International Journal of Robust and Nonlinear Control,
Год журнала:
2025,
Номер
unknown
Опубликована: Апрель 22, 2025
ABSTRACT
This
article
focuses
on
the
parameter
estimation
issues
for
dual‐rate
Volterra
fractional‐order
autoregressive
moving
average
models.
In
case
of
sampling,
we
derive
a
identification
model
system
and
implement
intersample
output
with
help
an
auxiliary
method.
Then,
combined
self‐organizing
map
technique,
propose
Aitken
multi‐innovation
gradient‐based
iterative
algorithm.
The
parameters
are
estimated
using
algorithm,
whereas
differential
orders
determined
Moreover,
computational
cost
proposed
algorithm
is
analyzed
floating
point
operation.
Finally,
convergence
analysis
simulation
examples
show
effectiveness
International Journal of Robust and Nonlinear Control,
Год журнала:
2023,
Номер
33(10), С. 5510 - 5535
Опубликована: Март 30, 2023
Abstract
For
equation‐error
autoregressive
moving
average
systems,
that
is,
Box–Jenkins
this
paper
presents
a
filtered
auxiliary
model
generalized
extended
stochastic
gradient
identification
method,
multi‐innovation
recursive
least
squares
and
method
by
using
the
filtering
idea
idea.
The
proposed
methods
can
be
to
other
linear
nonlinear
multivariable
systems
with
colored
noises.
International Journal of Robust and Nonlinear Control,
Год журнала:
2023,
Номер
33(13), С. 7755 - 7773
Опубликована: Июнь 3, 2023
Summary
This
paper
studies
the
parameter
estimation
problems
of
feedback
nonlinear
systems.
Combining
multi‐innovation
identification
theory
with
negative
gradient
search,
we
derive
a
gradient‐based
iterative
algorithm.
In
order
to
reduce
computational
burden
and
further
improve
accuracy,
decomposition
algorithm
is
proposed
by
using
technique.
The
key
transform
an
original
system
into
two
subsystems
estimate
parameters
each
subsystem,
respectively.
A
simulation
example
provided
demonstrate
effectiveness
algorithms.
International Journal of Adaptive Control and Signal Processing,
Год журнала:
2023,
Номер
37(8), С. 2247 - 2275
Опубликована: Июнь 6, 2023
Summary
Missing
data
often
occur
in
industrial
processes.
In
order
to
solve
this
problem,
an
auxiliary
model
and
a
particle
filter
are
adopted
estimate
the
missing
outputs,
two
unbiased
parameter
estimation
methods
developed
for
class
of
nonlinear
systems
(e.g.,
bilinear
systems)
with
irregularly
data.
Firstly,
is
constructed
unknown
output,
model‐based
multi‐innovation
recursive
least
squares
algorithm
presented
by
expanding
scalar
innovation
vector.
Secondly,
according
bias
compensation
principle,
proposed
compensate
caused
colored
noise.
Thirdly,
further
improving
accuracy,
true
output
estimated
filter,
filtering‐based
developed.
Finally,
numerical
example
selected
validate
effectiveness
algorithms.
The
simulation
results
indicate
that
algorithms
have
good
performance
identifying
Buildings,
Год журнала:
2022,
Номер
12(8), С. 1284 - 1284
Опубликована: Авг. 21, 2022
Building
energy
usage
has
been
an
important
issue
in
recent
decades,
and
prediction
models
are
tools
for
analysing
this
problem.
This
study
provides
a
comprehensive
review
of
building
uncertainties
the
models.
First,
paper
introduces
three
types
methods:
white-box
models,
black-box
grey-box
The
principles,
strengths,
shortcomings,
applications
every
model
discussed
systematically.
Second,
analyses
terms
human,
building,
weather
factors.
Finally,
research
gaps
predicting
consumption
summarised
order
to
guide
optimisation
methods.
IET Biometrics,
Год журнала:
2023,
Номер
12(2), С. 91 - 101
Опубликована: Март 1, 2023
Abstract
Non‐suicide
self‐injury
(NSSI)
can
be
dangerous
and
difficult
for
guardians
or
caregivers
to
detect
in
time.
NSSI
refers
when
people
hurt
themselves
even
though
they
have
no
wish
cause
critical
long‐lasting
hurt.
To
timely
identify
effectively
prevent
order
reduce
the
suicide
rates
of
patients
with
a
potential
risk,
detection
based
on
spatiotemporal
features
indoor
activities
is
proposed.
Firstly,
an
behaviour
dataset
provided,
it
includes
four
categories
that
used
scientific
research
evaluation.
Secondly,
algorithm
(NssiDetection)
NssiDetection
calculates
human
bounding
box
by
using
object
model
employs
extract
temporal
spatial
behaviour.
Thirdly,
optimal
combination
schemes
investigated
checking
its
performance
different
methods
training
strategies.
Lastly,
case
study
performed
implementing
prototype
system.
The
system
has
recognition
accuracy
84.18%
actions
new
backgrounds,
persons,
camera
angles.
