Electronic Research Archive,
Journal Year:
2024,
Volume and Issue:
32(11), P. 6364 - 6378
Published: Jan. 1, 2024
<p>Load
forecasting
is
an
important
part
of
microgrid
control
and
operation.
To
improve
the
accuracy
reliability
load
in
microgrid,
a
method
based
on
adaptive
cuckoo
search
optimization
improved
neural
network
(ICS-BP)
was
proposed.
First,
model
designed.
Then,
novel
step
adjustment
strategy
proposed
for
optimization,
position
update
law
loss
fluctuation
Finally,
weights
biases
were
optimized
by
algorithm.
The
results
showed
that
BP
enhanced
global
ability,
avoided
local
optima,
quickened
convergence
speed,
presented
excellent
performance
forecasting.
mean
absolute
percentage
error
(MAPE)
ICS-BP
1.13%,
which
very
close
to
ideal
prediction
model,
52.3,
32.8,
42.3%
lower
than
conventional
BP,
particle
swarm
respectively,
root
square
(RMSE),
(MAE),
(MSE)
reduced
75.6,
70.6,
94.0%,
compared
BP.
significantly
reliability,
can
effectively
realize
microgrid.</p>
International Journal of Renewable Energy Development,
Journal Year:
2024,
Volume and Issue:
13(2), P. 329 - 339
Published: Feb. 20, 2024
To
tackle
the
challenges
associated
with
variability
and
uncertainty
in
distributed
power
generation,
as
well
complexities
of
solving
high-dimensional
energy
management
mathematical
models
mi-crogrid
optimization,
a
microgrid
optimization
method
is
proposed
based
on
an
improved
soft
actor-critic
algorithm.
In
method,
algorithm
employs
entropy-based
objective
function
to
encourage
target
exploration
without
assigning
signifi-cantly
higher
probabilities
any
part
action
space,
which
can
simplify
analysis
process
generation
while
effectively
mitigating
convergence
fragility
issues
model
management.
The
effectiveness
validated
through
case
study
op-timization
results
revealed
increase
51.20%,
52.38%,
13.43%,
16.50%,
58.26%,
36.33%
total
profits
compared
Deep
Q-network
algorithm,
state-action-reward-state-action
proximal
policy
ant-colony
strategy
genetic
fuzzy
inference
system,
theoretical
retailer
stragety,
respectively.
Additionally,
com-pared
other
methods
strategies,
learn
more
optimal
behaviors
anticipate
fluctuations
electricity
prices
demand.
Entropy,
Journal Year:
2024,
Volume and Issue:
26(8), P. 699 - 699
Published: Aug. 17, 2024
To
meet
the
challenges
of
energy
sustainability,
integrated
system
(IES)
has
become
a
key
component
in
promoting
development
innovative
systems.
Accurate
and
reliable
multivariate
load
prediction
is
prerequisite
for
IES
optimal
scheduling
steady
running,
but
uncertainty
fluctuation
many
influencing
factors
increase
difficulty
forecasting.
Therefore,
this
article
puts
forward
multi-energy
approach
IES,
which
combines
fennec
fox
optimization
algorithm
(FFA)
hybrid
kernel
extreme
learning
machine.
Firstly,
comprehensive
weight
method
used
to
combine
entropy
Pearson
correlation
coefficient,
fully
considering
information
content
correlation,
selecting
affecting
prediction,
ensuring
that
input
features
can
effectively
modify
results.
Secondly,
coupling
relationship
between
learned
predicted
using
At
same
time,
FFA
parameter
optimization,
reduces
randomness
setting.
Finally,
utilized
measured
data
at
Arizona
State
University
verify
its
effectiveness
The
results
indicate
mean
absolute
error
(MAE)
proposed
0.0959,
0.3103
0.0443,
respectively.
root
square
(RMSE)
0.1378,
0.3848
0.0578,
weighted
percentage
(WMAPE)
only
1.915%.
Compared
other
models,
model
higher
accuracy,
with
maximum
reductions
on
MAE,
RMSE
WMAPE
0.3833,
0.491
2.8138%,
Discover Sustainability,
Journal Year:
2024,
Volume and Issue:
5(1)
Published: July 30, 2024
Abstract
Microgrids
have
emerged
as
a
promising
solution
for
enhancing
energy
sustainability
and
resilience
in
localized
distribution
systems.
Efficient
management
accurate
load
forecasting
are
one
of
the
critical
aspects
improving
operation
microgrids.
Various
approaches
prediction
using
statistical
models
discussed
literature.
In
this
work,
novel
framework
that
incorporates
machine
learning
(ML)
techniques
is
presented
an
solar
wind
generation.
The
anticipated
approach
also
emphasizes
time
series-based
microgrids
with
precise
estimation
State
Charge
(SoC)
battery.
A
unique
feature
proposed
utilizes
historical
data
employs
series
analysis
coupled
different
ML
to
forecast
demand
commercial
scenario.
Long
Short-Term
Memory
(LSTM)
Linear
Regression
(LR)
employed
experimental
study
under
three
cases,
such
(i)
generation,
(ii)
and,
(iii)
SoC
results
show
Random
Forest
(RF)
LSTM
performs
well
respectively.
On
other
hand,
Artificial
Neural
Network
(ANN)
model
exhibited
superior
accuracy
terms
estimation.
Further,
Graphical
User
Interface
(GUI)
developed
evaluating
efficacy
framework.
