13th IFAC Symposium on System Identification

MEASUREMENT OF YOUNG’S MODULUS VIA MODAL ANALYSIS

EXPERIMENTS: A SYSTEM IDENTIFICATION APPROACH

R. Pintelon(1), P. Guillaume(2), K. De Belder(1), and Y. Rolain(1)

(1)Vrije Universiteit Brussel, Department ELEC, Pleinlaan 2, 1050 Brussel, Belgium

(2)
Vrije Universiteit Brussel, Department WERK, Pleinlaan 2, 1050 Brussel, Belgium

Abstract: The stress-strain relationship of linear visco-elastic materials is characterized

by a complex-valued, frequency dependent elastic modulus E( jω) (Young’s modulus).
Using system identification techniques it is shown in this paper how E( jω) can be
measured accurately in a broad frequency band from forced flexural (transverse) and
longitudinal vibration experiments on a beam under free-free boundary conditions. The
advantages of the proposed method are (i) it takes into account the disturbing noise and
the nonlinear distortions, (ii) E( jω) is delivered with an uncertainty bound, (iii) the low
sensitivity to non-idealities of the experimental set up, and (iv) the ability to measure
lowly damped materials. The approach is illustrated on plexiglass and copper. Copyright
© 2003 IFAC

Keywords: Young’s modulus, modal analysis, frequency domain.

1. INTRODUCTION mass (e.g. for measuring rubber, plastic foams and

The complex modulus E( jω) is a widely used and
powerful tool for describing the linear dynamic elastic
and damping properties of visco-elastic materials. For
example, knowledge of E( jω) in the audio frequency
range is needed by acousticians when developing and
characterizing new materials for noise and vibration
control. The idea of measuring the complex moduli
(Young’s, shear, …) from vibration tests (flexural,
longitudinal, and torsional) is not new and dates from
the early sixties (see the references in [1]). From those
early days till now a continuing attention has been
paid to this measurement problem in the literature:
see [1-6] and the references therein for the
longitudinal vibration tests, and [1], [7-10] and the
references therein for the flexural vibration tests. The
longitudinal (wave) experiments assume that the
material is homogeneous and isotropic [8-10], while
the flexural (transverse or bending) experiments can
also handle layered structures [8], [10]. In the wave
experiments the beam is sometimes loaded by an end
This work is supported by the Belgium National Fund for
Scientific Research; the Belgian Government (IUAPV/22);
and the Flemish Community (GOA-IMMI).
mineral wools) [3, 4], while in the bending
experiments the beam is often clamped at one [8-10]
or two [7] sides. These boundary conditions are
always non-ideal and should be avoided especially
when measuring stiff and lowly damped materials.
According to the way Young’s modulus is obtained
from the measured frequency response function
(FRF), the different approaches can be split into two
classes: the first class uses the poles (resonant
frequencies) of the FRF and the partial differential
equation (PDE) model only [1, 4, 8, 10], while the
second class uses the whole FRF and PDE model [5,
6, 9]. It is clear that the second class will be much
more sensitive to non-idealities of the experimental
set up such as misalignment creating other vibration
modes than those described by the PDE model. These
errors are easily eliminated in the first class by
selecting the appropriate resonances (see also Section
3). The advantage of the second class is the much
higher frequency resolution of the E( jω)
measurement. On the other hand, except for highly
damped materials, the resonant frequencies (poles) of
the FRF contain most information about the material
damping.

389

Except [6], none of the existing methods take into
account the disturbing noise when estimating Young’s
modulus, and no uncertainty bound is given. In [6],
Young’s modulus is determined using strain
measurements from longitudinal impulse response
experiments (hammer and air gun) on a beam under
free-free boundary conditions. Uncertainty bounds on
E( jω) are calculated assuming that the disturbing
noise is spatially and temporarily white.
This paper uses frequency domain system
identification techniques to measure the elastic
modulus E( jω) of homogenous isotropic materials in
a broad frequency band from forced flexural and
longitudinal vibration experiments on a beam under
free-free boundary conditions. Special designed
broadband periodic excitation signals are used for the
measurement procedure. It allows to calculate
simultaneously the frequency response function
(FRF), the noise level, and the level of the nonlinear
distortions. This information is used to estimate the
poles (and their uncertainty) of the measured FRF.
The final result is Young’s modulus together with its
uncertainty due to the disturbing noise and/or the
nonlinear distortions. The presented method takes
into account the disturbing noise on the output and the
input, the noise colouring, the correlation between the
input and output disturbances, and the non-linear
distortions.
The outline of the paper is as follows. First the
theoretical resonances of the longitudinal (Love
model) and the flexural (Euler-Bernoulli and
Timoshenko models) vibration experiments are
calculated (Section 2). Next the identification
procedure for estimating Young’s modulus is
explained step by step and illustrated on a wave
experiment (Section 3). Finally some practical aspects
of the experimental setup are discussed and a
comparison between the wave and the transverse
experiments is made (Section 4).
2. THEORETICAL RESONANCE FREQUENCIES
2.1. Longitudinal (wave) vibrations
Consider a longitudinal vibration experiment on a
uniform (along its length) beam composed of a linear,
homogeneous and isotropic visco-elastic material (see
Fig. 1). According to the Love model [2] the poles s k
of the force u(t) to longitudinal displacement y(t, x)
transfer function are related to Young’s modulus E by
Ls k 2 kπ ν(s k)J 2
E(s k) = – ρ  -------- 1 +  ---------------------
 kπ   L
with k = ± 1, ± 2, … . L is the length of the beam, ρ
the density of the material, ν(s k) Poisson’s
13th IFAC Symposium on System Identification
y(t, x)
u(t)
0 x L
Fig. 1: Longitudinal (wave) vibration experiment of
a beam with length L (side view): u(t) is the
applied force per unit area and y(t, x) the
longitudinal displacement at position x .
0 x x0 L
y(t, x)
u(t)
Fig. 2: Flexural (transverse or bending) vibration
experiment on a beam with length L (top
view): u(t) is the applied force and y(t, x) the
transverse displacement as a function of the
position x .
coefficient, and J the radius of gyration of the cross
section of the beam about the x -axis ( J = r ⁄ 2 for
a circle with radius r ; and J = ( h y2 + h z2 ) ⁄ 12 for a
rectangle with sides h y and h z [11]).
2.2. Flexural (transverse or bending) vibrations
Consider a flexural vibration experiment on a uniform
(along its length) beam composed of a linear,
homogeneous and isotropic visco-elastic material
without axial loads (see Fig. 2). According to the
Euler-Bernoulli model [12] the poles of the force u(t)
to the transverse displacement y(t, x) transfer
function are related to Young’s modulus E by
ρ AL 4 s k2
E(s k) = – ------------------
- (2)
I ζ k4
with k = ± 1, ± 2, … . I is the moment of inertia of
the cross-section of the beam about the z -axis (axis
perpendicular to the x - and y - axes; I = πr 4 ⁄ 4 for
a circle with radius r , and I = h z h y3 ⁄ 12 for a
rectangle with sides h z and h y in the z - and y -
directions respectively [11]), and ζ k the wave number
of the k th resonance frequency
cosh(ζ k)cos(ζ k) = 1 (3)
Numerical values of ζ k are tabulated in the literature
(see, for example, Table 6.4 of [12]):
(1) ζ 1 = 4.730041 , ζ 2 = 7.853205 , ζ 3 = 10.995608 ,

and ζ k = ( 2k + 1 )π ⁄ 2 for k ≥ 4 within a relative
–7
error smaller than 1.2 ×10 .
The Euler-Bernoulli theory leading to (2) ignores the
390
 13th IFAC Symposium on System Identification

effects of shear deformation and rotary inertia, and is Laplace variable s (result: Young’s modulus E( jω)
accurate for thin beams only [12]. In practice (2) can and its uncertainty in the frequency band of interest).
be used to model the first few resonance frequencies These four main steps are explained in detail in the
corresponding to the simple mode shapes. sequel of this section, and each step is illustrated on
Timoshenko’s theory [12] considers the effects of the longitudinal vibrations of a plexiglass beam.
shear deformation and rotary inertia, giving the
following implicit relationship between the poles s k 3.1. Measurement of the frequency response function
and Young’s modulus E The frequency response function (FRF) of the system
is measured using random phase multisines [14, 15].
cosh(b 1(s k)L)cos(b 2(s k)L) – 1 + These are periodic signals consisting of the sum of
harmonically related sine waves with user defined
b 22(s k) – b 12(s k) (4)
----------------------------------- sinh(b 1(s k)L)sin(b 2(s k)L) = 0 amplitudes and random phases. The measurement
2b 1(s k)b 2(s k) procedure of [16] is followed to estimate the FRF and
its uncertainty. It consists of the following basic steps:
where
1 Choose the amplitude spectrum and the
frequency resolution of the random phase
b i2(s) = ( – 1 ) ( i + 1 ) c(s) + c 2(s) + a(s) multisine.
2 Make a random choice of the phases of the
 ρ
 c(s) = ------------- ( 1 + γ (s) )s 2 (5)
random phase multisine, and calculate the
 2E(s) corresponding time signal.

ρ A 2 ρ 2 γ (s) 4
 a(s) = –  ------------s 3 Apply the excitation to the system and measure
+ --------------
-s P ≥ 2 periods of the steady state response u(t) ,
  E(s)I E 2(s) 
 y(t) .
4 Repeat steps 2 and 3 M ≥ 6 times.
A is the cross section area of the beam, and γ (s) 5 Calculate the DFT spectra of the input u(t) , and
depends on the shape of the cross section and output y(t) signals for each period of each
Poisson’s coefficient ν(s) ( γ (s) = ( 12 + 11 ν(s) ) ⁄ 5 experiment at the excited DFT frequencies.
for a rectangle, and γ (s) = ( 7 + 6 ν(s) ) ⁄ 3 for a circle
[13]). From these M × P sets of noisy input/output spectra,
one can calculate for each experiment the average
2.3. Discussion FRF Ĝ [ m ] over the P periods and its sample variance
Equations (1), (2) and (4) establish a relationship σ̂ Ĝ
2
[ m ] ( m = 1, …, M ). An additional averaging over

between the poles s k of the longitudinal and flexural m gives the final FRF Ĝ of the whole measurement
vibration experiments and Young’s modulus E(s k) procedure
when the beam ( L , A , I , J ) and the material ( ρ ,
ν(s) ) properties are known. Fortunately, the cross Ĝ [ m ] Ĝ [ m ] – Ĝ 2
∑m = 1 ----------- ∑m = 1 ---------------------------
M M
Ĝ = , σ̂ Ĝ
2 = (6)
section dimensions of the beam can always be chosen M M (M – 1)
such that the terms in (1) and (4) depending on
Poisson’s coefficient ν(s) are correction terms. together with its sample variance σ̂ Ĝ
2 . If the system is
Hence, only a rough guess of ν(s) is required. This is linear, then σ̂ Ĝ should be equal to the mean value of
2
the basic idea used for identifying Young’s modulus σ̂ Ĝ
2
[ m ] divided by M
E(s) in Section 3.
1
σ̂ Ĝ - ∑M σ̂ 2 [ m ]
2 = ------- (7)
n M 2 m = 1 Ĝ
3. THE IDENTIFICATION PROCEDURE

The procedure consists of the follow main steps. (i) (the extra factor M copes with the averaging over the
Choice of the broadband periodic excitation signals M experiments). Indeed, in the linear case the
and measurement of the frequency response function variability of the FRF measurement over the M
(result: FRF with its uncertainty). (ii) Approximation experiments with different excitation signals can be
of the measured FRF by a rational form in the Laplace explained by the disturbing input/output noise only. In
variable s (result: poles and their uncertainty). (iii) that case σ̂ Ĝ
2 (6) and σ̂ 2 (7) are both a measure of
Ĝ n
Selection of the poles of the rational form the influence of the disturbing input/output noise on
corresponding to the longitudinal or flexural vibration the FRF. If σ̂ Ĝ
2 (6) is larger than σ̂ 2 (7), then this is
Ĝ n
modes and calculation of Young’s modulus E(s k) via an indication that the systems behaves nonlinearly.
eq. (1), (2) or (4) (result: Young’s modulus E(s k) and Indeed, the difference
its uncertainty at the values of the poles s k ). (iv)
Approximation of E(s k) by a rational form in the σ̂ ĜS
2 = σ̂ 2 – σ̂ 2
Ĝ Ĝ n
(8)

391
 13th IFAC Symposium on System Identification

40 40 40

0 0 0

Amplitude (dB)
Amplitude (dB)

Amplitude (dB)
-40 -40 -40

-80 -80 -80

-120 -120 -120

0 1000 2000 3000 4000 0 1000 2000 3000 4000 0 1000 2000 3000 4000
frequency (Hz) frequency (Hz) frequency (Hz)

Fig. 3: Force-to-acceleration FRF (bold black line)
and its standard deviation (thin black line: (7)
noise errors only, and gray line: (6) noise errors
+ nonlinear distortions) of the longitudinal
vibration experiments. Left: plexiglass
( L = 1.983 m ), right: copper ( L = 2.209 m ).
40
Fig. 5: Comparison between the measured and
modelled FRF of the longitudinal vibration
experiment on the plexiglass beam. Bold black
line: Ĝ( jω) and G( jω, θ̂) , gray line: the 95%
confidence bound of the FRF measurement
( 3σ̂ Ĝ ), and black line: the complex error
Ĝ( jω) – G( jω, θ̂) .
40

0 0
estimate θ̂ together with its covariance matrix
Cov(θ̂) . Finally, from θ̂ and Cov(θ̂) the poles and
Amplitude (dB)
Amplitude (dB)

-40
-40
their uncertainty are calculated.
-80 -80

-120
0 1000 2000 3000
frequency (Hz)
4000
Fig. 4: Force-to-velocity FRF (bold black line) and
its standard deviation (6) (gray line) of the
flexural vibration experiments. Left: plexiglass
( L = 1.372 m ), right: copper ( L = 2.209 m ).
quantifies the contribution of the stochastic non-linear
distortions to the FRF measurement; while the non-
linear bias contributions G B are bounded by
1 the FRF. Next the index k of each observed
--- Mσ̂ ≤ G B ≤ γ Mσ̂
γ ĜS ĜS
where γ typically lies between two and ten (see [15,
17 and 18] for the details).
The measurement procedure is illustrated in Figures 3
and 4. For each measurement P = 10 and M = 25 .
It can be seen that the stochastic nonlinear distortions
are mostly dominant, and that the signal-to-noise ratio
Ĝ ⁄ σ̂ Ĝ of the FRF measurement (6) varies between
40 to 60 dB.
3.2. Approximation of the FRF by a rational form
According to the Mittag-Leffler theorem [19] the
infinite dimensional transfer functions shown in
Figures 3 and 4 can be approximated arbitrary well by
a rational form of finite order in a particular frequency
band
B(s, θ)
G(s, θ) = ---------------- =
A(s, θ)
The parameters θ of the rational form are found by
minimizing the sample maximum likelihood cost
function (see [15] for the details). The result is an
-120
Fig. 5 illustrates the procedure on the longitudinal
0 1000 2000 3000 4000
frequency (Hz) vibration experiment of a plexiglass beam. Using a
rational form (10) of order n a = n b = 34 , the
approximation error is about at the level of the
uncertainty of the FRF measurement (between -50 dB
and -60 dB).
3.3. Calculation of Young’s modulus at the poles
The first step consists in selecting the poles s k of
G(s, θ̂) (10) corresponding to the longitudinal or
flexural vibration modes. This can easily be done by
comparing s k ⁄ ( 2π ) to the resonance frequencies of
(9)
resonance peak (pole) is determined. This is done by
comparing the first few peaks of the measured FRF to
the resonance frequencies predicted using eq. (1), (2)
or (4), the beam parameters, and a rough guess of
Young’s modulus and Poisson’s coefficient. Finally
Young’s modulus is calculated at the values of the
poles s k via eq. (1), (2) or (4), where (1) and (4) need
a rough guess of ν(s k) ( ν = 0.33 for plexiglass and
ν = 0.3 for copper). By an appropriate choice of the
beam cross section dimensions, the correction term in
(1) depending on ν(s k) can often be neglected.
Equation (4) is a nonlinear algebraic equation in
E(s k) , and is solved via the Newton-Raphson root
finding algorithm [20]. As starting value we use the
Euler solution (2) or the previous solution E(s k – 1) of
(4). Uncertainty bounds on E(s k) are obtained from
the uncertainty of the poles through a first order
sensitivity analysis of eq. (1), (2) and (4).
Fig. 6 gives the result for the longitudinal vibration
nb
∑ n=0 n
b sn experiments. The following standard deviations are
---------------------------- (10) found
na
∑ n=0 n
a sn
5
plexiglas: std(E(s k)) = 1 ×10 N/m 2
6
(11)
copper: std(E(s k)) = 1 ×10 N/m 2

392
 13th IFAC Symposium on System Identification

3
E
194.8 195
DI
E E E
4
I II I I IEI IEI I EI
E E IE E DIDIDI

Amplitude (dB)
Amplitude (dB)
E E
III DIEDI
IEIII D D
194.6 2.5 194.5 3

Phase (°)

Phase (°)
E E DI DIDIDIEDIDI DIEDIDI DI
E E E D EDI DIEDI DIEDI DI EDI
194.4
E 2 E 194 IIDDDDDD D DDDD D 2 DI
IDIDD
D DD
DD
DI
194.2 1.5 193.5 1
0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4
|poles|/(2π) (kHz) |poles|/(2π) (kHz) |poles|/(2π) (kHz) |poles|/(2π) (kHz)
221.49 0.04 222 0.1

E E
Amplitude (dB)

Amplitude (dB)
E E I EIIIIIEI I I DI
DIDIDIDIDIDIDIDIDIDIDIIIIIIIII I
0.03

Phase (°)
221.5

Phase (°)
E E DDDDDD DI
221.48
E E E E DDDDDD
0.05
DIDIDIDIDIDIDIDIE DI
DDDDD DI DIDIDIDIDIDIEDIDIDIDIDIDIDIDIDIDI DI DI
0.02 221
D E E
221.47 0.01 220.5 0
0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4
|poles|/(2π) (kHz) |poles|/(2π) (kHz) |poles|/(2π) (kHz) |poles|/(2π) (kHz)

Fig. 6: Comparison between the measured and the Fig. 7: Comparison between the elastic moduli (N/
modelled elastic modulus (N/m2) at the poles s k m2) obtained from the longitudinal (o: Love
of the longitudinal vibration experiments: model (1)) and the flexural (+: Euler model
plexiglass (row 1), and copper (row 2). Circles: (2), x: Timoshenko model (4)) vibration
measurement E(s k) (1), and solid line: model experiments.
E(s k, θ̂) (row 1 (10), row 2 (12))

perfectly by (10) or (12).

3.4. Modelling of Young’s modulus
In Section 3.3 Young’s modulus E and its variance
var(E) are obtained at the value of the poles of the 4. EXPERIMENTAL RESULTS
longitudinal or flexural vibration modes. In practise
Young’s modulus should be known along the jω - The following test beams are used for the
axis, which requires an additional modelling step. experiments: plexiglass ( ρ = 1200 kg/m 3 ,
Based on the spring-dashpot representations of linear rectangular cross section with sides h y = 10.3 mm
visco-elastic materials [21] Young’s modulus is and h z = 20.0 mm , L = 1.372 m and
modelled as a rational from E(s, θ) (10). For metals L = 1.983 m ), and copper ( ρ = 8900 kg/m 3 ,
Young’s modulus is modelled as a complex, rectangular cross section with sides h y = 10.0 mm
frequency independent constant and h z = 30.0 mm , L = 2.209 m ). The beams are
hung by two or three nylon threads perpendicular to
the excitation direction. They are excited in the
E(s, θ) = R + Ij (12)
longitudinal or transverse direction by a mini-shaker.
A stinger rod connects the mini-shaker to the
The parameters θ in (10) and (12) are found by
impedance head, which is attached as close as
minimizing the following cost function w.r.t. θ
possible to the test specimen (distance less than 7
mm).
K
E(s k) – E(s k, θ) 2
∑ ------------------------------------------
var(E(s k))
(13) The results of the longitudinal and flexural vibration
k=1 experiments are compared in Fig. 7. It can be seen
that the amplitude of Young’s modulus obtained from
The procedure is illustrated in Fig. 6 on the the Timoshenko (4) and Love (1) models coincide
longitudinal vibration experiments shown in Fig. 3 (errors of less than 0.1 dB = 1%), while that of the
For plexiglass a rational form (10) with Euler model (2) is accurate for the first few (five)
n a = n b = 2 is sufficient to model the elastic values only. The phase of Young’s modulus matches
modulus. The poles and zeroes of the estimated very well for the three models. This can be explained
second order model have multiplicity one, lie on the by the fact that Timshenko’s solution (4) does almost
negative axis, and are alternating. This proves that not change the phase of the poles. Note, however, that
E(s, θ̂) can be represented by an RC-network [22] or the differences between the Timoshenko and Love
an equivalent spring-dashpot scheme [21]. Although models are larger than the uncertainty of the E(s k)
models (10) and (12) are quite accurate (model errors measurement (compare eq. (11) with Fig. 7). For
less than 0.05 dB = 5‰ for plexiglass, and less than plexiglass this can be due to the variability of the
0.008 dB = 0.8‰ for copper), the model errors are material properties over the two test beams, and/or a
significantly larger than the uncertainty of E(s k) temperature difference between the two experiments;
(compare eq. (11) and Fig. 6). It can be due to the while for the copper beam it can be due to non-
non-idealities of the experimental set up and the test isotropy caused by the rolling process (the side walls
specimen (non-homogeneity of the material, non- are stiffer).
uniform cross section beam, …), and/or the fact that
the true material behaviour cannot be described

393

5. CONCLUSION
A system identification approach for measuring the
elastic modulus of homogeneous visco-elastic
materials from longitudinal and flexural vibration
experiments has been presented. The final result is a
parametric model for Young’s modulus over a broad
frequency band together with an uncertainty bound.
The uncertainty bound takes into account the
disturbing input/output (measurement) noise and the
stochastic nonlinear distortions. As such it is a
measure for the repeatability of the experiment over
the class of random excitation signals with the same
power spectrum. It can be used to study the influence
(non-idealities) of the measurement set up, the
temperature, the nonlinear material behaviour, the
variability of the material over different test
specimen, the value of Poisson’s coefficient …
Note that the same ideas could be used to measure the
shear modulus G( jω) from torsional vibration
experiments.
REFERENCES
[1] ASTM Standard E 1876-99 (1999). Standard
Test Method for Dynamic Young’s Modulus,
Shear Modulus, and Poisson’s Ratio by Impulse
Excitation of Vibration, Annual book of ASTM
Standards, American Society for Testing and
Materials: West Conshohocken (PA).
[2] T. Pritz (1981). Apparent complex Young’s
modulus of a longitudinally vibrating
viscoelastic rod, Journal of Sound and Vibration,
vol. 77, no. 1, pp. 93-100.
[3] T. Pritz (1982). Transfer function method for
investigating the complex modulus of acoustic
materials: rod-like specimen, Journal of Sound
and Vibration, vol. 81, no. 3, pp. 359-376.
[4] T. Pritz (1994). Dynamic Young’s modulus and
loss factor of plastic foams for impact sound
isolation, Journal of Sound and Vibration, vol.
178, no. 3, pp. 315-322.
[5] L. Hillström, M. Mossberg, and B. Lundberg
(2000). Identification of complex modulus from
measured strains on an axially impacted bar
using least squares, Journal of Sound and
Vibration, vol. 230, no. 3, pp. 689-707.
[6] M. Mossberg, L. Hillström, and T. Söderström
(2001). Non-parametric identification of
viscoelastic materials from wave propagation
experiments, Automatica, vol. 37, no. 4, pp. 511-
521.
[7] T. Miyazaki, J. Muroi, Y. Yamamoto, and T.
Uyemura (1994). Measurement of complex
elastic modulus by holography - effect of
moment caused under incorrect driving
conditions, Int. J. Japan Soc. Prec. Eng., vol 28.
no. 3, pp. 243-248.
394
13th IFAC Symposium on System Identification
[8] Y. C. Xu, and A. Nashif (1996). Measurement,
analysis and modeling of the dynamic properties
of materials, Sound and Vibration, vol. 30, no. 7,
pp. 20-23.
[9] R. Caracciolo, A. Gasparetto, and M.
Giovagnoni (2000). Measurement of the
isotropic dynamic Young’s modulus in a
seismically excited cantilever beam using a laser
sensor, Journal of Sound and Vibration, vol. 231,
no. 5, pp. 1339-1353.
[10] C. L. Sisemore and C. M. Darvennes (2002).
Transverse vibration of elastic-viscoelastic-
elastic sandwich beams: compression-
experimental and analytical study, Journal of
Sound and Vibration, vol. 252, no. 1, pp. 155-
167.
[11] R. D. Blevins (2001). Formulas for Natural
Frequency and Mode Shape, Krieger Publishing
Compagny, Malabar (Fl., USA).
[12] D. J. Inman (1996). Engineering Vibration,
Prentice Hall, Englewood Cliffs (N.J., USA).
[13] G. R. Cowper (1966). The shear coefficient in
Timoshenko’s beam theory, Transactions of the
ASME: Journal of Applied Mechanics, vol. 33,
pp. 335-340.
[14] J. Schoukens, T. Dobrowiecki and R. Pintelon
(2002). Linear modeling in the presence of
nonlinear distortions, IEEE Trans. Instrum. and
Meas., vol. 51, no. 4, pp. ?-?.
[15] R. Pintelon, and J. Schoukens (2001). System
Identification: A Frequency Domain Approach,
IEEE Press, Piscataway (N.J., USA).
[16] R. Pintelon, J. Schoukens, W. Van Moer and Y.
Rolain (2001). Identification of linear systems in
the presence of nonlinear distortions, IEEE
Trans. Instrum. and Meas., vol. 50, no. 4, pp.
855-863.
[17] J. Schoukens, T. Dobrowiecki, and R. Pintelon
(1998). Parametric Identification of Linear
Systems in the Presence of Nonlinear
Distortions. A Frequency Domain Approach,
IEEE Trans. Autom. Contr., vol. AC-43, no. 2,
pp. 176-190.
[18] R. Pintelon, and J. Schoukens (2002).
Measurement and modeling of linear systems in
the presence of non-linear distortions. MSSP, vol.
16, no. 5, pp. 785-801.
[19] P. Henrici (1974). Applied and Computational
Complex Analysis, vol. 1, John Wiley & Sons,
New York (USA).
[20] Fletcher, R. (1991). Practical Methods of
Optimization (2nd ed.), Wiley, New York (USA).
[21] W. Flügge (1967). Viscoelasticity, Blaisdell
Publishing Compagny, London (UK).
[22] N. Balabanian (1964). Network Synthesis,
Prentice-Hall, Englewood Cliffs (USA).