Global Journal of researches in engineering

Mechanical and mechanics engineering

Volume 12 Issue 1 Version 1.0 January 2012
Type: Double Blind Peer Reviewed International Research Journal
Publisher: Global Journals Inc. (USA)
Online ISSN: 2249-4596 Print ISSN:0975-5861

Experimental and Analytical Modal Analysis of Welded

Structure Used For Vibration Based Damage Identification
By Putti Srinivasa Rao, Ch.Ratnam
Andhra University, Visakhapatnam, Andhra Pradesh, India
Abstract - This paper presents a systematic procedure and details of the use of experimental and
analytical modal analysis of a welded structure used for vibration based damage identification. First an
experimental modal analysis was conducted on the undamaged welded structure model used for
vibration based damage identification. Using impact hammer test, when the test structure is fitted to the
multi-axis electro dynamic vibration shaker .In this experimental modal analysis frequency response
functions are calculated from measured input force and output response of a structure using piezoelectric
accelerometers from the frequency response function the peaks were identified. To compare the
experimental modal analysis values finite element modal analysis was conducted using ANSYS software
version 11.0. It has been observed from the results the natural frequencies obtained from the experimental
modal analysis and ANSYS software version 11.0 shows a good consistency in comparison.

Keywords : Experimental Modal Analysis, Theoretical Modal Analysis, Finite Element Method,
Impact Hammer, Vibration Based damage identification, Eigenvalues and Eigenvectors,
frequency response function, Resonance condition.
GJRE-A Classification : FOR Code: 091304

Experimental and Analytical Modal Analysis of Welded Structure Used For Vibration Based Damage Identification

Strictly as per the compliance and regulations of:

© 2012 Putti Srinivasa Rao, Ch.Ratnam. This is a research/review paper, distributed under the terms of the Creative Commons
Attribution-Noncommercial 3.0 Unported License http://creativecommons.org/licenses/by-nc/3.0/), permitting all non commercial
use, distribution, and reproduction in any medium, provided the original work is properly cited.
 Experimental and Analytical Modal Analysis of
Welded Structure Used For Vibration Based
Damage Identification
Putti Srinivasa Rao α, Ch.Ratnam Ω

January 2012
Abstract - This paper presents a systematic procedure and development of methods that examine changes in the
details of the use of experimental and analytical modal vibration parameter of the structure [1-6].
analysis of a welded structure used for vibration based The actual implantation of vibration based
damage identification. First an experimental modal analysis
damage identification using statistical process control
was conducted on the undamaged welded structure model
for mechanical, aerospace, marine and civil engineering
used for vibration based damage identification. Using impact
hammer test, when the test structure is fitted to the multi-axis infrastructures starts with designing a proof of concept 45
electro dynamic vibration shaker .In this experimental modal experiment. First, an excitation mechanism for vibration

Global Journal of Researches in Engineering ( A ) Volume XII Issue vI Version I

analysis frequency response functions are calculated from testing should be selected. The excitation methods fall
measured input force and output response of a structure using into the two general categories of ambient and forced
piezoelectric accelerometers from the frequency response excitation methods. During ambient excitation, the input
function the peaks were identified. To compare the to a system is not generally measured. In contrast,
experimental modal analysis values finite element modal excitation forces are usually applied in a controlled
analysis was conducted using ANSYS software version 11.0. It
manner and measured when the forced excitation
has been observed from the results the natural frequencies
method is employed [7]. A resonance condition exists
obtained from the experimental modal analysis and ANSYS
software version 11.0 shows a good consistency in when the frequency of excitation due to any source
comparison. coincides with a natural frequency of the structure.
Keywords : Experimental Modal Analysis, Theoretical Therefore it is necessary to know the natural frequencies
Modal Analysis, Finite Element Method, Impact of the structure to be monitored prior to the
Hammer, Vibration Based damage identification, experimentation. The modal analysis primarily concerns
Eigenvalues and Eigenvectors, frequency response determination of natural frequencies and mode shapes
function, Resonance condition. of a dynamic structure. Once the modes are
determined, they can be used in understanding the
I. INTRODUCTION dynamic nature of the structure. Therefore Modal
analysis is an important tool in vibration analysis,

O
ur dependency on mechanical, aerospace,
diagnosis, design, and control.
marine and civil engineering infrastructures is
Vibration is a repetitive, periodic, or oscillatory
increasing day by day. These structures
response of an engineering structure. The rate of the
continue to be used, despite aging and the associated
vibration cycles is termed “frequency”. Vibrations can
potential for damage accumulation. All these
naturally occur in an engineering structure and may be
infrastructures are subjected to damage as a result of
representative of its free and natural dynamic behavior.
fatigue, overloading conditions, material degradation
Vibrations may also be forced onto a structure through
through environmental effects and unanticipated
some form of excitation [8]. The excitation forces may
discrete events such as impacts or seismic events.
be either generated internally within the dynamic
Damage adversely affects the current or future
system, or transmitted to the structure through an
performance of these infrastructures. Therefore, the
external source. When the frequency of the forcing
interest in the ability to monitor the health of these
excitation coincides with that of the natural motion, the
infrastructures and damage identification at the earliest
structure will respond more vigorously with increased
possible stage is very important to ensure performance
amplitude. This condition is known as resonance, and
standards, extend the operational lifespan, economical
the associated frequency is called the resonant
and maintain life-safety. Therefore the need for robust
frequency. Natural, free vibration is a manifestation of
global damage identification methods that can be
applied to complex structures has led to the the oscillatory behavior in engineering structures, as a
result of repetitive interchange of kinetic and potential
energies among components in the structure.
Author α Ω : Department of Mechanical Engineering, Andhra University,
Visakhapatnam, Andhra Pradesh, India-530003. Complex vibrating structures usually consist of
E-mail : [email protected] components that possess distributed energy storage

© 2012 Global Journals Inc. (US)

 Experimental and Analytical Modal Analysis of Welded Structure Used For Vibration Based Damage
Identification

and energy dissipative characteristics. An engineering II. SPECIFICATIONS OF THE TEST

structure, when given an initial disturbance and allowed
STRUCTURE
to execute free vibrations without a subsequent forcing
excitation, will tend to do so at a particular “preferred” The applicability of the proposed vibration
frequency and maintaining a particular “preferred” based damage identification technique to structural
geometric shape. This frequency is termed a “natural health monitoring problems is implemented using the
frequency” of the structure, and the corresponding cantilever plate like structure. The geometry considered
shape (or motion ratio) of the moving parts of the for this purpose is shown in Fig 1, where the two plates
structure is termed a “mode shape.” Any arbitrary are welded to form a cantilever plate. The thicknesses of
motion of a vibrating structure can be represented in both the plates are 12 mm and are made up of carbon
terms of its natural frequencies and mode shapes. steel. The dimension of the first plate is 150 mm x 150
January 2012

The limitations of the human mind are such that mm square plate and it is drilled with four 8 mm through
it cannot grasp the behavior of its complex structures in holes at each corner of the plate. The centre of the
one operation. Thus the process of subdividing all drilled holes is at 18 mm from each corner of the plate.
complex structures into their individual components or The dimension of the second plate is 150 mm x 100 mm
“Finite elements” whose behavior is understood very and is welded to the plate one to form a cantilever plate
easily, then rebuilding the original complex structures like structure. The elastic module of the structure
46
from the individual components or “Finite elements” to material (E) is 200 x 109 N/m2, Poisson’s ratio 0.3 and the
study its behavior. The term finite element was first used mass density is 7850 kg/m3. The test structure is fitted to
Global Journal of Researches in Engineering ( A ) Volume XII Issue vI Version I

by Clough in 1960 [9] and gives the basic idea of Finite the multi axis electro dynamic vibration shaker with the
element method. The finite element method is a help of four bolts and nuts as shown in Fig. 1
numerical method but is more general and powerful in
its application to real world problems that involves
complicated physics, geometry and/or boundary
conditions. Engineering application of the Finite element
method may be used in the three major categories of
boundary value problems, namely 1) Equilibrium
problems 2) Eigenvalue problems 3) Propagation or
Transient problem
The generalized problem in free vibration is that
of evaluating an Eigenvalue which is a measure of the
frequency of vibration together with the corresponding
eigenvector indicating the mode shapes. Actually the
Eigenvalue problem may be considered as extension of
the equilibrium problem in which critical values of certain
parameters are determined in addition to the steady
state configurations. The Eigenvalue-eigenvector
calculation procedure falls into the three basic Fig.1 : Test Structure fitted to the Multi Axis Electro
categories namely characteristic polynomial technique, Dynamic Vibration Shaker
vector iteration method and transformation method.
Aiming to investigate the vibration phenomena III. EXPERIMENTAL MODAL ANALYSIS(EMA)
occurring in test structure first an experimental modal
analysis (EMA) was conducted. The rapid development Modal analysis is vital to understand and
of finite-element techniques accompanied by optimize the inherent dynamic behavior of structures,
tremendous technological progress in the field of leading to lighter, stronger, and safer structures with
personal computers allowed structural designers to use better performance. Experimental modal analysis is
software packages like ANSYS for accurate simulation based on determining the modal parameters by testing,
of structural behavior. In this work the experimental unlike analytical modal analysis, where the modal
modal analysis (EMA) values are compared with the parameters are derived from finite element models
results obtained from ANSYS software version 11.0. (FEMs). There are two ways of doing experimental
The main purpose of this paper is to present our modal analysis [10-15]. They are 1) Classical modal
perspective concerning the evolution of modal analysis analysis and 2) Operational modal analysis. In classical
of the test structure used for vibration based damage modal analysis frequency response functions (or
identification, experimentally and compared the results impulse response functions) are calculated from
obtained from ANSYS software package version 11.0. measured input forces and output responses of a
structure. Most of the analysis in modal testing is
performed in the frequency domain inside the analyzer.

© 2012 Global Journals Inc. (US)

 Experimental and Analytical Modal Analysis of Welded Structure Used For Vibration Based Damage
Identification

The task of the analyzer is to convert analog time force transducer located behind the tip, the impact
domain signal into digital frequency domain information hammer measures the force used to excite the structure.
compatible with digital computing and then to perform The force input and corresponding responses
the required computations with these signals. are then used to compute the FRFs. The FRFs obtained
Operational Modal Analysis is based on measuring only from the impact hammer test is shown in Fig.4. Testing
the output of a structure and using the ambient and with impact hammer has some very distinct advantages.
operating forces as unmeasured input. It is used instead The input spectrum from the impact is flat out to the roll-
of classical mobility-based modal analysis for accurate off frequency with no holes in the spectrum. The
modal identification under actual operating conditions, technique can be very efficient and portable compared
and in situations where it is difficult or impossible to to the aligning and moving of shakers and their
control an artificial excitation of the structure. Classical associated control systems.

January 2012
modal analysis is a more mature technique in
comparison to operational modal analysis, and is
extremely useful in the design of engineering structures.
Enhanced computing power and advances in finite
element analysis (FEA) techniques have increased the
fidelity of analytical model and in several cases have
47
reduced the need for classical modal analysis,
especially with legacy structures. However, classical

Global Journal of Researches in Engineering ( A ) Volume XII Issue vI Version I

testing will continue to be required to give engineers the
confidence they need to continue to bring new product
into development in today’s competitive market.
Classical modal analysis relies heavily on adhering to
the four primary assumptions: 1) observability, 2)
linearity. 3) time invariant and 4) reciprocity.
When a modal test is performed on a test
structure the objective is to measure data from which Fig. 3 : Experimental setup for Impact Hammer test
the modal parameters - modal frequencies can be
estimated. The most typical data used for parameter The natural frequencies of the test structure
estimation are frequency response functions (FRFs), which were identified with the peaks in FRFs plot and
which use excitation input and the corresponding output the values up to five kHz are tabulated in table 1.
of the test structure. Transient excitation is an input of
short duration relative to the measured time record in Table 1 : Result Obtained From Experimental Modal
contrast to random or sine inputs. The versatility of Analysis (Impact Hammer Test)
transient excitation techniques allows for several Mode Number Natural Frequency in
advantages over typical vibration shaker input. Quick Hz
diagnostics of structures with short setup times are
1 862.5
possible. The most commonly used method of transient
2 1737.5
excitation for modal testing is the impact hammer. The
impact hammer used to excite the test structure during 3 2159.4
experimental modal analysis is shown in Fig. 2. 4 2631.3
5 4050.0

FRF PLOT FOR THE TEST STRUCTURE IN FOUR BOLT FIXED CONDITION
40

35

30
m/s^2/N

25

20

15

10

Fig. 2 : Impact Hammer 5

0
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000
The idea of exciting a structure with an impact NATURAL FREQUENCY Hz
hammer actually involves striking the structure at a
particular location and in particular direction with an Fig.4 : Impact Hammer FRFs Measured from the Test
impact hammer as shown in Fig. 3. Instrumented with a Structure
© 2012 Global Journals Inc. (US)
 Experimental and Analytical Modal Analysis of Welded Structure Used For Vibration Based Damage
Identification

IV. MODAL ANALYSIS USING ANSYS The total number of nodes generated in the
meshing of the test structure is 1289, and the total
In this application the plate is subjected to number of elements is found to be 1164. The first nine
transverse loads and in-plane loads and at any point natural frequencies for the test structure are then
inside the plate experiences both in-plane and lateral calculated and the values are tabulated in table 2. The
displacements. The natural frequencies for the test resultant deformation at each natural frequency and
structure are calculated using SHELL63 element in corresponding figures are given in Fig 6.
ANSYS software version 11.0. SHELL63 has both
bending and membrane capabilities. Both in-plane and
normal loads are permitted for the SHELL63 element.
The element has six degrees of freedom at each node
January 2012

i.e., translations in the nodal x, y, and z directions and

rotations about the nodal x, y, and z-axes.
The geometry, node locations, and the
coordinate system for SHELL63 element are shown in
Fig.5. The SHELL63 element is defined by four nodes,
four thicknesses, elastic foundation stiffness, and the
48
orthotropic material properties. Orthotropic material
directions correspond to the element coordinate
Global Journal of Researches in Engineering ( A ) Volume XII Issue vI Version I

directions. In this application the thickness of the

element is constant. The elastic foundation stiffness
(EFS) is defined as the pressure required to produce a
First Natural Frequency Second Natural Frequency
unit normal deflection of the foundation. The elastic
foundation capability is bypassed if EFS is less than, or
equal to, zero.

Third Natural frequency Fourth Natural Frequency

Fig.5 : SHELL63 Element Geometry

Table 2 : Results Obtained From Ansys Software
Package Version 11.0
Mode Number Natural Frequency in Hz
1 885.2
2 1731.0
3 2164.0
4 2626.0
5 4053.0
6 5454.0
7 6752.0 Fifth Natural Frequency Sixth Natural frequency
8 6859.0
9 6957.0

© 2012 Global Journals Inc. (US)

 Experimental and Analytical Modal Analysis of Welded Structure Used For Vibration Based Damage
Identification

experiment are almost coinciding with the ANSYS

results. The table 3 shows the comparison of natural
frequencies obtained from the experiment and ANSYS;
and it shows a quite satisfactory correlation.

Table 3 : Comparisons Of Natural Frequencies

Natural Frequencies of the test
Mode structure in Hz
Number Experimentally ANSYS
1 862.5 885.2

January 2012
2 1737.5 1731.0
3 2159.4 2164.0
4 2631.3 2626.0
Seventh Natural Frequency Eighth Natural Frequency 5 4050.0 4053.0
6 -- 5454.0 49
7 -- 6752.0

Global Journal of Researches in Engineering ( A ) Volume XII Issue vI Version I

8 -- 6859.0
9 -- 6957.0

VI. CONCLUSIONS
This paper presents a systematic procedure
and details of the use of experimental and analytical
modal analysis of a welded structure used for vibration
based damage identification. It has been concluded
from the results that the natural frequencies obtained
from the experimental modal analysis and ANSYS
software version 11.0 shows a good consistency in
comparison.
Ninth Natural frequency
REFERENCES REFERENCES REFERENCIAS
Fig.6 : First Nine Natural Frequencies of the Test
Structure 1. Putti srinivasa Rao and Ch.Ratnam “Damage
Identification of Welded Structures Using Time
Series Models and Exponentially Weighted Moving
V. RESULTS AND DISCUSSIONS Average Control” Jordan Journal of Mechanical and
The actual implantation of vibration based Industrial Engineering, Vol.4, (2010) 701-710.
damage identification using statistical process control 2. Putti srinivasa Rao and Ch.Ratnam “Vibration Based
for mechanical, aerospace, marine and civil engineering Damage Identification Using Burg’s Algorithm and
infrastructures starts with designing a proof of concept Shewhart Control Charts” Journal of ASTM
experiment. First, an excitation mechanism for vibration International, Vol. 8, No. 4, (2011).
testing should be selected. The excitation methods fall 3. Ch.Ratnam, J.Srinivas and B.S.N Murthy, “Damage
into the two general categories of ambient and forced detection in mechanical system using Fourier co-
excitation methods. A resonance condition exists when efficients” Journal of sound and vibration, Vol. 303,
the frequency of excitation due to any source coincides (2007) 909-917.
with a natural frequency of the structure. Therefore it is 4. A. Deraemaeker, E. Reynders, G. De Roeck, J.
necessary to know the natural frequencies of the Kullaa “Vibration-based structural monitoring using
structure to be monitored prior to the experimentation. output-only measurements under changing
Experimental and analytical modal analysis of a welded environment” Mechanical systems and signal
structure used for vibration based damage identification processing,Vol.22, (2008) 34-56.
was conducted. 5. I. Trendafilova, “Vibration-based damage detection
The natural frequencies obtained from the in structures using time series analysis”
experimental modal analysis using impact hammer test Proceedings of the Institute of Mechanical
and finite element modal analysis using ANSYS version Engineering Part C- Journal of mechanical
11.0 software package were compared. It has been engineering science, Vol.220, no.3, (2006) 261-272.
observed that the natural frequencies obtained from the 6. J.M.Nichols, S.T.Trickey, M.Seaver, S.R.Motley and
© 2012 Global Journals Inc. (US)
 Experimental and Analytical Modal Analysis of Welded Structure Used For Vibration Based Damage
Identification

E.D.Eisner “using ambient vibrations to detect

loosening of a composite-to-metal bolted joint in the
presence of strong temperature fluctuations” ASME
journal of vibration and acoustics, Vol. 129, (2007)
710-717.
7. Hoon Sohn, Charles R. Farrar,Francois M.
Hemez,Devin D. Shunk,Daniel W. Stinemates, Brett
R. Nadler and Jerry J. Czarnecki “A Review of
Structural Health Monitoring Literature: 1996–2001”
LA-13976-MS, Los Alamos National Laboratory,
USA, 2004.
January 2012

8. Clarence W. de Silva “Vibration Fundamentals and

Practice” Boca Raton: CRC Press LLC, International
Standard Book Number 0-8493-1808-4, Printed in
the United States of America, 2000.
9. Singiresu S.Rao “The finite Element method in
Engineering” Butterworth-Heinemann Publisher, a
50
member of Reed Elsevier group, 3rd Edition, 2001.
10. M.J.Carre and S.J.Haake “An examination of the
Global Journal of Researches in Engineering ( A ) Volume XII Issue vI Version I

clegg impact hammer test with regard to the playing

performance of synthetic sports surfaces” sports
engineering vol.7, 121-129 2004.
11. Anayet U.Patwari, Waleed F.Faris, A.K.M.Nurul
Amin, and S.K.Loh “Dynamic Modal Analysis of
Vertical Machining Centre Components” Advances
and Vibration vol.2009, Article ID 508076, 10 pages.
12. Alvaro Cunha and Elsa Caetano” Experimental
Modal Analysis of Civil Engineering Structures”.
13. C.Delprete, A.Galeazzi and F.prengo”Experimental
Modal Analysis of an Automotive Power train”,
Applied Mechanics and Materials Vol.24-25, 71-76
2010.
14. T.G.Carne and E.C.Stasiunas” Lessons Learned In
Modal Testing- Part 3: Transient Excitation For
Modal Testing, More Than Just Hammer Impacts”
Society for Experimental Mechanics, May/June
2006.
15. Kenneth A.Ramsey, Structural Measurement
Systems, San Jose, California” Experimental Modal
Analysis, Structural Modifications and FEM Analysis
on a Desktop Computer” Sound and Vibration,
February 1983.

© 2012 Global Journals Inc. (US)