence-based
insights.
Despite
exciting
scientific
advances
this
field,
there
still
lacks
a
comprehensive
review
on
topic
summarize
achievements,
synthesize
knowledge
gained,
navigate
future
research
directions.
In
paper,
we
try
how
accelerated
CE.
We
conducted
critical
where
helped
with
focus
four
areas
including
(1)
characterizing
socioeconomic
metabolism,
(2)
reducing
unnecessary
waste
generation
by
enhancing
material
efficiency
optimizing
(3)
extending
lifetime
repair,
(4)
facilitating
reuse
recycling.
also
introduced
limitations
challenges
current
applications
discussed
opportunities
clear
roadmap
for
field.
Energy Conversion and Management,
Journal Year:
2024,
Volume and Issue:
315, P. 118795 - 118795
Published: July 16, 2024
Nowadays,
residential
households,
including
both
consumers
and
emerging
prosumers,
have
exhibited
a
growing
demand
for
active/reactive
power.
This
surge
arises
from
activities
such
as
charging
electrical
devices,
leveraging
flexible
resources,
integrating
renewable
energy
sources.
To
meet
this
escalating
effectively,
operators
must
ensure
the
provision
of
an
ample
supply
Achieving
necessitates
identification
influential
factors
generation
precise
forecasts
power
demand.
Hence,
work
proposes
efficient
hybrid
deep
learning
model
consisting
long
short-term
memory
self-Attention
(LSTM-Attention).
incorporates
explicit
time
encoding
to
forecast
one-hour-ahead
consumption
active
reactive
using
real-time
data
units.
The
integration
models
represents
strategic
development
robustness.
Leveraging
inherent
strengths
architectures
allows
synergistic
compensation
that
addresses
limitations
within
each,
contributing
overall
effective
forecasting
model.
Moreover,
Shapley
Additive
Explanations
(SHAP)
framework
was
employed
interpretability,
investigation
underscores
pivotal
role
incorporating
temporal
features
into
forecasting.
SHAP
findings
can
be
effectively
applied
in
management
strategies
optimally
enhance
response.
Finally,
evaluate
effectiveness
proposed
model,
comprehensive
array
performance
metrics
employed.
results
demonstrate
superior
accuracy
compared
alternative
models.
achieved
lowest
root
mean
square
error
(RMSE),
absolute
(MAE),
percentage
(MAPE)
with
value
0.0256,
0.0181,
14.255
%,
respectively.
formulated
method
also
significantly
contribute
industrial
sector
by
improving
forecasting,
thereby
enhancing
interpretability
identifying
most
critical
factors.
Sensors,
Journal Year:
2022,
Volume and Issue:
22(19), P. 7692 - 7692
Published: Oct. 10, 2022
Building
energy
consumption
prediction
has
become
an
important
research
problem
within
the
context
of
sustainable
homes
and
smart
cities.
Data-driven
approaches
have
been
regarded
as
most
suitable
for
integration
into
houses.
With
wide
deployment
IoT
sensors,
data
generated
from
these
sensors
can
be
used
modeling
forecasting
patterns.
Existing
studies
lag
in
accuracy
various
attributes
buildings
are
not
very
well
studied.
This
study
follows
a
data-driven
approach
this
regard.
The
novelty
paper
lies
fact
that
ensemble
model
is
proposed,
which
provides
higher
performance
regarding
cooling
heating
load
prediction.
Moreover,
influence
different
features
on
investigated.
Experiments
performed
by
considering
such
glazing
area,
orientation,
height,
relative
compactness,
roof
surface
wall
area.
Results
indicate
area
play
significant
role
selecting
appropriate
building.
proposed
achieves
0.999
R2
0.997
prediction,
superior
to
existing
state-of-the-art
models.
precise
load,
help
engineers
design
energy-efficient
buildings,
especially
future
homes.
