MT Lectures 11-20

Modal Testing
(Lecture 11)
Dr. Hamid Ahmadian

School of Mechanical Engineering
Iran University of Science and Technology
ahmadian@iust.ac.ir
Response Function
Measurement Techniques
Introduction
Test Planning
Basic Measurement System
Structure Preparation
Excitation of the structure
Response Function Measurement

Techniques
IUST ,Modal Testing Lab ,Dr H Ahmadian
Introduction
The measurements techniques used for

modal testing are discussed:
Response measurement only

Force and response measurement
The 2nd type of measurement

techniques is of our concern:
Single-point excitation( SISO/SIMO)

Multi-point excitation (MIMO)

Techniques
Test Planning
Objective of the test
Level
Levels according to Dynamic Testing Agency:

Natural
Freq
Damping
ratio
Mode Shapes
Usabe for
validation
Out of range
residues
Updating
0
1
Only in few
points
2
3
4
Techniques
Test Planning
Extensive test planning is required

before full-scale measurement:
Method of excitation
Signal processing and data analysis
Proper selection of pickup points
Excitation location
Suspension method

Techniques
Quality of measured data
Signal quality
Signal fidelity
Sufficient strength and clarity/noise free

No cross sensitivity
Measurement repeatability
Measurement reliability
Measurement data consistency,
including reciprocity

Techniques
An excitation mechanism
A transduction mechanism
An Analyzer

Techniques
Source of excitation
signal:
Sinusoidal
Periodic (with specific
freq. content)
Random
Transient
Transducers
Condition Amplifiers
Analyzers
Power Amplifier
Exciter

Techniques
Structure Preparation
Free Supports
Grounded Support
Loaded Support
Perturbed Support

Techniques
Free Supports
Theoretically the structure will possess 6 rigid

body modes @ 0 Hz.
In practice this is provided by a soft support
Rigid body modes are less then 10% of strain
modes
Suspending from nodal points for minimum

interference
The suspension adds significant damping to the
lightly damped structures

Techniques
Free Supports
Suspension wires,
should be normal to
the primary
vibration direction
The mass and
inertia properties
can be determined
from the RBMs.

Techniques
Free Supports

Techniques
Free Supports

Techniques
Grounded Support
The structure is fixed to the ground at selected

points.
The base must be sufficiently rigid to provide
necessary grounding.
Usually is employed for large structures
Another application is simulating the operational

condition
Parts of power generation station

Civil engineering structures
Turbine Blade
Static stiffness can be obtained from low frequency

mobility measurements.

Techniques
Loaded Support
The structure is connected to a simple

component with known mobility
A specific mass
The effect of added mass can be removed

analytically
More modes are excited in a certain
frequency range compared to free suspension
The modes of structure are quite different

Techniques
Perturbed Support
The data base for the structure can be

extended by repetition of modal tests for
different boundary conditions

Techniques
Perturbed Support

Techniques
Perturbed Support

Techniques
Excitation of the structure
Various devices are available for

exciting the structure:
Contacting
Mechanical (Out-of-balance rotating masses)

Electromagnetic (Moving coil in magnetic field)
Electrohydraulic
Non-Contacting
Magnetic excitation

Techniques
Electromagnetic Exciters
Supplied input to the shaker is converted to

an alternating magnetic field acting on a
moving coil.

Techniques
Electromagnetic Exciters
There is a small difference between the force

generated by the shaker and the applied
force to the structure
The force required to accelerate the shaker

moving
The force required to excite the structure

sharply reduces near the resonance point,
Much smaller than the generated force in the

shaker and the inertia of the drive rod
Vulnerable to noise or distortion

Techniques
Attachment to the structure
Push rod or stingers:
Applying force in
only one direction
Flexible drive
rod/stinger
introduces its own
resonance into the
measurement.

Techniques
Support of shakers

Techniques
Support of shakers

Techniques
Hammer or Impactor
Excitation

Techniques
Other excitation methods
Step Relaxation/sudden release

Charge/Explosive impactor
.

Techniques
Moving Support
Corresponds to
grounded model
Only responses are
measured
When the mass
properties are known,
the modal properties can
be calculated from
measured data

Techniques
Moving Support

Techniques
Modal Testing
(Lecture 12)
Dr. Hamid Ahmadian

ahmadian@iust.ac.ir
Digital Signal Processing
Introduction
Basics of Discrete Fourier Transform
(DFT)
Aliasing
Leakage
Windowing
Filtering
Improving Resolution
FRF Measurement Techniques
Introduction
The measured force or

accelerometer signals
are in time domain.
The signals are
digitized by an A/D
converter
And recorded as a set
of N discrete values
evenly spaced in the
period T
Basics of DFT
The spectral properties of the recorded

signal can be obtained using Discrete
Fourier Transform/Series (DFT/DFS):
The DFT assumes the signal x(t) is periodic

In the DFT there are just a discrete
number of items of data in either form
There are just N values xk

The Fourier Series is described by just N values
Basics of DFT
x (t ) = x (t + T )
a0
x (t ) = + an cos( n t ) + bn sin( n t )
2 n =1
2n
,
n =
T
2 T
an = x (t ) cos( n t )dt
T 0
2 T
bn = x (t ) sin( n t )dt
T 0
or
x (t ) =
X e
i
n =
nt
1 T
i n t
x
t
e
dt
(
)
T 0
= X n*
Xn =
X n
xk =
N 1
2
a0
2nk
2nk
) + bn sin(
)
+ an cos(
2 n =1
N
N
N 1
2nk
)
N
2
an =
N
xk cos(
2
bn =
N
or
2nk
)
xk sin(
N
k =0
k =0
N 1
N 1
xk = X n e 2ink / N
n =0
1
Xn =
N
N 1
2ink / N
,
x
e
X
=
X
k
N r
r
k =0
Basics of DFT
1
Xn =
N
N 1
x e
k =0
2ink / N
X 0 L x0
x
X
1 N L O 1
M
M
K
K
M
M
X N 1 K K N x N 1
Basics of DFT
The sampling frequency:

1 N
2 2N
fs = = s =
=
ts
ts
Nyquist Frequency
The range of frequency spectrum:

s N
fs
f max =
= max =
=
T
2
2
The resolution of frequency spectrum:
1
2
f = , =
T
T
Basics of DFT
There are a number of features of DF

analysis which if not properly treated,
can give rise to erroneous results:
Aliasing
Mis-interoperating a high frequency component

as a low frequency one
Leakage
Periodicity of the signal
Aliasing
Digitizing a low
frequency signal
produces exactly the
same set of discrete
values as result from
the same process
applied to a higher
frequency signal
<
s
2
s
Aliasing
Compare :
k
k
sin(2p ) sin(2 ( N p ) N )
N
2pk
sin(2k
)
N
2pk
)
sin(
N
p<N
2
Aliasing
Aliasing
The solution to the

problem is to use an
anti-aliasing filter
Subjecting the original

signal to low pass with
sharp filter
Filters have a finite cutoff rate; it is necessary to
reject the spectral range
near Nayquist frequency
> (08 1.0) s

2
Leakage
A direct consequence of
taking a finite length of
time history coupled with
assumption of periodicity
Energy is leaked into a
number of spectral lines
close to the true
frequency.
Leakage
Leakage
To avoid the leakage there are number

of scenarios:
Increasing the record time T

Windowing
Multiply the time record by a function that is

zero at the ends of the time record and large in
the middle, the FFT content is concentrated on
the middle of the time record
Windowing
Windowing involves the imposition of a

prescribed profile on the time signal prior to
performing the FT
x(t ) = w(t ) x(t )
a0 a1 cos(0t ) + a1 cos(0t ) + a2 cos(20t )
0 < t < T,
w(t ) = a3 cos(30t ) + a4 cos(40t )
0
elsewhere
0 = 2 T
Windowing
Windowing
Windowing
Function
a0
a1
a2
a3
a4
Rectangular
Hanning
KaserBessel
Flat top
1.298 0.244 0.003
1.933 1.286 0.388 0.032

Windowing
Windowing
Improving Resolution (Zoom)
There arises limitations of inadequate

frequency resolution
at the lower end of the frequency range

For lightly-damped systems
A common solution is to concentrate all

spectral lines into a narrow band
Within fmin-fmax
Instead of 0-fmax
Zoom
Method 1:
Shifting the frequency origin of the spectrum
x(t ) = A sin(t )
x(t ) = A sin(t ) cos(min t )
A
[sin( min )t + sin( + min )t ]
2
The modified signal is then analysed in the
range of 0-(fmax-fmin)
Zoom
Method 2:
A controlled aliasing
effect
Applying a band pass

filter
Because of the
aliasing phenomenon,
the frequency
component between
f1 and f2 will appear
aliased between 0-(f2f1)
Modal Testing
(Lecture 13)
Dr. Hamid Ahmadian

ahmadian@iust.ac.ir
Use of Different Excitation

Signals
Introduction
Stepped-Sine Testing
Slow Sine Sweep Testing
Periodic Excitation
Random Excitation
Transient Excitation
Introduction
There are three different classes of

excitation signals used:
Periodic
Transient
Random
Introduction
Periodic:
Stepped sine
Slow sine sweep
Periodic
Pseudo-random
Introduction
Transient:
Burst sine
Burst random
Chirp
Impulse
Introduction
Random:
(true) random
White noise
Classical method of FRF measurement

To encompass a frequency range of
interest, the command signal frequency is
stepped from one frequency to another
The excitation/response(s) are measured

(amplitudes and phase(s)) .
It is necessary to ensure that the steady-state
condition is attained before the measurement.
The extent of unwanted transient

response depends on:
Proximity of excitation frequency to a

natural frequency,
The abruptness of the changeover from
the previous command signal to the new
one,
The lightness of the damping of nearby
modes.
An advantage
of stepped-sine
testing is the
facility of taking
measurement
where and as
they are
required.
No. point
between
HPPs
Largest Error
%
dB
30
10
0.5
0.2
0.1
Involves the use of a sweep oscillator
Provides a sinusoidal signal

Its frequency is varied slowly but
continuously
If an excessive sweep rate is used then

distortions of FRF plot are introduced
One way of
checking the
suitability of a
sweep rate is to
make the
measurement
twice:
Once sweeping up
And the 2nd time
sweeping down
It is possible to
prescribe an
optimum sweep rate
for a given structure
taking into account
its damping levels
Recommended sweep rate:
ISO prescribes
maximum linear and
log sweep rate
through a resonance
as: Linear
Smax < 216 ( r r ) 2 Hz / min

Log
Smax < 310 ( r )
2
r
Octaves / min
Periodic Excitation
A natural extension of the sine wave

test methods:
To use a complex periodic input signal

which contains all the frequencies of
interest,
The DFT of both input and output signals
are computed and the ratio of these gives
the FRF
Both signal have the same frequency
contents
Periodic Excitation
Two types of periodic signals are used:
A deterministic signal (square wave)
Some frequency components are inevitably

weak.
Pseudo-Random type of signal
The frequency components may be adjusted to

suit a particular requirements-such as equal
energy at each frequency,
Its period is exactly equal to the sampling time
resulting zero leakage .
Random Excitation
S xx ( ) = H ( ) S ff ( )
H1 ( ) =
S fx ( ) = H ( ) S ff ( )
S xx ( ) = H ( ) S xf ( )
S fx ( )
S ff ( )
S xx ( )
H 2 ( ) =
S xf ( )
H1 ( )
=
H 2 ( )
2
Random Excitation
There may be noise on one of the two

signals
Near resonance this is likely to influence

the force signal
At anti-resonances it is the response signal
which will suffer
Random Excitation
H2 might be a better indication near

resonances while H1 is a better indication
near anti-resonances:
S fx ( )
S xx ( ) + Smm ( )
H1 ( ) =
, H 2 ( ) =
S ff ( ) + Snn ( )
S xf ( )
Auto-spectra of noise on the output signal
Auto-spectra of noise on the input signal
Random Excitation
A closer optimum formula for the FRF is

defined as the geometric mean of the two
standard estimates
Phase is identical to that in the two basic

estimates
H v ( ) = H1 ( ) H 2 ( )
Random Excitation
Typical measurement made using random

excitation:
Random Excitation
Details from previous plot around a

resonance:
H2
H1
Random Excitation
Use of zoom spectrum analysis:
Improving the resolution removes the major

source of low coherence
Random Excitation
Effect of averaging:
The excitation
and the
response are
contained
within the
single
measurement
Burst excitation signals:
A short section of a continuous signal (sin,

random, ) followed by a period of zero wave.
Chirp excitation:
The spectrum can be strictly controlled to be such

within frequency range of interest
Impulsive excitation
by Hammer:
Different impulsive
excitations
Signals and spectra
for double hit case
Impulsive excitation by Shaker:
Modal Testing
(Lecture 14)
Dr. Hamid Ahmadian

ahmadian@iust.ac.ir
RESPONSE FUNCTION
MEASUREMENT TECHNIQUES
3.9 Calibration
3.10 Mass Cancellation
3.11 Rotational FRF Measurement
3.12 Measurement on Nonlinear
Structures
Effects of Different Excitations

Level Control in FRF Measurement
Calibration
In all measurement systems it is

necessary to calibrate the equipment.
There should be two levels of
calibration:
Absolute calibration of individual

transducers
The overall sensitivity of instrumentation
system
Calibration
The overall system

calibration
The scale factor should

be checked against
computed factor using
manufacturers stated
sensitivity
Should be carried out
before & after each test
Mass Cancellation
Near resonance the actual applied force

becomes very small and is thus very prone to
inaccuracy.
Some applied mass is used to move
additional transducer mass
fT = f M m &x&
X&&
T =
required
FT
X&&
M =
measured
FM
Mass Cancellation
Added mass to be
cancelled and the
typical analogue
circuit
At deriving point a
relation between
measured and
required FRFs can
be obtained
Mass Cancellation
Re( FT ) = Re( FM ) m Re( X&& )
Im( F ) = Im( F ) m Im( X&& )
or
Re(1 / T ) = Re(1 / M ) m
Im(1 / T ) = Im(1 / M )
Rotational FRF Measurement
Measurement of rotational FRFs using two or

more transducers:
x A + xB
xo =
2
A + B
o =
L
Rotational FRF Measurement
Application of moment excitation
X X
, , ,
F M F M
Measurement on Nonlinear
Structures
Many structures, especially in vicinity of

resonances, behave in a nonlinear way:
Natural frequency varies with position

and strength of excitation
Distorted frequency responses (near
resonances)
Unstable or unrepeatable data
Measurement on Nonlinear
Structures
Examples of
different nonlinear
system response for
different excitation
levels
Softening effect
Increase in damping
FRF measurement
on nonlinear
system:
Sinusoidal Excitation
Random Excitation
Compatible with
theory
Linearized system
Most types of nonlinearity are amplitude

dependent:
A linearized behaviour is observed when

the response level is kept constant
The obtained linear model is valid for that
particular vibration level
Level Control in FRF

Measurement
Response level
control,
Best linear representation

(nonlinearities are
displacement dependent)
Force level control

Or no level control

Measurement
Inverse FRF plots for a SDOF
Real part is expected to be liner wrt frequency squared

Imaginary part should be linear/constant
Any deviation from the expected behaviour can be detected
as nonlinearity in the system

Measurement
Use of Hilbert transform to detect non-linearity
The Hilbert transform express the relations between

real and imaginary parts of the Fourier Transform
Notes: Hilbert Transform
The Hilbert transform express the

relations between real and imaginary
parts of the Fourier Transform
Fourier Transform is considered to map

functions of time to functions of frequency
and vice versa
Hilbert transform map functions of time or
frequency to the same domain
For causal functions:
g (t ) = g even (t ) + g odd (t ),
g ( t ) / 2, t > 0
g even (t ) =
g ( t ) / 2, t < 0
g ( t ) / 2, t > 0
g odd (t ) =
g ( t ) / 2, t < 0

Re G ( ) = {g even (t )} = {g odd (t ) sign(t )},
Im G ( ) = {g odd (t )} = {g even (t ) sign(t )},
Since {sign(t )} =
based on convolution theorm :
Re G ( ) = i Im G ( )
Im G ( ) = Re G ( )
.
Modal Testing
(Lecture 15)
Dr. Hamid Ahmadian

ahmadian@iust.ac.ir
Modal Parameter Extraction
Introduction
Preliminary checks of FRF data
Visual checks
Assessment of multiple-FRF data set using SVD
Mode indicator functions
SDOF modal analysis methods
Peak amplitude method

Circle fit method
Inverse or line fit method
Modal Parameter Extraction Methods
Introduction
Some of the many available procedures for

fitting a model to the measured data are
discussed:
Their various advantages and limitations are

explained,
No single method is best for all cases.
This phase of the modal test procedure is

often called modal parameter extraction
or modal analysis
Introduction
Types of modal analysis:
Frequency domain (of FRFs)

Time domain (of Impulse Response
Function)
The analysis will be performed using
SDOF methods, and

MODF methods.
Introduction
Another classification of methods

relates to the number of FRFs used in
the analysis:
Single-FRF methods, and

Multi-FRF methods:
Global methods which deals with SIMO data

sets
and Polyreference which deals with MIMO data
Introduction
Difficulty due to damping:
In practice we are obliged to make certain

assumption about the damping model,
Significant errors can be incurred in the
modal parameter estimates as a result of
conflict between assumed and actual
damping effects.
Decision on the issue of real and complex
modes.
Preliminary checks of FRF

data
Low-frequency asymptotes,
High-frequency asymptotes,
Stiffness-like characteristics for grounded

structures
Mass-line asymptotes for free structures
Mass line or stiffness line
Incidence of antiresonances
For a point FRF there must be a resonance

after each antiresonance
Preliminary checks of FRF

data
Mode Indicator Functions:
The Peak-Picking Method
Sum of amplitudes of all measured FRFs to

locate the resonance points
The frequency-domain decomposition

method
Defined by the SVD of the FRF matrix
Case Study: MODES OF A

RAILWAY VEHICLE
Case Study: Test set-up
Case Study: Sensor Locations
Case Study: Sensor Locations
Case Study: Excitation
Case Study: Excitation
Case Study: Measurements
The Peak-Picking Method
Sum of amplitudes of all measured FRFs to

locate the resonance points
# Mode
(Hz)
Frequency
2.67
4.67
4
5.33
8.33
12.33
13.33
14.00
16.00
24.67
The frequency-domain
decomposition method
A more advanced method consists of

computing the Singular Value
Decomposition of the spectrum matrix.
The method is based on the fact that
the transfer function or spectrum matrix
evaluated at a certain frequency is only
determined by neighboring modes.
The frequency-domain
decomposition method
[H ( )] = [{H11 ( )} {H 21 ( )}
[H ( )] = [U ( )][( )][V ( )]T
T
[MIF ( )] = [( )] [( )]
{H
np
( )}
SDOF modal analysis

methods
The SDOF assumption

N
jk ( ) =
s =1
jk ( ) =
jk ( ) =
A jk
s2 + i s s2
r
A jk
+ i r
2
r
2
r
A jk
+ i r
2
r
2
r
+
s =1
sr
A jk
s2 + i ss2
+ r B jk
SDOF modal analysis

methods
SDOF modal analysis methods
Peak amplitude method

Circle fit method
Inverse or line fit method
SDOF modal analysis

methods: Peak Amplitude
Individual resonance peaks are

detected from the FRF
The frequency of the maximum responses

is takes as the natural frequency of that
mode,
The peak amplitude and the half power
points are determined,
SDOF modal analysis

r
knowns H ,
,
a b
a2 b2 a b
,
2 r = r
=
r =
2
r
2r
then
A
H = r 2 ,
Ar = rr2 H
rr
SDOF modal analysis

Another estimate for

modal residue:
H = ( max(Re) + min(Re) )
Ar = rr2 ( max(Re) + min(Re) )
Modal Testing
(Lecture 16)
Dr. Hamid Ahmadian

ahmadian@iust.ac.ir
Circle-fit method
Properties of the modal circle

Circle-fit analysis procedure
Interpretation of damping plots
Assuming a system with structural damping

the basic function to deal with is:
( ) =
Ajk
(1 ( / ) + ir )
2
r
2
r
Since the effect of modal constant is to scale

the size and rotate the circle, we consider:
1
( ) = 2
2
2
r (1 ( / r ) + ir )
Finding the natural frequency:
r
1 ( / r ) 2
tan =
, tan(90 ) = tan( ) =
2
1 ( / r )
2
r
2
2
1 ( / r )
d
r
2
2

= r (1 r tan( ))
=
1 +
2
2
d
r
2
r

2
2
r
1 ( / r )
1
d

=
1 +
( sweep rate)
r
2
d
2
r
d d 2
= 0. @ = r Natural frequency
d d
2
d
=
Damping
2
2
rr
d =r
b 1 (b / r )2
tan( 2 ) =
a2 b2
r
r =
2
a
b
2
tan( a ) = ( a / r ) 1
r (tan( ) + tan( ))
2
2
2
r
forr 2% K 3%
r =
2( a b )
r (tan(
a
2
) + tan(
b
2
))
when a = b = 90o
r =
a b
r
The final property relates to the diameter of

the circle (D):
D
=
r
jk
Ajk
2
r
Select points to
be used
Fit circle,
calculate quality
of fit
Locate natural
frequency,
Obtain damping
estimates
Calculate
multiple
damping
estimate and
scatter
Determine modal
constant module
and argument.
Interpretation of damping
plots
Noise may contribute

to the roughness of
the surface.
Systematic distortions
due to:
Leakage
Erroneous estimates for
natural frequency
Nonlinearity
Circle-fit
Minimizing the algebraic distance:

( x + a ) 2 + ( y + b) 2 = R 2
x 2 + y 2 + Ax + By + C = 0.
Least Squares Solution :
x12
2
x2
2
xn
+ y12
+ y 22
M
+ y n2
x1
x2
M
xn
y1 1 1

y 2 1 A
= 0.
M M B

y n 1 C
Circle-fit
Minimizing the geometric distance:
( x1 + a ) 2 + ( x2 + b) 2 = R 2
d =
2
i
u1
u1

X i u3 , Let U = u2
u
2
u
3
2
min d i (U ) 2
i =1
Circle-fit
d 0
p + L
d = d0 +
p
d 0
p
min d min d 0 +
p
u1 x11
2
2
u
x
u
x
(
)
(
)
1
11
2
12
d 0
M
=
p
u1 xm1
2
2
u
x
u
x
(
)
(
)
m1
m2
1
2
u2 x12
(u1 x11 ) 2 + (u2 x12 ) 2
M
u 2 xm 2
(u1 xm1 ) 2 + (u2 xm 2 ) 2
Circle-fit
Least-Squares Fitting of
Circles and Ellipses By: Walter
Gander, Gene H. Golub, and Rolf

Strebel
You may find it in ftpmech.iust.ac.ir
Home work 1
Determine the modal properties of the

beam tested in the lab
Frequency range of 0-400Hz
Natural frequencies
Damping (carpet plots)
Mode Shapes
Due time 87/2/22
Importing the ASCII files
Importing the ASCII files
Modal Testing
(Lecture 18)
Dr. Hamid Ahmadian

ahmadian@iust.ac.ir
MDOF Modal Analysis in the

Frequency Domain (SISO)
In some cases the SDOF approach to

modal analysis is simply inadequate or
inappropriate:
closely-coupled modes,
the natural frequencies are very closely spaced,

or
which have relatively heavy damping,
those with extremely light damping

Frequency Domain (SISO)
One step MDOF curve fitting methods:
Non-linear Least Squares Method

Rational Fraction Polynomial Method
A method particularly suited to very lightly
damped structures
Global Modal Analysis in Frequency

Domain
Global Rational Fraction Polynomial Method

Global SVD Method
Non-linear Least Squares

Method
m2
Ajk
1
1
H jk ( l ) = H l = 2
+
+ 2
2
2
K r l M r
r = m1 r l + i r r
r
l = H lm H l
The difference between
measurement and analytical
model
p
E = w ,
l =1
2
l l
dE
= 0.,
dq
q =1 Ajk ,2 Ajk ,3 Ajk ,L, 1 , etc

Non-linear Least Squares

Method
The set of obtained equations are

nonlinear
No direct solution (iterative procedures)

Non-uniqueness of solution
Huge computational load
Rational Fraction Polynomial

Method
N
H ( ) =
r =1
r
2
Ajk
+ 2i r r
2
r
b0 + b1 (i ) + b2 (i ) 2 + L + b2 N 1 (i ) 2 N 1
H ( ) =
a0 + a1 (i ) + a2 (i ) 2 + L + a2 N (i ) 2 N

Method
Order of model is selected
b0 + b1 (i k ) + b2 (i k ) + L + b2 m 1 (i k )
ek =
Hk
2
2m
a0 + a1 (i k ) + a2 (i k ) + L + a2 m (i k )
2
2 m 1
or
ek = (b0 + b1 (i k ) + b2 (i k ) 2 + L + b2 m1 (i k ) 2 m 1 )
H k (a0 + a1 (i k ) + a2 (i k ) 2 + L + a2 m (i k ) 2 m )

Method
b0
b
1
ek = 1 (ik ) (ik ) 2 L (ik ) 2 m 1 b2
M
b2 m 1
a0
a
1
H k 1 (ik ) (ik ) 2 L (ik ) 2 m 1 a2 a2 m H k (ik ) 2 m
M
a2 m 1

Method
A set of linear equations using each

individual measured FRF is formed.
The unknowns ai and bi are obtained
using a least square solution.
The modal properties are extracted
from obtained coefficients ai and bi.
The analysis may repeat for a different
model order.
Lightly Damped Structures
In these structures it is easy to locate

the natural frequencies,
Its accuracy is equal to the frequency

resolution of the analyzer
The damping ratio is assumed to be

zero.
The modal constants are obtained using
curve fittings.
Lightly Damped Structures

N
H ( ) =
r
2
r
A jk
r =1
(
(
H (1 ) 12 12
H ( ) 2
2
2
1
2
=

M
M
M
M
The natural frequencies are known
) (
) (
1
1
2
2
2
2
)
)
2 1
1
2 1
2
M
M
L 1 A jk
A
L 2 jk
M M
M M
Global Modal Analysis in

Frequency Domain
So far each measured FRF is curve

fitted individually,
Multi-estimates for global parameters

(natural frequencies and damping)
Another way is to use measured FRF

curves collectively.
Frequency and damping characteristics

appear explicitly.
Global Rational Fraction

Polynomial Method
If we take several FRFs from the same

structure then the denominator
polynomial will be the same in every case.
A natural extension of RFP method is to
fit all n FRFs simultaneously
2m-1 values of ai and,

and n(2m-1) values of bi
Global SVD Method
{H ( )}k
H1k ( )
H ( )
2k
=
H nk ( )
n1
= [ ]n N [(i sr )]
1
N N
{k }N 1 + {Rk ( )}N 1
Global SVD Method
{H ( )}k = [ ]n N [(i sr )] {k }N 1 + {Rk }

1
&
{H ( )}k = []nN [sr ][(i sr )]N N {k }N 1 + {Rk }
1
{g k ( )} = [(i sr )]N N {k }N 1
{H ( )}k = [ ]n N {g k ( )}+ {Rk }
{H& ( )}k = []nN [sr ]{g k ( )}+ {Rk }
1
NN
Global SVD Method
{H (i )}k = {H (i )}k {H (i +c )}k
{H (i )}k = [ ]n N {g k (i )}
{H& (i )}k = []nN [sr ]{g k (i )}
Global SVD Method
Consider data from several different frequencies

to obtain frequencies and damping:
[H ( )k ]nL = [ ]n N [g k ( )]N L
[H& ( )k ]nL = []nN [sr ][g k ( )]N L
T
T
+T
&
([H k ] sr [H k ] ){zr } = 0, [z] = []
Global SVD Method
The eigen-problem is solved using the SVD.

The rank of the FRF matrices and eigenvalues are
obtained.
Then the modal constants can be recovered from:
Modal Testing
(Lecture 18-1)
Dr. Hamid Ahmadian

ahmadian@iust.ac.ir

Time Domain
The basic concept: Any Impulse Response

Function can be expressed by a series of
Complex Exponentials
2N
h jk (t ) = r A jk e sr t ; sr = r r + i 1 r2
r =1
The Complex Exponential Series contain the

eigenvalues and eigenvectors information.
The IRF is obtained by taking inverse Fourier
transform of the measured FRF.
Complex Exponential Method

N
FRF jk ( ) =
r =1
A jk
i s r
2N
or : jk ( ) =
r =1
IRF
jk
i s
A jk
i s r
2N
h jk (t ) = r A jk e
sr t
r =1

(Single FRF)
2N
h(t ) = Ar e
r =1
2N
sr t
hl = Ar e
1
h0 1
h
1 V1 V2
2
2
h
2 = V1 V2
M M
M

q
q
h
q V1 V2
s r lt
r =1
= ArV
r =1
K
L
K
L
2N
l
r
1 A1
V2 N A2
2
V2 N M
M M
q
V2 N A2 N

(Single FRF)
0

1
2
M

q
h0 0
h
1 1
h2 = 2
M M

hq q
1
1
V V
2
1
V12 V22
M
M
V1q V2q
L 1 A1
L V2 N A2
K V2 N M
L M M
q
L V2 N A2 N
L
L
K
L
L
i
i hi = A j iV j
i =0
j =1
i =0
q
2N

(Single FRF)
The i are selected to be coefficients of the

polynomial:
0 + 1V + 2V 2 + L + qV q = 0.
2N
i
V
q
q
i j = 0,
2N
i =0
i
Set : q = 2 N i hi = A j iV j 2 N
i =0
j =1
i =0
h = 0.
i i

i =0
2 N 1
h
i =0
i i
= h2 N

(Single FRF)
2 N 1
h
i =0
i i
h0
h
1
M
M
h2 N 1
= h2 N
h1
h2
h2
h3
h2 N
h2 N +1
L h2 N 1 0
h2 N
L h2 N 1
h2 N +1
L
M M = M
M
L
M M
h4 N 1
L h4 N 2 2 N 1

(Single FRF)
The values Vi = e
from:
sr t
and Ai are obtained
0 + 1V + 2V 2 + L + 2 NV 2 N = 0.
1
h0 1
h V
V2
1 1
2
2
=
h
V
V
2
1
2
M M
M
h2 N 1 V12 N 1 V22 N 1
L
K
L
L
L
K
L
L
1 A1
V2 N A2
2
V2 N M
M M
2 N 1
V2 N A2 N

(Single FRF)
Implementation Procedure:
Order of modal model is selected,

Modal model is identified using the defined
steps in previous slides,
FRF is regenerated from modal information
and compared with the measured FRF
The procedure repeated using another
order for the modal model until stable
results are obtained.
Stabilization Diagram
Global Analysis in Time Domain

(Ibrahim Time Domain Method)
The basic concept is to obtain a unique

set of modal parameters from a set of
vibration measurements:
Scaled (mass normalized) mode shapes

when the force is known,
Un-scaled mode shapes when the force is
not measured.
Ibrahim Time Domain Method

2m
xi (t ) = ir e
sr t
r =1
x1 (t1 ) x1 (t 2 )
x (t ) x (t )
2 2
2 1
M
M
xn (t1 ) xn (t 2 )
L x1 (t q ) 11 12
L x2 (t q ) 21 22
=
L
M M
M

L xn (t q ) n1 n 2
L 1, 2 m e s1t1
L 2, 2 m e s2t1
L
M M
s2 mt1
L n , 2 m e
st
e 1q
s2t q
L L e
L L
M
s2 mt q
L L e
L L
[X ] = [ ] [ ]
A 2nd set of eqns:

2m
xi (tl + t ) = ir e
s r ( tl + t )
r =1
2m
= ir e
r =1
s r t
)e
2m
sr tl
= ir e
[X ] = [ ] []
sr tl
r =1
[
[A] [ ] = ]
[X ] = [ ] [ ]
[
]
[
]
[
]
X
=
X
[X ] = [ ] []
+
[A] = [X ] [X ]
[A]{ r } = e { r }
s r t
Eigenvectors of matrix [A] are the mode

shapes,
The natural frequencies and damping ratios
are obtained from eigenvalues of [A].
Modal
oda Testing
est g
(MDOF Modal Analysis in the Time Domain)
Dr. Hamid Ahmadian

Iran Universityy of Science and Technology
gy
ahmadian@iust.ac.ir

Time Domain
The basic concept: Any Impulse Response

Function can be expressed by a series off
Complex Exponentials
2N
h jk (t ) r A jk e ; sr r r i 1
sr t
r 1
2
r
The Complex Exponential Series contain the

eigenvalues and eigenvectors information
information.
The IRF is obtained by taking inverse Fourier
transform of the measured FRF
FRF.
Dr H Ahmadian, Modal Testing Lab, IUST
Complex Exponential
Method
h d
(CE)
Dr H Ahmadian,
Modal Testing Lab,
IUST

Methods

N
FRF jk ( )
r 1
A jk
i s r
2N
or : jk ( )
r 1
IRF
jk
i s
A jk
i s r
2N
h jk (t ) r A jk e
sr t
r 1

(Single FRF)
2N
h(t ) Ar e
r 1
2N
sr t
hl Ar e
1
h0 1
h
1 V1 V2
2
2
h
2 V1 V2

q
q
h
q V1 V2
r 1
s r lt
2N
ArV
r 1
l
r
1 A1
V2 N A2
2
V2 N
q
V2 N A2 N

(Single FRF)
0

1
2
h0 0
h
1 1
h2 2

hq q
1
1
V V
2
1
V12 V22

V1q V2q
1 A1
V2 N A2
2
V2 N
q
V2 N A2 N
q
i
i hi A j iV j
i 0
j 1
i 0
2N

(Single FRF)
The i are selected to be coefficients of the

polynomial:
0 1V 2V 2 qV q 0.
2N
i
V
q
2N
i j 0,
q
Set : q 2 N i hi A j iV ji i2N0
i 0
j 1
i 0
h 0.
i i

i 0
2 N 1
h
i 0
i i
h2 N

(Single FRF)
2 N 1
h
i 0
i i
h0
h
1

h2 N 1
h2 N
h1
h2
h2
h3
h2 N
h2 N 1
h2 N 1 0
h2 N
h
h2 N 1
2 N 1
1
h4 N 1
h4 N 2 2 N 1

(Single FRF)
The values Vi e
f
from:
sr t
and Ai are obtained
0 1V 2V 2 2 NV 2 N 0.
1
h0 1
h V
V2
1 1
2
2
h
V
V
2
1
2

h2 N 1 V12 N 1 V22 N 1
1 A1
V2 N A2
2
V2 N
V22NN 1 A2 N

(Single FRF)
Implementation Procedure:
Order of modal model is selected,

Modal model is identified using
g the defined
steps in previous slides,
FRF is regenerated
g
from modal information
and compared with the measured FRF
The p
procedure repeated
p
using
g another
order for the modal model until stable
results are obtained.
The Least Squares

q
Complex Exponential
Method
h d
(LSCE)
Dr H Ahmadian,
Modal Testing Lab,
IUST

Methods
The Least Squares Complex

Exponential Method (LSCE)
The LSCE is the extension of CE to a

global procedure.
It processes several IRFs
IRF s obtained
using SIMO method.
The coefficients that provide the
polynomial
y
are
solution of characteristic p
global quantities.
The Least Squares Complex

Exponential Method (LSCE)
One Typical IRF
h0
h
1

h2 N 1
h1
h2
h2
h3
h2 N
h2 N 1
h2 N 1 0
h2 N
h
h2 N 1
2 N 1
, or hq hq
h4 N 2 2 N 1
h4 N 1
Extending to all measured IRFs
h1
h1
h
h
2 2 , or h h h T h 1 h T h
G
G
G
G
G
G

hq
hp
The PolyReference
y
Complex Exponential
Method
h d (PRCE)
( C )
Dr H Ahmadian,
Modal Testing Lab,
IUST

Methods
The PolyReference Complex

Exponential Method (PRCE)
Constitutes the extension of LSCE to

MIMO.
A general and automatic way of
analyzing dynamics of a structure.
MIMO test method overcomes the
problem of not exciting
p
g some modes as
usually happens in SIMO.

Considering q input reference points:
2N
h j1 (t )
r A j1e sr t
r 1
2N
h j 2 (t ) r A j 2 e sr t
h jq (t )
2N
r 1
2N
A jll Qr jrlrl
A jk rWkl r A jl
rWkl
r A jq e sr t
r 1
kr
lr
h j1 (t )
r A j1e sr t
r 1
2N
h j 2 (t ) rW21 r A j1e sr t
r 1
2N
h jq (t ) rWq1
r 1
sr t
A
e
r j1
Modal Participation factor


2N
h j1 (t )
r A j1e sr t
r 1
2N
h j 2 (t ) rW21 r A j1e sr t h (t ) W e t A
j
j1
r 1
2N
h jq (t ) rWq1 r A j1e sr t
r 1
h j1 (t ) 1
h (t ) W
j 2 1 21

h jq (t ) 1Wq1
W
2 N 21
2 NWq1
1
2W21
2Wq1
e s1t
0 1 A j1
A
0 2 j1
e s2 N t 2 N A j1
0
e s 2t

h (0) W A
h (t ) W V A
j
j1
j1
L
h j ( Lt ) W V Aj1
, V e
0 1 W V 2 W V 2 L W V L 0, L q 2 N

0 h j (0) 0 W Aj1
1 h j (t ) 1 W V Aj1
2
2 h j (t ) 2 W V Aj1
h ( Lt ) W V A
L
j1
h (kt ) W V A
L
k 0
k 0
j1

h (kt ) 0, I
L
k 0
L 1
h (kt ) h ( Lt )
k 0
0
0 1
h j (t )
h j (( Nt 1)t )
h j (0)
h (t )
h
(
2
t
)
h
(
N
t
)
j
j
j
t
L 1
h
((
L
1
)
t
)
h
(
L
t
)
h
((
L
2
)
t
)
j
j
j
t
h j (( L Nt 1)t )
h j ( Lt ) h j (( L 1)t )
BT h j hj

BT h j hj
Considering
g for each response
p
location jj=1,,p:
, ,p
BT h1 h2 hp h1 h2
BT hT hT
T
T 1
BT hT hT hT hT
h
p
Knowing the coefficient matrix [B], we must now determine [V]


0 1 W V 2 W V
W V
1
k 0
0 L
L
k W V k k e s1t k W1 0
k 0
k 0
0
0
L
L
k 1
k W V k e s2t k W2 0
k 0
k 0
0
0
L
L
k 1
k W V k e s2 N t k W2 N 0
k 0
k 0
0
L W V 0,
L
L
k
k Vr Wr 0
k 0

L
k
k Vr Wr 0
k 0
V V
2
z0 Wr
z1 Vr Wr Vr z0
z2 Vr 2 Wr Vr z1
zL 1 Vr L1 Wr Vr zL2
zL Vr L Wr Vr zL1
L 1
L
Wr
V
W
V
L 11
r
0 z0 1 z1
L 1 z L 1 Vr z L 1

An standard eigenvalue
g
problem
p
to obtain Vr
L 1 L 2
0
I

0
0
1
0
z L 1
0
z L 2
0 V
z1
0 z
0
z L 1
z L 2

z
1
z0
The eigenvetors z0 correspond to Wr

h k t
h k t W V A , k 0,1,, L
h k t
h 0 W
h k t
h t
W
V

A or H W A h k t
h L t W V
k
j1
j1
j2
j1
j1
jq
A W H
j1
The residue calculation is repeated for

all measured
meas ed points
points, j=1,2,,p
j 1 2 p.

The method provide more accurate modal

representation off the structure.
It can determine multiple roots or closely
spaced modes.
Shortcomings:
Sensitive to nonlinearities and any lack of
reciprocity in frequency responses
responses,
Some difficulties in analyzing structures
with more than 5% viscous damping
damping.
Global Analysis in Time

Domain
i
(Ibrahim Time Domain
Method)
Dr H Ahmadian,
Modal Testing Lab,
IUST

Methods
Global Analysis in Time Domain

(Ibrahim Time Domain Method)
The basic concept is to obtain a unique

set of modal parameters from a set of
vibration measurements:
Scaled (mass normalized) mode shapes

when the force is known,
known
Un-scaled mode shapes when the force is
not measured
measured.

2m
xi (t ) ir e
sr t
r 1
x1 (t1 ) x1 (t 2 )
x (t ) x (t )
2 2
2 1

xn (t1 ) xn (t 2 )
x1 (t q ) 11 12
x2 (t q ) 21 22

xn (t q ) n1 n 2
1, 2 m e s1t1
2, 2 m e s2t1

n , 2 m e s2 mt1
st
e 1q
s2t q
e

s2 mt q
e
X
A 2nd set of eqns:

2m
xi (tl t ) ir e
s r ( tl t )
r 1
2m
ir e
r 1
s r t
2m
sr tl
ir e
X
sr tl
r 1
A
X
A X X
A X X
A r e r
s r t
Eigenvectors off matrix [[A]] are the

h mode
d
shapes,
The natural frequencies and damping ratios
are obtained from eigenvalues of [A].
Modal Testing
(Lecture 19)
Dr. Hamid Ahmadian

ahmadian@iust.ac.ir
Derivation of Mathematical
Models
Spatial Models (mass, stiffness, damping)
Response Models (FRF)
Needs measurement of most of the modes

Requires measurement in many DOFs
Needs measurement in frequency range of interest
Requires measurement in selected DOFs
Modal Models (natural frequencies and mode

shapes)
Needs measurement of only one mode

Requires measurement in handful of DOFs
Derivation of Mathematical Models
Models
Modal Models
Requirements to construct Modal Models

Refinement of Modal Model
Conversion to real modes

Compatibility of DOFs
Reduction
Expansion
Response Models
FRF
Transmissibility
Base Excitation
Requirement to construct
Modal Models
Minimum requirements
One column in case of fixed excitation or

One row when response is measured at a fixed point.
Modal Models
Proof:
mn
X m ( )
=
Fn ( )
=
X m ( ) X i ( ) Fi ( )
Fi ( ) Fn ( ) X i ( )
mi ni
=
ii
Modal Models
Several additional elements of FRF or even columns

are measured to:
Replace poor data,

To provide checks
Modes have not been missed
Refinement of Modal Models
Complex to real conversion:
Taking the modulus of each element and

assigning a phase of 0 or 180.
Finding a real mode with maximum projection to
the measured one:
max
RT C
R C
Multi point excitation (Ashers method)
Compatibility of DOFs
Employment of the measured modes in

updating/modification of analytical
models requires the compatibility of
DOFs.
There are two approaches in
compatibility excursive:
Analytical model reduction

Expansion of measured modes
Reduction of Analytical Model

(Guyan Reduction)
K11 K12 x1 f1
K T K x = 0

22 2
12
x2 = K 221 K12T x1
I
x1
x1
{x }, = [T ]{x1}
=
1 T 1
x2 K 22 K12
x2
K
K12
T 11
T {x1} = { f1}
T T
K12 K 22
Dynamic Model Reduction

K11
T
K
12
K12
M 11
2
T
K 22
M 12
M 12 1
= 0.
M 22 2
2 = (K 22 M 22 ) (K12T 2 M 12T )1
2
I
1
1
{ }, = [T ]{1}
1
=
2
2
T
T 1
2 K 22 M 22 K12 M 12
2
K
K12
M 11 M 12
2
T 11
T {1} = 0.
T T
T
K
K
M
M
22
22
12
12
)(
Expansion of Models
In order to compare analytical model

with the measured modal data on may
expand the measured data by:
Geometric interpolation using spline

functions
Using analytical model spatial model
Using analytical model modal model
Expansion in Spatial Domain

K11
T
K
12
M 12 1
=
0
.
M 22 2
K12
M 11
2
T
K 22
M 12
2 = (K 22 M 22 ) (K M )1
2
T
12
T
12
Expansion in Modal Domain
= 0R
Sol :
Assuming Mass
Matrix is correct
~
~
T
st : R R = I .
min R 0
R
~T ~
T
T
0 = UV R = VU ,
Response Model
Frequency response functions
[H ] = [ ][(
2
r
) ][]
2 1
Transmissibilities
T jk ( ) =
X je
X ke
it
it
T ( ) =
i jk
H ji ( )
H ki ( )
Transmissibility Plots
Response Model
The amplitudes peaks

of the transmissibilities
do not correspond with
the resonant
frequencies.
iT
Transmissibilities cross
each other at the
resonant frequencies
(becomes independent
of the location of the
input)
jrir
2
2
H ji ( )
r r
=
jk ( ) =
krir
H ki ( )
r 2 2
r
T jk ( )
jr
kr
Base Excitation
An application area
of transmissibility.
Input is measured
as response at the
drive point.
1
1
{x} = {xrel }+ xref

M
1
Base Excitation
1
1
[M ]{&x&rel }+ [K ]{xrel } = &x&ref [M ]{g}, {g} = .

M
1
[H ( )] ({X } xref {g}) = 2 xref [M ]{g},

(
{X } xref {g})
= [H ( )][M ]{g }.
or
2
1
xref
Base Excitation
({X } x {g}) = ([H ( )])([M ]{g}),

ref
xref
2
(
{X } x {g})
{Q} =
, {u} = [M ]{g },
ref
xref
2
ir jr
Qi ( ) = 2
u
2 j
j
r r
Spatial Models
[M ] = [ ] [ ]
T
1
[K ] = [ ] [r ][ ]
T
Modal Testing
(Lecture 20)
Dr. Hamid Ahmadian

ahmadian@iust.ac.ir
Models
Introduction
Equation Error Method (Sec. 6.3.6 page 456)
Identification of Rod FE Model

Parameter Identification
Solution of Over-determined set of Equations

Solution of Under-determined set of
Equations
Error Analysis
Introduction
Construction of Spatial Model from modal

data:
T
K = , M = , C =
Modal model must be complete:
All modes must be present

Mode shapes are measured in all DOFs
Measurement of complete Modal Model is

impractical.
Introduction
Alternative methods are required to construct

the spatial model from
incomplete and
noisy measured modes.
The difficulty with incompleteness is removed

by reducing the number of unknowns in
spatial model.
The noise effects are removed by averaging.
Equation Error Method
We have some information regarding

the spatial model format:
Symmetry
Pattern of zeros
We may incorporate these information

into the identification procedure and
reconstruct the spatial model.
Equation Error Method
In this method the eigen problem is

rearranged to obtain the spatial model:
K M = 0.
K
= 0
M
T
The DOFs of measured modes must be

compatible with the DOFs of spatial model.
Rearrangement Example:
k1 + k 2
k
2
or :
k2
m
2 1
r
k2
0
1 1 2
0 2 1
r21
0 1
= 0,
m2 2
k1
0 k2
= 0.
2
r 2 m1
m2
Identification of A Rod FE
Model
Consider a fixed-free rod with n elements.

The mass and stiffness matrices are:
k2
k1 + k2
k
k 2 + k3 k3
2
K=
L
L
L
kn 1 kn 1 + kn
kn
M = diag ( m1 , m2 ,L, mn 1 , mn )
kn
kn
Model
The equilibrium state at modes r and s are:
(K r M )r = 0.
(K s M )s = 0.
The last rows of equilibrium state equations

are:
knr ,n 1 + (kn r mn )r ,n = 0,
kns ,n 1 + (kn r mn )s ,n = 0.
Model
r ,n r ,n 1 rr ,n kn
= 0.

ss ,n mn
s ,n 1
s ,n
r ,n (s ,n s ,n 1 )
s
r
s ,n (r ,n r ,n 1 )
kn 0, mn 0
rr ,n
k
n
=
mn r ,n r ,n 1
Model
From other rows one obtains:
k1 k2
kn 1 m1 m2
mn 1
, ,K,
, , ,K,
mn mn
mn mn mn
mn
Using total mass information mm is obtained:
ml
= mn 1 +
l =1 mn
n 1
mtotal
Model
Only two modes and one natural

frequency are required to construct the
mass and stiffness matrices.
More details can be found in:
GML Gladwell, YM Ram, Constructing

Finite Element Model of a Vibrating Rod,
Journal of Sound and Vibration, 169,229237,1994.
In a general case the mass and stiffness

matrices are parameterized and are
obtained by rearranging:
Equation of motion in modal domain

Orthogonality requirements,
etc.
Parameterization :
K = K ( k1 , k2 ,K, kn ), M = M ( m1 , m2 ,K, mn ).
EOM :
K M = 0, M = I , K = .
T
Extras :
K R = 0,
M R m, I xx , I xy ,K
T
R
Re arrangement :
Ax = b
A = A(, R , ), b = b(, mtotal , I xx , K)
x = x(k1 , k 2 , K , k n , m1 , m2 , K , mn )
A full
m = n x = A1b
rank (n m) m > n Under det er min ed

m < n Over det er min ed
Solution of underdetermined
case
min x x0 , ST : Ax = b
or
min x , ST : Ax = b Ax0 = b
Solution :
min(x T x 2(x T AT b T ) ) 2x 2 AT = 0.
AA = b = (AA
T
T 1
b x = A (AA
T
T 1
Solution of Over-determined
set of Equations
Ax = b,
Ax b =
m<n
[ ]
Assume E [ i ] = 0., E j = ij
Solution min = min T
= x A Ax 2 x A b + b b
T
( )
1 T

T
T
T
= 2 A Ax 2 A b x = A A A b
x
T
Error Analysis
b = Ax +
E [(A A)
A b] = E [x ] + E [(A A)
E [x ] = E [x ] + E [(A A) A ]
T
A
T
as m , E [ ] 0 E [x ] = E [x ]
[ ]
If A noisy E A 0 E [x ] E [x ]
T
Example:
The parameters to be
updated are the 10
stiffness and 6 masses
The measured data
consists of the 1st
three natural
frequencies and mode
shapes (added with
uniformly distributed
random noise)
Example:
Eigenvalue equations arearrangment:
31 equations ( 3*6 equations for each eigenvector term, 2*6

symmetric orthogonality equations, and 1 total mass equation)
16 parameters and
The terms in A and b contain noisy data.
Parameters : x = {k1 , k 2 , K , k10 , m1 , m2 , K , m6 },
EOM : K M = 0, T M = I , T K = .
Extras : MR = m.
T
R
Example
k1
k2
k3
k4
k5
k6
k7
k8
k9
k10 m1 m2 m3 m4 m5 m6
Exact
1000
1250
1500
10000
1000
1000
5000
7000
1000
1000
0.2
0.1
0.1
0.1
S/N=
100
1041
1266
1518
10008
1006
1008
9914
-414
2035
-63
0.2
0.1
0.99
0.2
-0.01
S/N=
20
954
1243
1502
9844
1160
1011
-1024
21352
-83
2323
0.97
0.21
0.1
0.96
-0.02
0.28
Regularized Solution
Ahmadian, Mottershead,
and Friswell,
REGULARISATION
METHODS FOR FINITE
ELEMENT MODEL
UPDATING, Mechanical
Systems and Signal

Processing (1998)
12(1),47-64
Home Work 3
Develop a procedure to construct the

FE model of a fixed-free beam from
minimum modes.
How many modes are required to obtain EI

an m of each element?
Add some noise to the modes and try to
reconstruct the model.
Investigate the correlated noise effects?

MT Lectures 11-20

Загружено:

Сведения о документе

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

MT Lectures 11-20

Загружено:

Авторское право:

Доступные форматы

Modal Testing

Dr. Hamid Ahmadian

Response Function Measurement

IUST ,Modal Testing Lab ,Dr H Ahmadian

The measurements techniques used for

Response measurement only

The 2nd type of measurement

Single-point excitation( SISO/SIMO)

Response Function Measurement

IUST ,Modal Testing Lab ,Dr H Ahmadian

Objective of the test

Levels according to Dynamic Testing Agency:

IUST ,Modal Testing Lab ,Dr H Ahmadian

Extensive test planning is required

Response Function Measurement

IUST ,Modal Testing Lab ,Dr H Ahmadian

Quality of measured data

Sufficient strength and clarity/noise free

Response Function Measurement

IUST ,Modal Testing Lab ,Dr H Ahmadian

Basic Measurement System

Response Function Measurement

IUST ,Modal Testing Lab ,Dr H Ahmadian

Basic Measurement System

Response Function Measurement

IUST ,Modal Testing Lab ,Dr H Ahmadian

Response Function Measurement

IUST ,Modal Testing Lab ,Dr H Ahmadian

Theoretically the structure will possess 6 rigid

Suspending from nodal points for minimum

Response Function Measurement

IUST ,Modal Testing Lab ,Dr H Ahmadian

Response Function Measurement

IUST ,Modal Testing Lab ,Dr H Ahmadian

Response Function Measurement

IUST ,Modal Testing Lab ,Dr H Ahmadian

Response Function Measurement

IUST ,Modal Testing Lab ,Dr H Ahmadian

The structure is fixed to the ground at selected

Another application is simulating the operational

Parts of power generation station

Static stiffness can be obtained from low frequency

Response Function Measurement

IUST ,Modal Testing Lab ,Dr H Ahmadian

The structure is connected to a simple

The effect of added mass can be removed

Response Function Measurement

IUST ,Modal Testing Lab ,Dr H Ahmadian

The data base for the structure can be

Response Function Measurement

IUST ,Modal Testing Lab ,Dr H Ahmadian

Response Function Measurement

IUST ,Modal Testing Lab ,Dr H Ahmadian

Response Function Measurement

IUST ,Modal Testing Lab ,Dr H Ahmadian

Excitation of the structure

Various devices are available for

Mechanical (Out-of-balance rotating masses)

Response Function Measurement

IUST ,Modal Testing Lab ,Dr H Ahmadian

Supplied input to the shaker is converted to

Response Function Measurement

IUST ,Modal Testing Lab ,Dr H Ahmadian

There is a small difference between the force