Welcome to Scribd, the world's digital library. Read, publish, and share books and documents. See more
Download
Standard view
Full view
of .
Save to My Library
Look up keyword
Like this
1Activity
0 of .
Results for:
No results containing your search query
P. 1
Singh Article J096

Singh Article J096

Ratings: (0)|Views: 5 |Likes:

More info:

Published by: shedat_utopian872941 on Apr 14, 2010
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less

04/14/2010

pdf

text

original

 
Journal of Sound and Vibration
(1998)
213
(4), 673–708
IDENTIFICATION OFMULTI-DEGREE-OF-FREEDOM NON-LINEARSYSTEMS UNDER RANDOM EXCITATIONS BYTHE REVERSE PATHSPECTRAL METHOD
C. M. R
 
R. S

Acoustics and Dynamics Laboratory
,
Department of Mechanical Engineering
,
The Ohio State University
,
Columbus
,
OH 
43210-1107,
.
.
A
.(
Received 
11
June
1997,
and in final form
20
January
1998)Conventional frequency response estimation methods such as the ‘‘
1
’’ and ‘‘
2
’’methods often yield measured frequency response functions which are contaminated by thepresence of non-linearities and hence make it difficult to extract underlying linear systemproperties. To overcome this deficiency, a new spectral approach for identifyingmulti-degree-of-freedom non-linear systems is introduced which is based on a ‘‘reversepath’’ formulation as available in the literature for single-degree-of-freedom non-linearsystems. Certain modifications are made in this article for a multi-degree-of-freedom‘‘reverse path’’ formulation that utilizes multiple-input/multiple-output data fromnon-linear systems when excited by Gaussian random excitations. Conditioned ‘‘
c
1
’’ and‘‘
c
2
’’ frequency response estimates now yield the underlying linear properties withoutcontaminating effects from the non-linearities. Once the conditioned frequency responsefunctions have been estimated, the non-linearities, which are described by analyticalfunctions, are also identified by estimating the coefficients of these functions. Identificationof the local or distributed non-linearities which exist at or away from the excitationlocations is possible. The new spectral approach is successfully tested on several examplesystems which include a three-degree-of-freedom system with an asymmetric non-linearity,a three-degree-of-freedom system with distributed non-linearities and a five-degree-of-free-dom system with multiple non-linearities and multiple excitations.
7
1998 Academic Press Limited
1. INTRODUCTION
The properties of multi-degree-of-freedom linear systems are typically identified using timeor frequency domain modal parameter estimation techniques [1]. The frequency domaintechniques extract modal parameters from ‘‘
1
’’ and ‘‘
2
’’ estimated frequency responsefunctions in the presence of uncorrelated noise [2, 3]. However, if the system underidentification also possesses non-linearities, these conventional estimates often yieldcontaminated frequency response functions from which accurate modal parameters cannotbe determined [4, 5]. Such conventional methods are also incapable of identifying thenon-linearities.To accommodate the presence of non-linearities, several researchers have developedmethods to improve frequency domain analysis of non-linear systems [6–11]. For example,the functional Volterra series approach for estimating higher order frequency responsefunctions of non-linear systems has gained recognition [6]. This method has been used toestimate first and second order frequency response functions of a non-linear beamsubjected to random excitation [7], where curve fitting techniques were used for parametric
0022–460X/98/240673+36 $25.00/0/sv981522
7
1998 Academic Press Limited
 
.
.
  
.

674estimation of an analytical model. However, the method is very computationally intensiveand estimation of third and higher order frequency response functions has beenunsuccessful. To alleviate this problem, sinusoidal excitation was used to estimate only thediagonal second and third order frequency response functions of the Volterra series [8].Other higher order spectral techniques have also been employed for the analysis of non-linear systems [9]. For instance, the bi-coherence function has been used to detect thesecond order non-linear behaviour present in a system [10]. Also, the sub-harmonicresponses of a high speed rotor have been studied using bi-spectral and tri-spectraltechniques [11].An alternative approach has recently been developed by Bendat
et al 
. [12–15] forsingle-input/single-output systems which identifies a ‘‘reverse path’’ system model. Asimilar approach has been used for the identification of two-degree-of-freedom non-linearsystems where each response location is treated as a single-degree-of-freedom mechanicaloscillator [16]. Single-degree-of-freedom techniques are then used to identify systemparameters [17]. However, this approach requires excitations to be applied at everyresponse location and it also inhibits the use of preferred higher dimensional parameterestimation techniques that are commonly used for the modal analysis of linear systems [1].The literture review reveals that there is clearly a need for frequency domain systemidentification methods that can identify the parameters of non-linear mechanical andstructural systems. Also, improvements to the frequency response estimation methods suchas the ‘‘
1
’’ and ‘‘
2
’’ methods are necessary when measurements are made in the presenceof non-linearities. The primary purpose of this article is to introduce an enhancedmulti-degree-of-freedom spectral approach based on a ‘‘reverse path’’ system model.Additional discussion is included to justify the need for spectral conditioning andcomputational results are given to illustrate the performance on several non-linear systems.However, focus of this article is on the mathematical formulation for multi-degree-of-free-dom non-linear systems. Specific objectvies include the following: (1) to accommodate forthe presence of non-linearities so that improved estimates of the linear dynamic compliancefunctions can be determined from the input/output data of multi-degree-of-freedomnon-linear systems when excited by Gaussian random excitations; (2) to estimate theunderlying linear systems’ modal parameters from these linear dynamic compliancefunctions using higher dimensional modal analysis parameter estimation techniques;(3) to determine the coefficients of the analytical functions which describe local ordistributed non-linearities at or away from the locations where the excitations areapplied; (4) to assess the performance of this new method via three computationalexamples with polynomial type non-linearities. Comparison of this method with anexisting time domain method is in progress, and ongoing research is being conducted toconsider both correlated and uncorrelated noise. Issues such as the spectral variability of coefficient estimates as well as other errors are currently being examined and will beincluded in future articles. However, these issues have been omitted from this article sothat focus can be kept on introducing an analytical approach to multi-degree-of-freedomsystems.
2. PROBLEM FORMULATION
2.1.
 
The equations of motion of a discrete vibration system of dimension
with localizednon-linear springs and dampers can be described in terms of a linear operator
L
[
x
(
t
)] and
 
- 
675a non-linear operator
[
x
(
t
),
x˙ 
(
t
)]:
L
[
x
(
t
)]+
[
x
(
t
),
x˙ 
(
t
)]=
(
t
),
L
[
x
(
t
)]=
Mx¨
(
t
)+
Cx˙ 
(
t
)+
Kx
(
t
),
[
x
(
t
),
x˙ 
(
t
)]=
s
n j 
=1
A
 j 
y
 j 
(
t
), (1ac)where
M
,
C
and
K
are the mass, damping and stiffness matrices, respectively,
x
(
t
) is thegeneralized displacement vector and
(
t
) is the generalized force vector. Also refer to theappendix for the identification of symbols. The non-linear operator
[
x
(
t
),
x˙ 
(
t
)] containsonly the non-linear terms which describe the localized constraint forces and this operatoris written as the sum of 
n
unique non-linear function vectors
y
 j 
(
t
) representing each
thtype of non-linearity present (e.g., quadratic, cubic, fifth order, etc.). Considering onlynon-linear elastic forces, each
y
 j 
(
t
) is defined as
y
 j 
(
t
)=
{
D
x
k
(
t
)
m
 j 
}
, where
D
x
k
(
t
) is therelative displacement across the
k
th junction where the
th type of non-linearity exists, and
m
 j 
is the power of the
th type of non-linearity. These vectors
y
 j 
(
t
) are column vectors of length
q
 j 
, where
q
 j 
is the number of locations the
th type of non-linearity exists. Note thata single physical junction may contain more than one type of non-linearity (e.g., aquadratic and cubic); therefore, more than one
y
 j 
(
t
) is necessary to describe the non-linearconstraint force across that particular junction, as illustrated in the examples to follow.The coefficient matrices
A
 j 
contain the coefficients of the non-linear function vectors andare of size
by
q
 j 
. Inserting equations (1b) and (1c) into equation (1a), the non-linearequations of motion take the form
Mx¨
(
t
)+
Cx˙ 
(
t
)+
Kx
(
t
)+
s
n j 
=1
A
 j 
y
 j 
(
t
)=
(
t
). (2)From a system identification perspective, it is assumed that the types of non-linearities andtheir physical locations are known. Therefore the
n
non-linear function vectors
y
 j 
(
t
) canbe calculated; also, the coefficients of 
y
 j 
(
t
) can be placed in the proper element locationsof the coefficient matrices
A
 j 
. This assumption renders limitations on the practical use of this method since various types of non-linearities at each location are not always known.Therefore, research is currently being conducted to alleviate this limitation. However, itshould be noted that this restriction is currently true for any identification scheme whenapplied to practical non-linear systems.Consider several multi-degree-of-freedom non-linear systems as illustrated in Figure 1.The first example as shown in Figure 1(a) possesses an asymmetric quadratic–cubicnon-linear stiffness element which exists between the second and third masses and aGaussian random excitation is applied to the first mass:
 f 
e
23
(
t
)=
k
2
(
x
2
(
t
)
x
3
(
t
))+
a
2
(
x
2
(
t
)
x
3
(
t
))
2
+
b
2
(
x
2
(
t
)
x
3
(
t
))
3
,
(
t
)=[
1
(
t
) 0 0]
. (3a, b)Assuming that the form of the non-linear elastic force
e
23
(
t
) is known, the non-linearoperator
[
x
(
t
),
x˙ 
(
t
)], the non-linear functions (
y
1
(
t
) and
y
2
(
t
)) and their respectivecoefficient matrices (
A
1
and
A
2
) take the form
[
x
(
t
),
x˙ 
(
t
)]=
A
1
y
1
(
t
)+
A
2
y
2
(
t
),
y
1
(
t
)=(
x
2
(
t
)
x
3
(
t
))
2
,
 y
2
(
t
)=(
x
2
(
t
)
x
3
(
t
))
3
,
A
1
=(0
a
2
a
2
)
,
A
2
=(0
b
2
b
2
)
. (4a–e)

You're Reading a Free Preview

Download
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->