Академический Документы
Профессиональный Документы
Культура Документы
ScienceDirect
Gudrun
Mikota*
Gudrun
Mikota*
*Institute of Machine Design and Hydraulic
Drives, Johannes Kepler University Linz,
*Institute
of Machine
Design
and Hydraulic Drives,
Johannes
Kepler University
4040
Linz,
Austria
(Tel:
+43-732-2468-6538;
e-mail:
gudrun.mikota@
jku.at)Linz,
*Institute of Machine Design and Hydraulic Drives,
Johannes
Kepler University
Linz,
4040 Linz, Austria (Tel: +43-732-2468-6538; e-mail: gudrun.mikota@ jku.at)
4040 Linz, Austria (Tel: +43-732-2468-6538; e-mail: gudrun.mikota@ jku.at)
Abstract: For hydraulic pipeline systems, a multi-degrees-of-freedom model is developed from the
Abstract:
For hydraulic
systems,
a multi-degrees-of-freedom
model and
is developed
from the
modal decomposition
of pipeline
the transfer
function
between flow rate excitation
pressure response.
Abstract:
For hydraulic
pipeline
systems,
a multi-degrees-of-freedom
model and
is developed
from the
modal
decomposition
thethe
transfer
function
flow rate excitation
pressure
response.
Eigenvectors
are taken of
from
undamped
case. between
Natural frequencies
and damping
ratios
are calculated
modal
decomposition
of
the
transfer
function
between
flow
rate
excitation
and
pressure
response.
Eigenvectors
are taken from the
undamped
case.pipelines
Natural frequencies
damping
ratios are calculated
from modal approximations
of the
individual
with single and
frequency
approximations
at the
Eigenvectors
are taken from the
undamped
case.pipelines
Natural frequencies
and
damping
ratios are calculated
from
modal
approximations
of
the
individual
with
single
frequency
approximations
at the
pipeline system resonances in the low frequency range. The multi-degrees-of-freedom pipeline system
from
modal approximations
thelow
individual
pipelines
withmulti-degrees-of-freedom
single frequency approximations system
at the
pipeline
inofthe
frequency
range. The
model is system
rebuilt resonances
from its modal
description
and evaluated
for a network that connects a pipeline
pump with two
pipeline
system
resonances
in
the
low
frequency
range.
The
multi-degrees-of-freedom
pipeline
system
model
is rebuilt from its modal description and evaluated for a network that connects a pump with two
closed volumes.
model
is rebuilt from its modal description and evaluated for a network that connects a pump with two
closed volumes.
closed
2015,volumes.
IFAC
(International
Federation
of Automatic
Control)
Hosting by
ElsevierEigenvalues,
Ltd. All rights
reserved.
Keywords:
Hydraulic
pipelines,
Approximate
analysis,
Eigenmode
analysis,
Eigenvectors,
Keywords:
Hydraulic pipelines, Approximate analysis, Eigenmode analysis, Eigenvalues, Eigenvectors,
Transfer
functions
Keywords: Hydraulic pipelines, Approximate analysis, Eigenmode analysis, Eigenvalues, Eigenvectors,
Transfer functions
Transfer functions
range.
werethe
explained
by the fact that
damped
pipelineThey
systems,
modal approximation
of anforindividual
1. INTRODUCTION
range.
They
werethe
explained
by the fact that
forindividual
damped
pipeline
systems,
modal
approximation
of
an
1.
INTRODUCTION
is
rather
inaccurate
in
the
frequency
range
below
the
Undamped fluid flow in a pipeline is described by the wave pipeline systems, the modal approximation of an individual
pipeline
is
rather
inaccurate
in
the
frequency
range
below
first
pipeline
resonance.
If
the
pipeline
becomes
part
of
Undamped
fluid Ingard
flow in(1988)
a pipeline
by model
the wave
equation, which
treatsisasdescribed
a common
for pipeline is rather inaccurate in the frequency range below the
thea
Undamped
fluid Ingard
flow in(1988)
a pipeline
isasdescribed
by model
the wave
first
pipeline
resonance.
If
the
pipeline
becomes
part
of
network,
this
frequency
range
will
contain
one
or
more
equation,
which
treats
a
common
for
electromagnetic waves on a cable, sound waves in a fluid, first pipeline resonance. If the pipeline becomes part of aa
equation,
which Ingard
(1988)
treatssound
as a common
model
for network, resonances,
this frequency
range will
contain one orwill
more
for which
the approximation
be
electromagnetic
waves
a cable,
waveswaves
in
a fluid,
longitudinal waves
on aonsolid
bar, and
torsional
on a network
network,
this frequency
range will
contain one orwill
more
electromagnetic
waves
on
a
cable,
sound
waves
in
a
fluid,
network
resonances,
for
which
the
approximation
wrong.
longitudinal
waves
on
a
solid
bar,
and
torsional
waves
on
a
rod. If damping is taken into account, the model for oil network resonances, for which the approximation will be
be
longitudinal
wavesis on
a solid
and torsional
waves onoila wrong.
rod.
If damping
taken
intobar,
account,
the Itmodel
hydraulic
applications
becomes
more
specific.
can befor
based wrong.
rod.
If damping is taken
into more
account,
the Itmodel
for
oil In this paper, the modal approximations from Mikota (2013)
hydraulic
becomes
specific.
can be
based
on
laminarapplications
flow conditions
since
turbulence
increase
paper, in
thea modal
approximations
fromfrequency
Mikota (2013)
hydraulic
applications
becomes
more
specific. would
It can be
based In
arethis
modified
way that
corrects the low
errors
on laminar
turbulence
would increase
this
paper, in
thea modal
approximations
fromfrequency
Mikota (2013)
pressure
lossflow
andconditions
is usuallysince
avoided
in oil hydraulics.
The In
are
modified
way
that
corrects
the
low
errors
on laminar
flow
conditions
since
turbulence
would increase
for
a
predefined
pipeline
system.
Proportional
damping
is
pressure laminar
loss andflow
is usually
avoided inNewtonian
oil hydraulics.
modified
in a pipeline
way that system.
corrects Proportional
the low frequency
errors
transient
of a compressible
fluid The
in a are
for
a predefined
damping
is
pressure
loss
and
is
usually
avoided
in
oil
hydraulics.
The
enforced
on
the
pipeline
system
model
so
that
all
transient circular
laminar flow
of a compressible
fluid in a for a predefined pipeline system. Proportional damping is
straight
pipeline
was describedNewtonian
by transcendental
enforced
on can
thebe pipeline
system
model so
that
all
transient
laminar
flow
of
a
compressible
Newtonian
fluid
in
a
eigenvectors
taken
from
the
undamped
case.
In
the
straight
circular pipeline
was described
by and
transcendental
enforced on the pipeline system model so that all
transfer functions
from D'Souza
et al. (1964)
has further eigenvectors
be taken
the undamped
case. In
the
straight
circular pipeline
was described
by and
transcendental
low
frequencycan
range,
naturalfrom
frequencies
and damping
ratios
transfer
functions
from
D'Souza
et
al.
(1964)
has
further
eigenvectors
can
be taken
from
the undamped
case. In
the
been modelled by rational fraction modal approximations. low
frequency from
range,single
naturalfrequency
frequencies
and damping of
ratios
transfer
functions
from
D'Souza
et
al.
(1964)
and
has
further
are
calculated
approximations
the
been
modelled
by
rational
fraction
modal
approximations.
low
frequency
range,
natural
frequencies
and
damping
ratios
Such models were published by Almondo et al. (2006), are calculated from single frequency approximations of the
been
by rational
fraction modal approximations.
individual pipelines. Higher natural frequencies and
Such modelled
models
published
et al. (2006),
calculated
from single
frequency
approximationsand
of the
Ayalew
et al. were
(2005),
Hsue etbyal. Almondo
(1983), Mkinen
et al. are
individual
pipelines.
Higher
natural
the
Such
models
were
published
by
Almondo
et
al.
(2006),
damping ratios
are
takenfrequencies
from the viscous
Ayalew van
et al.
(2005), (1997),
Hsue etand
al. Yang
(1983),
Mkinen
etthey
al. respective
individual
pipelines.
Higher
natural
frequencies
and
the
(2000),
Schothorst
et
al.
(1991);
respectiveapproximaton
damping ratios
are
taken
from
the
viscous
Ayalew et al. (2005), Hsue et al. (1983), Mkinen et al. damping
as used by Mikota (2014). The new
(2000),
van for
Schothorst
(1997),
and Yang
etare
al. well
(1991);
they
respective
damping
ratios
are
taken
from
the
viscous
can
be
used
time-domain
simulation
and
suited
to
by Mikota
(2014).
new
(2000), van Schothorst (1997), and Yang et al. (1991); they damping
isapproximaton
applied to as
theused
pipeline
network
fromThe
Mikota
can be used
time-domain
simulation
and are wellpipelines.
suited to method
damping
as
by Mikota
(2014).
The
new
study
the for
dynamic
behaviour
of individual
method
isapproximaton
applied
tosignificant
theused
pipeline
network
from
Mikota
can
be
used
for
time-domain
simulation
and
are
well
suited
to
(2014)
and
leads
to
a
improvement
of
the
multistudy the todynamic
behaviour model
of individual
pipelines.
Compared
the transcendental
by D'Souza
et al. method
is
applied
to
the
pipeline
network
from
Mikota
(2014) and leads to pipeline
a significant
improvement
of the multistudy
the todynamic
behaviour model
of individual
pipelines.
system
model.
Compared
the
by
D'Souza
al. degrees-of-freedom
(2014) and leads to pipeline
a significant
improvement
of the multi(1964), Kojima
et transcendental
al. (2002) encountered
large
errors et
when
degrees-of-freedom
system
model.
Compared
to
the
transcendental
model
by
D'Souza
et
al.
(1964),
Kojima modal
et al. (2002)
encountered
errors when
they combined
approximations
for large
the simulation
of degrees-of-freedom pipeline system model.
(1964),
Kojima modal
et al. (2002)
encountered
errors when
2. EXAMPLE SETUP AND PREVIOUS RESULTS
they
combined
approximations
for large
the simulation
compound
pipeline systems;
they therefore
suggested of
to
2. EXAMPLE SETUP AND PREVIOUS RESULTS
they
combined
modal
approximations
for
the
simulation
of
compound
pipeline systems;
theyfunctions
therefore ofsuggested
to
calculate transcendental
transfer
the entire
2. EXAMPLE SETUP AND PREVIOUS RESULTS
compound
pipeline
systems;
they
therefore
suggested
to
calculate
transcendental
transfer
functions
of the entire
system
and
approximate
the
result
in
a
second
step.
calculate transcendental transfer functions of the entire
system and approximate the result in a second step.
system
and approximate
theinjected
result inflow
a second
For
a closed-end
pipeline,
rate step.
excitations and
For
a closed-end
injected
flow (2013)
rate excitations
resulting
pressurepipeline,
responses,
Mikota
derived and
the
For
a closed-end
injected
flow (2013)
rate excitations
and
resulting
pressurepipeline,
responses,
Mikota
derived
the
modal decomposition
of the transcendental
pipeline
model.
resulting
pressure responses,
Mikota (2013)
derived
the
modal
decomposition
the transcendental
pipeline
model.
Transcendental
modal of
transfer
functions were
approximated
modal
decomposition
of
the
transcendental
pipeline
model.
Transcendental
modalexpressions,
transfer functions
by rational fraction
whichwere
lead approximated
to a multiTranscendental
modalexpressions,
transfer functions
were
approximated
by
rational
fraction
which
to a Mikota
multidegrees-of-freedom description of the lead
pipeline.
by rational fraction description
expressions, ofwhich
to a Mikota
multidegrees-of-freedom
the a lead
pipeline.
(2014)
used this model to investigate
specific pipeline
degrees-of-freedom
description
of
the
pipeline.
Mikota
(2014)
used
this model to
investigate
a specific
pipeline
network
and experienced
similar
problems
as Kojima
et al. Fig. 1. Hydraulic pipeline network.
(2014) used
this model to
investigate
a specific
pipeline
network
and
experienced
similar
problems
as Kojima etand
al.
(2002).
However,
by
comparing
transcendental
1. Hydraulic pipeline network.
network and experienced similar problems as Kojima et al. Fig.
Fig. motivate
1. Hydraulic
network.
(2002). However,
comparing
and To
approximated
transferby
functions
of the transcendental
network, it became
the pipeline
necessity
of a low frequency correction,
(2002).
However,
by
comparing
transcendental
and
approximated
functions
of theinnetwork,
became To
motivate
the necessity
of a low
frequency
correction,
clear that thesetransfer
problems
were located
the low itfrequency
example
and results
from Mikota
(2014)
are summarized
in
approximated
functions
of theinnetwork,
became To
motivate
the necessity
of a low
frequency
correction,
clear that thesetransfer
problems
were located
the low itfrequency
example
and results
from Mikota
(2014)
are summarized
in
clear that these problems were located in the low frequency example and results from Mikota (2014) are summarized in
MATHMOD 2015
February 18 - 20, 2015. Vienna, Austria Gudrun Mikota et al. / IFAC-PapersOnLine 48-1 (2015) 435440
436
P( x j , s )
Qex ( xk , s)
Als
(1)
nx j
2E
Gn ( s) cos
Al n 1
l
nxk
cos
l
with the modal transfer function
= 510-5 m2s-1.
Gn ( s)
s
nc
s
l
2
(2)
E
T
1 1 1
Al
(3)
and
2E
Al
T
nx1
nx N 1
nx2
cos
,
cos
cos
l
l
l
n 1,2, , N ,
(4)
nc
, n 0,1, , N .
l
Using the (N+1)(N+1) matrix
n
(5)
0 1 N
(6)
M T
Ku T
436
[n2 ] 1 ,
(7)
(8)
MATHMOD 2015
February 18 - 20, 2015. Vienna, Austria Gudrun Mikota et al. / IFAC-PapersOnLine 48-1 (2015) 435440
Gnv ( s)
f (in )
K ug K
i
(10)
K ug 0
(11)
s
nc
s2
Cv T
[02n ] 1
(17)
[2 n 0n ] 1.
(18)
(12)
(i )
,
enlarged to obtain the respective global matrices M g(i ) , K vg
(i )
and C vg
. These are assembled into the global mass
f (s)2
(16)
with
J 0 ir
f (s)
J 2 ir
(15)
Im
2
f (in )
.
n
2 Re
2
(
)
f
i
Kv T
Gn ( s)
(14)
and
(9)
(i )
ug
0 n n Re
Mg M
s 2 n0n s 02n
where
s
2
437
(i )
K vg K vg
(19)
(i )
C vg C vg
(20)
Av Bv
(21)
C vg
Av
M g
and
437
Mg
0
(22)
MATHMOD 2015
February 18 - 20, 2015. Vienna, Austria Gudrun Mikota et al. / IFAC-PapersOnLine 48-1 (2015) 435440
438
K vg
Bv
0
0
M g
(23)
results in complex conjugate eigenvalues v and eigenvectors v . For undercritical damping, the eigenvalues v are
(24)
and
Im v 1 v2 v .
(25)
f
(
i
)
a
(27)
and
1a
f (ia ) .
Re
2
f (ia )
1 Im
2a
(28)
Rec cc
(29)
and
Im c 1 c2 c .
(30)
C al(i ) Cvl(i ) . If
(26)
with
01a 1 Re
and
G1a ( s)
K al(i ) K vl(i )
Rev vv
MATHMOD 2015
February 18 - 20, 2015. Vienna, Austria Gudrun Mikota et al. / IFAC-PapersOnLine 48-1 (2015) 435440
439
un u u T M g u ,
where u is a matrix composed
1
(31)
of arbitrarily scaled
eigenvectors of the undamped pipeline system and the square
root is taken per matrix element. Squared natural frequencies
c2 and v2 are collected into a diagonal matrix in the
order of the corresponding eigenvectors; the values 2 cc
K cg un
T 1
un 1
(32)
Ccg un
T 1
Dun .
1
(33)
If the flow rates injected at all system nodes are collected into
a global flow rate excitation vector Qex, g and the pressures at
all system nodes are collected into a global pressure vector
Pg , the low frequency correction of the multi-degrees-offreedom model for hydraulic pipeline systems is described by
s M
2
sCcg K cg Pg sQex,g .
(34)
s s 2 M g sCcg K cg
(35)
5. NUMERICAL EVALUATION
For the hydraulic pipeline network described in Section 2, the
first resonance of pipeline 1 lies at 307.4 Hz, and the first
resonance of pipelines 2 and 3 lies at 235.7 Hz. Numerical
investigations are performed with mesh nodes at every 0.1 m.
An undamped eigenvalue analysis of the network yields
natural frequencies at 63.5 Hz, 106.3 Hz, and 202.2 Hz,
which are below the first resonances of pipelines 1, 2, and 3,
and a natural frequency at 259.0 Hz, which is still below the
first resonance of pipeline 1. Single frequency approximations are used at these frequencies, and the viscous
damping approximation above. From 500 Hz upwards, 10 %
damping is assumed to capture the contribution of high
frequency modes while avoiding the tedious transfer of
individual modal parameters.
439
MATHMOD 2015
440
February 18 - 20, 2015. Vienna, Austria Gudrun Mikota et al. / IFAC-PapersOnLine 48-1 (2015) 435440