Energy and Built Environment,
Journal Year:
2023,
Volume and Issue:
5(1), P. 143 - 169
Published: June 16, 2023
Advanced
data
mining
methods
have
shown
a
promising
capacity
in
building
energy
management.
However,
the
past
decade,
such
are
rarely
applied
practice,
since
they
highly
rely
on
users
to
customize
solutions
according
characteristics
of
target
systems.
Hence,
major
barrier
is
that
practical
applications
remain
laborious.
It
necessary
enable
computers
human-like
ability
solve
tasks.
Generative
pre-trained
transformers
(GPT)
might
be
capable
addressing
this
issue,
as
some
GPT
models
GPT-3.5
and
GPT-4
powerful
abilities
interaction
with
humans,
code
generation,
inference
common
sense
domain
knowledge.
This
study
explores
potential
most
advanced
model
(GPT-4)
three
scenarios
management,
i.e.,
load
prediction,
fault
diagnosis,
anomaly
detection.
A
performance
evaluation
framework
proposed
verify
capabilities
generating
prediction
codes,
diagnosing
device
faults,
detecting
abnormal
system
operation
patterns.
demonstrated
can
automatically
tasks
domain,
which
overcomes
domain.
In
exploration
GPT-4,
its
advantages
limitations
also
discussed
comprehensively
for
revealing
future
research
directions
Energy and Buildings,
Journal Year:
2023,
Volume and Issue:
292, P. 113171 - 113171
Published: May 18, 2023
In
an
increasingly
digital
world,
there
are
fast-paced
developments
in
fields
such
as
Artificial
Intelligence,
Machine
Learning,
Data
Mining,
Digital
Twins,
Cyber-Physical
Systems
and
the
Internet
of
Things.
This
paper
reviews
discusses
how
these
new
emerging
areas
relate
to
traditional
domain
building
performance
simulation.
It
explores
boundaries
between
simulation
other
order
identify
conceptual
differences
similarities,
strengths
limitations
each
areas.
The
critiques
common
notions
about
domains
they
simulation,
reviewing
field
may
evolve
benefit
from
developments.
Energy and Buildings,
Journal Year:
2023,
Volume and Issue:
303, P. 113768 - 113768
Published: Nov. 22, 2023
Stakeholders
such
as
urban
planners
and
energy
policymakers
use
building
performance
modeling
analysis
to
develop
strategic
sustainable
plans
with
the
aim
of
reducing
consumption
emissions
from
built
environment.
However,
inconsistent
data
lack
scalable
models
create
a
gap
between
traditional
planning
practices.
An
alternative
approach
is
conduct
large-scale
usage
survey,
which
time-consuming.
Similarly,
existing
studies
rely
on
machine
learning
or
statistical
approaches
for
calculating
performance.
This
paper
proposes
solution
that
employs
data-driven
predict
residential
buildings,
using
both
ensemble-based
end-use
demand
segregation
methods.
The
proposed
methodology
consists
five
steps:
collection,
archetype
development,
physics-based
parametric
modeling,
analysis.
devised
tested
Irish
stock
generates
synthetic
dataset
one
million
buildings
through
19
identified
vital
variables
four
archetypes.
As
part
process,
study
implemented
an
method,
including
heating,
lighting,
equipment,
photovoltaic,
hot
water,
at
scale.
Furthermore,
model's
enhanced
by
employing
approach,
achieving
91%
accuracy
compared
approach's
76%.
Accurate
prediction
enables
stakeholders,
planners,
make
informed
decisions
when
retrofit
measures.
Applied Energy,
Journal Year:
2020,
Volume and Issue:
279, P. 115834 - 115834
Published: Sept. 9, 2020
Urban
planners,
local
authorities,
and
energy
policymakers
often
develop
strategic
sustainable
plans
for
the
urban
building
stock
in
order
to
minimize
overall
consumption
emissions.
Planning
at
such
scales
could
be
informed
by
modeling
using
existing
data
Geographic
Information
System-based
mapping.
However,
implementing
these
processes
involves
several
issues,
namely,
availability,
inconsistency,
scalability,
integration,
geocoding,
privacy.
This
research
addresses
aforementioned
information
challenges
proposing
a
generalized
integrated
methodology
that
implements
bottom-up,
data-driven,
spatial
approaches
multi-scale
System
mapping
of
modeling.
study
uses
Irish
map
performance
multiple
scales.
The
data-driven
approximately
650,000
Energy
Performance
Certificates
buildings
predict
more
than
2
million
buildings'
performance.
In
this
case,
approach
delivers
prediction
accuracy
88%
deep
learning
algorithms.
These
results
are
then
used
from
individual
level
national
level.
Furthermore,
maps
coupled
with
available
resources
(social,
economic,
or
environmental
data)
planning,
analysis,
support
decision-making.
identify
clusters
have
significant
potential
savings
within
any
specific
region.
aids
stakeholders
identifying
priority
areas
efficiency
measures.
target
communities
retrofit
campaigns,
which
would
enhance
implementation
policy
decisions.
