See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/278173441

On the use of Optical Fiber Bragg Grating (FBG) Sensor Technology for Strain
Modal Analysis

Conference Paper · May 2014

DOI: 10.1063/1.4879567

CITATIONS READS
8 940

4 authors, including:

Bart Peeters Fabio Luis Marques dos Santos

Siemens Industry Software Siemens Industry Software
301 PUBLICATIONS 10,078 CITATIONS 68 PUBLICATIONS 601 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Bart Peeters on 10 September 2015.

The user has requested enhancement of the downloaded file.

 On the use of optical fiber Bragg grating (FBG) sensor technology for strain modal
analysis
Bart Peeters, Fábio Luis Marques dos Santos, Andreia Pereira, and Francisco Araujo

Citation: AIP Conference Proceedings 1600, 39 (2014); doi: 10.1063/1.4879567

View online: http://dx.doi.org/10.1063/1.4879567
View Table of Contents: http://scitation.aip.org/content/aip/proceeding/aipcp/1600?ver=pdfcov
Published by the AIP Publishing

Articles you may be interested in

High resolution interrogation system for fiber Bragg grating (FBG) sensor application using radio frequency
spectrum analyser
AIP Conf. Proc. 1528, 444 (2013); 10.1063/1.4803642

DYNAMIC STRAIN SENSING IN A LONGSPAN SUSPENSION BRIDGE USING FIBER BRAGG GRATING
SENSORS
AIP Conf. Proc. 1335, 1418 (2011); 10.1063/1.3592098

CRYOGENIC FIBER OPTIC SENSORS BASED ON FIBER BRAGG GRATINGS

AIP Conf. Proc. 985, 940 (2008); 10.1063/1.2908693

Cryogenic Fiber Optic Temperature Sensors Based on Fiber Bragg Gratings

AIP Conf. Proc. 823, 267 (2006); 10.1063/1.2202425

Fiber optic sensor for dual measurement of temperature and strain using a combined fluorescence lifetime decay
and fiber Bragg grating technique
Rev. Sci. Instrum. 72, 3186 (2001); 10.1063/1.1372171

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
151.5.180.15 On: Wed, 25 Jun 2014 11:05:38
 On the use of Optical Fiber Bragg Grating (FBG) Sensor
Technology for Strain Modal Analysis
Bart Peetersa, Fábio Luis Marques dos Santosa, Andreia Pereirab and Francisco
Araujob
a
LMS, A Siemens Business, LMS International, Interleuvenlaan 68, B-3001 Leuven, Belgium
([email protected])
b
FiberSensing, Rua Vasconcelos Costa, 277, PT-4470-640 Maia, Portugal

Abstract. This paper discusses the use of optical fiber Bragg grating (FBG) strain sensors for structural dynamics
measurements. For certain industrial applications, there is an interest to use strain sensors rather than or in combination
with accelerometers for experimental modal analysis. Classical electrical strain gauges can be used hereto, but optical
strain sensors are an interesting alternative with some very specific advantages. This paper gives an overview of dynamic
strain measurements in industrial applications, discusses the benefits of FBG sensors and reviews their measurement
principle. Finally, the concept of strain modal analysis is introduced and a helicopter main rotor blade vibration testing
and analysis case study is presented.
Keywords: Fiber Bragg Grating, dynamic strain measurements, Experimental Modal Analysis.
PACS: 07, 42, 45

INTRODUCTION

Experimental and Operational Modal Analysis typically use structural response measurements in the form of
accelerations as data type. However, in quite some applications, also (dynamic) strain measurements are acquired. In
durability applications, fatigue failure is related to stress (strain) cycles; in Transfer Path Analysis, there is the
tendency to use strains for load identification (as an alternative for using acceleration measurements combined with
matrix inversion or mount stiffness methods); in gas turbine blade testing, vibration levels and resonance
frequencies are typically obtained from strain measurements at the root of the blades; in helicopter or wind turbine
blade operational testing, accelerometers are often not practical so strain sensors are used; in wind-tunnel testing
frequently a combination of strains (at “clamped” locations on the wind tunnel structural model) and acceleration (at
“flexible” locations) are used [1].
The use of strain sensors for modal analysis was already discussed in [2,3]. The increased interest from both
industry and academia on assessing and evaluating structural integrity on design prototype stages and also
monitoring in real-time (with Structural Health Monitoring (SHM) systems), has led to an increase in the number of
dynamic strain applications, to the development of improved identification and measurement techniques, as well as
to improved sensor technology [4]. Strain gauges have been commonly used for static load testing of mechanical
products in the aeronautic, automotive and mechanical industry. Moreover, fatigue testing, durability analysis and
lifetime prediction have also been applications where strain gauges are used [5,6,7,8]. This sort of testing is a
common part of the product development process, and additional information on product durability and dynamic
performance can be assessed by obtaining the modal parameters of the system, while still using the same
instrumentation.
Another application of dynamic strain measurements is related to the strain displacement relations [9,10]. In
many systems, strain gauges are used as the standard vibration sensor, especially when size or sensor location is an
issue. Such is the case in aerospace applications, like gas turbines, wind turbines and helicopters [11], where size
and weight are very restricted, and any sensor place on a blade should affect its aerodynamic properties as little as
possible. One particular use of the strain measurements and strain to displacement relations is the strain pattern
analysis (SPA), where strain measurements are used to predict blade displacements.
A very important contribution to the field of strain measurement is the development of optical fiber sensors, of
which fiber Bragg grating (FBG) sensors are the most widespread [12,13]. Advantages of optical fiber sensors are
insensitivity to electromagnetic interference, small dimensions, light-weight, multiplexing capabilities, resistance to
11th International Conference on Vibration Measurements by Laser and Noncontact Techniques - AIVELA 2014
AIP Conf. Proc. 1600, 39-49 (2014); doi: 10.1063/1.4879567
© 2014 AIP Publishing LLC 978-0-7354-1234-7/$30.00

39 to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

This article is copyrighted as indicated in the article. Reuse of AIP content is subject
151.5.180.15 On: Wed, 25 Jun 2014 11:05:38
 corrosion and, in case of the need for embedding sensors: the compatibility of these sensors with the manufacturing
process of fibrous composite materials [14].
FBG sensors have already been used in multiple industrial applications (see [15] for an overview), for instance
for Structural Health Monitoring of civil structures (bridges, buildings, towers, dams, tunnels) or other special
structures (e.g. cranes) during construction, at commissioning, during service life. Sometimes, hundreds of sensors
are deployed for long term dynamic or quasi-static measurements of different physical parameters (vibrations,
strains, temperature, …) in extremely harsh environments [16]. Also for wind turbine monitoring, the use of FBG
sensors has been reported [17]: monitored components include blades (transport, operation), tower, foundations.
FBG sensors are also applied to large-scale electric power generators for vibration and temperature monitoring,
analysis and alarming. In the oil and gas sector, they are used for down-hole pressure and temperature monitoring as
well as for offshore platform monitoring [18]. FBG sensors have also been used as alternative for electrical strain
gauges during vibration environmental testing of light-weight space structures [19]. In the same field, FBG sensors
combined with modal analysis constitute an SHM system for composite aerospace structures that proofs to be useful
during ground qualification testing and on-board during the mission [20,21].
The paper is organized as follows. In next section, the FBG measurement principle will be reviewed. Afterwards,
the concept of strain modal analysis is introduced and, finally, a helicopter main rotor blade vibration testing and
analysis case study is presented.

FBG STRAIN MEASUREMENT PRINCIPLE

The fiber Bragg grating (FBG) is a periodic microstructure that acts as a wavelength selective mirror. This means
that if light from a broadband source is injected in the optical fiber, only light within a very narrow spectral width,
centered at the Bragg wavelength, will be back-reflected by the grating. The remaining light will continue its way
through the optical fiber without experiencing any loss (FIGURE 1). The fiber Bragg grating is a symmetric
structure, so it will always reflect light at the Bragg wavelength no matter which side the light is coming from.
The Bragg wavelength (ɉ஻ ) is essentially defined by the period of the microstructure (Ȧ) and the index of
refraction of the core (݊ୣ୤୤ ):

ɉ஻ ൌ ʹ݊ୣ୤୤ Ȧ (1)

A fiber Bragg grating has unique characteristics to perform as a sensor. For example, when the fiber is stretched
or compressed, the FBG will measure strain. This happens essentially because the deformation of the optical fiber
leads to a change in the period of the microstructure and, consequently, of the Bragg wavelength. There is also some
contribution from the variation of the index of refraction, through the photo-elastic effect.
Sensitivity to temperature is also intrinsic to a fiber Bragg grating. In this case, the main contributor to Bragg
wavelength change is the variation of the silica refraction index, induced by the thermo-optic effect. There is also a
contribution from the thermal expansion, which alters the period of the microstructure. This effect is, however,
marginal given the low coefficient of thermal expansion of silica.

FIGURE 1. A Fiber Bragg Grating structure, with refractive index profile and spectral response (Source: Wikipedia).

40 to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

This article is copyrighted as indicated in the article. Reuse of AIP content is subject
151.5.180.15 On: Wed, 25 Jun 2014 11:05:38
 FBG Strain Sensitivity

The strain dependence of a fiber Bragg grating can be determined by differentiating the wavelength:

ȟɉ஻ ȟሺ݊ୣ୤୤ Ȧሻ ͳ ߲݊ୣ୤୤

ൌ ൌ ൬ͳ ൅ ൰ ȟɂ ൌ ሺͳ ൅ ‫݌‬௘ ሻȟɂ ൌ Ⱦக ȟɂ (2)
ɉ஻ ݊ୣ୤୤ Ȧ ݊ୣ୤୤ ߲ɂ

where Ⱦக is the strain sensitivity of the Bragg grating; ‫݌‬௘ is the photo-elastic constant (i.e. variation of the index of
refraction with axial tension), which is for the optical fiber: ‫݌‬௘ ൎ െͲǤʹͳʹ, meaning that strain sensitivity of a FBG is
given by the expression:

ȟɉ஻
ൌ Ⱦக ɉ஻ ൌ ͲǤ͹ͺͺɉ஻ (3)
ȟɂ
So, for an FBG with ɉ஻ ൌ ͳͷͷͲ, the following strain sensitivity holds:

ȟɉ஻ ’
ൌ ͳǤʹ (4)
ȟɂ Ɋɂ

FBG Temperature Sensitivity

Similarly to the strain dependence of a fiber Bragg grating, the temperature dependence can be determined by
differentiating the wavelength expression (1):

ȟɉ஻ ȟሺ݊ୣ୤୤ Ȧሻ ͳ ߲Ȧ ͳ ߲݊ୣ୤୤

ൌ ൌ൬ ൅ ൰ ȟܶ ൌ ሺȽ ൅ Ƀሻȟܶ ൌ Ⱦ ் ȟܶ (5)
ɉ஻ ݊ୣ୤୤ Ȧ Ȧ ߲ܶ ݊ୣ୤୤ ߲ܶ

where Ⱦ ் is the thermal sensitivity of the Bragg grating; Ƚ is the coefficient of thermal expansion of the fiber; Ƀ is
the thermo-optic coefficient (i.e. variation of the index of refraction with temperature). For a temperature sensitivity
approximation, it can be assumed that these values are constant for the temperature range of interest (below ʹͲͲι):
Ƚ ൌ ͲǤͷͷ ൈ ͳͲି଺ Ȁι and Ƀ ൌ ͷǤ͹͹ ൈ ͳͲି଺ Ȁι, meaning that the approximate temperature sensitivity is given by:

ȟɉ஻
ൌ Ⱦ ் ɉ஻ ൌ ͸Ǥ͵ʹɉ஻ (6)
ȟܶ
So, for an FBG with ɉ஻ ൌ ͳͷͷͲ, the following temperature sensitivity holds:

ȟɉ஻ ’
ൌ ͻǤͺ (7)
ȟܶ ι

Optical data acquisition system

Next to the sensors, also an optical data acquisition system is required. The FiberSensing FS2200 Industrial
BraggMETER (FIGURE 2) is specifically designed to interrogate FBG-based sensors in industrial environments.
The BraggMETER employs tunable laser sweeping and includes a NIST traceable wavelength reference that
provides continuous calibration to ensure system accuracy over long-term operation. The high dynamic range and
output power allow high resolution to be attained even for long fiber leads and fibers with optical connectors where
losses are expected. Up to 4 parallel optical channels are available allowing the simultaneous measurement of a
large number of sensors, up to 500 Hz. The recommended maximum number of sensors per channel is 25,
considering sensor wavelengths equally spaced over the operating range and maximum measurement range of
േ2 nm per sensor.

41 to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

This article is copyrighted as indicated in the article. Reuse of AIP content is subject
151.5.180.15 On: Wed, 25 Jun 2014 11:05:38
 Optical
Fiber Coupler Circulator Sensor 2
Tunable
Laser
Sensor 1 Sensor 3

O O B2 > O B3 > O B1

Gas-Cell

t
PD PD V

Wavelength Sensor 1 3 2 1 3 2 t
Reference Signals

FIGURE 2. Operating principle of the FiberSensing BraggMETER multipurpose measurement unit for FBG sensors. (Insert)
Photo of the system.

STRAIN MODAL ANALYSIS

In previous sections in was explained how optical strain data can be acquired. In this section, it will be discussed
how this data can be analyzed, leading to a dynamic characterization of the tested structure by means of
experimental modal analysis (EMA). EMA is a well-known technology that relies most typically on force and
acceleration measurements. The use of strain data in an EMA context will be explained in the following.
To obtain the strain modal formulation, one can start with the fundamental theory of modal analysis. Modal
theory states that for instance the displacement ‫( ݑ‬in ‫ ݔ‬direction) of a given coordinate ‫ ݔ‬can be approximated by the
summation of ܰ௠ modal contributions:
ே೘

‫ݑ‬ሺ‫ݔ‬ǡ ‫ݐ‬ሻ ൌ ෍ ߶௜ ሺ‫ݔ‬ሻ‫ݍ‬௜ ሺ‫ݐ‬ሻ (8)

௜ୀଵ

th
where ߶௜ ሺ‫ݔ‬ሻ is the ݅ (displacement) vibration mode shape at location ‫ݔ‬, and ‫ݍ‬௜ ሺ‫ݐ‬ሻ is the generalized modal
coordinate at time instant ‫ݐ‬. For small displacements, given the theory of elasticity, the strain (ɂ௫ ) - displacement (‫)ݑ‬
relation is:

߲
ɂ௫ ሺ‫ݔ‬ǡ ‫ݐ‬ሻ ൌ ‫ݑ‬ሺ‫ݔ‬ǡ ‫ݐ‬ሻ (9)
߲‫ݔ‬
By applying this spatial derivative to (8) and defining the strain vibration mode shape as:

߲
߰௜ ሺ‫ݔ‬ሻ ൌ ߶ ሺ‫ݔ‬ሻ (10)
߲‫ ݔ‬௜
the following strain modal decomposition is obtained:
ே೘

ɂ௫ ሺ‫ݔ‬ǡ ‫ݐ‬ሻ ൌ ෍ ߰௜ ሺ‫ ݔ‬ሻ‫ݍ‬௜ ሺ‫ݐ‬ሻ (11)

௜ୀଵ

Due to the similarity of the strain modal formulation and the displacement modal formulation, the Strain
Frequency Response Function (SFRF) ‫ ݈݇ܪ‬ሺ‫ݏ‬ሻ between a (translation) force at coordinate ݈ and a strain at coordinate
݇ can be written as:
ே೘
߰௜ ሺ݇ሻ߶௜ ሺ݈ሻ ߰௜‫ כ‬ሺ݇ሻ߶௜‫ כ‬ሺ݈ሻ
‫ܪ‬௞௟ ሺ‫ݏ‬ሻ ൌ ෍ ቆ ൅ ቇ (12)
‫ ݏ‬െ ɉ௜ ‫ ݏ‬െ ɉ‫כ‬௜
௜ୀଵ

42 to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

This article is copyrighted as indicated in the article. Reuse of AIP content is subject
151.5.180.15 On: Wed, 25 Jun 2014 11:05:38
 where ‫ ݏ‬ൌ ݆ɘ is the Laplace variable; ɉ௜ the ݅th pole of the system; and ሺǤ ሻ‫ כ‬is the complex conjugate operator.
In classical force-displacement modal analysis, ߰௜ would be substituted by ߶௜ in (12). Again, due to this similarity,
the classical modal parameter estimation procedures can be followed, except for the notions driving point FRF and
reciprocity (unlike the classical FRF matrix, the SFRF matrix is not symmetric). More details can be found in [22].

CASE STUDY: HELICOPTER BLADE VIBRATION MEASUREMENT AND ANALYSIS

FIGURE 3. The PZL W-3 Sokół (Polish for “Falcon”) is a Polish medium-size, twin-engine, multipurpose helicopter
manufactured by PZL-Świdnik (now AgustaWestland Świdnik) (©PZL-Świdnik).

A main rotor blade of the PZL W-3 helicopter (FIGURE 3) was selected as a test object to verify the
performance of the FBG strain sensors and compare them with classical electrical strain gauges. A vibration test was
set up hereto. The blade was flexibly suspended (approaching free-free conditions: i.e. the rigid body frequencies are
much lower than the frequencies of the first deformation modes of the structure) by means of bungees (FIGURE 4 –
Left). The blade is excited by an electrodynamic shaker located at an outer point of the blade. The same blade was
already the subject of a previous study on new dynamic measurement techniques (though not involving FBG sensors
at that time) [23]. In [24], the blade was used to validate modal parameter-based Structural Health Monitoring
concepts.
2 Optical fibers of 10 sensors each are glued to the blade. The curing took 8-10 hours (overnight). The sensors
could have been directly inscribed in the long fiber, but in this case, multiple individual FBG sensors have been
spliced (i.e. fused together) on-site into 1 fiber. Next to the 20 FBG sensors, also 4 classical electric strain gauges
have been used, as well as 16 accelerometers (FIGURE 5). The electrical strain gauges in quarter-bridge
configuration are positioned close to FBGs (FIGURE 4 – Right) for comparison and for post-synchronizing the
signals because two data acquisition systems have been used. The FBGs have been measured by the FiberSensing
BraggMETER whereas the electrical signals (shaker force, acceleration responses, and electrical strain gauge
responses) have been acquired by the LMS SCADAS front-end. The accelerometers are distributed along the
structure for characterizing the identified mode shapes.
A post-synchronization procedure was elaborated based on a pair of collocated sensors (an electrical and an FBG
strain sensor in this case). This procedure is essential for calculating Frequency Response Functions (FRFs) between
a force input measured by one system and an (optical) strain response measured by another system. Inaccuracies in
aligning both signals have a significant impact on the FRF quality. Evidently, these synchronization challenges can
also be tackled at hardware level. The FRFs have been estimated using the so-called ‫ܪ‬ଵ estimator:

ܵመ௬௨ ሺɘሻ
෡ ሺɘሻ ൌ
‫ܪ‬ (13)
ܵመ௨௨ ሺɘሻ

43 to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

This article is copyrighted as indicated in the article. Reuse of AIP content is subject
151.5.180.15 On: Wed, 25 Jun 2014 11:05:38
 where ܵ෠ ‫ ݑݕ‬ሺɘሻ is the cross spectrum estimate between output (in this case strain or acceleration) and input (in this
case force) and ܵ෠ ‫ ݑݑ‬ሺɘሻ is the input power spectrum estimate. These spectra are calculated as:
ே್
ͳ
ܵመ௬௨ ሺɘሻ ൌ ‫ כ‬ሺ ሻ
෍ ܻሺ௜ሻ ሺɘሻܷሺ௜ሻ ɘ (14)
ܰ௕
௜ୀଵ

with ܻሺ݅ሻ ሺɘሻ the Fourier transform of output time data block ݅ and ܾܰ the number of data blocks (the
number of averages). The coherence ɀʹ is a quality indicator of the FRF. It is a value between Ͳ (low-
quality) and ͳ (high-quality) and can be computed as:
ଶ
หܵመ௬௨ ሺɘሻห
ɀଶ ሺɘሻ ൌ (15)
ܵመ௬௬ ሺɘሻܵመ௨௨ ሺɘሻ

FIGURE 4. (Left) Helicopter main rotor blade flexibly supported; (Right) (Nearly) collocated strain sensors on the helicopter
blade: optical FBG and electrical.

Elec. strain gauge FBG sensor Accelerometer Excitation point

20 19 18 17 16 15 14 13 12 11

SG 36 SG 34
SG 35 SG 33

10 9 8 7 6 5 4 3 2 1

FIGURE 5. Sensors used for the blade vibration test: electrical strain gauges, FBG sensors, accelerometers, and force cell.

As dynamic excitation signals, both sine sweeps from 9 to 80 Hz and burst random (broadband signal with
frequency range 7-80 Hz) have been used. The excitation signals have been repeated 20 times, allowing for
averaging in the FRF estimation procedure (13). The strain time domain responses to both excitation signals are
represented in FIGURE 6. After synchronization, the collocated FBG and electrical strain gauges signals agreed
very well. FIGURE 7 shows the estimated acceleration-over-force and strain-over-force FRFs (13) and related
coherences (15). Both acceleration and strain responses yield high-quality FRFs. The structural resonances can be
observed as peaks in both types of FRFs. It is typical that the acceleration peaks have similar amplitudes across the
entire frequency range, whereas the strain peaks decrease with the square of the frequency. In FIGURE 8, a
comparison is made between electrical strain gauge FRFs and FBG strain FRFs at (nearly) collocated points (see
FIGURE 5 for locations). The agreement is excellent. Small differences can be explained by the fact that both
sensors are not at exactly the same location.

44 to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

This article is copyrighted as indicated in the article. Reuse of AIP content is subject
151.5.180.15 On: Wed, 25 Jun 2014 11:05:38
 42.00 60.00
Time FBG:4:S Time FBG:4:S
Real Time Point33:S Time Point33:S

Real
uE

uE
-42.00 -60.00
18.00 s 35.00 19.40 s 38.00
27.00 43.00
Time FBG:4:S Time FBG:4:S
Time Point33:S Time Point33:S
Real

Real
uE

uE
-27.00 -40.00
25.40 s 26.40 20.50 s 21.50

FIGURE 6. Comparison between FBG and electrical strain gauge time series: (Left) sine sweep excitation; (Right) burst random
excitation. (Top) 1 block; (Bottom) 1 s zoom.

0.00 1.00 20.00 1.00

Amplitude

Amplitude
uE/N
g/N
dB

dB
/

/
F FRF blade1:+X/Point1:-X F FRF Point34:S/Point1:-X
B Coherence blade1:+X/Point1:-X B Coherence Point34:S/Point1:-X
-70.00 0.00 -60.00 0.00
5.00 Hz 80.00 5.00 Hz 80.00
FIGURE 7. Burst random FRF amplitude and coherence: (Left) acceleration FRF; (Right) strain FRF.

45 to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

This article is copyrighted as indicated in the article. Reuse of AIP content is subject
151.5.180.15 On: Wed, 25 Jun 2014 11:05:38
 40.00 30.00
FRF Point33:S/Point1:-X FRF Point35:S/Point1:-X
FRF FBG:4:S/Point1:-X FRF FBG:7:S/Point1:-X
uE/N

uE/N
dB

dB
-50.00 -70.00
180.00 180.00
Phase

Phase
°

°
-180.00 -180.00
5.00 Hz 80.00 5.00 Hz 80.00
30.00 30.00
FRF Point34:S/Point1:-X FRF Point36:S/Point1:-X
FRF FBG:16:S/Point1:-X FRF FBG:19:S/Point1:-X
uE/N

uE/N
dB

dB

-60.00 -50.00
180.00 180.00
Phase

Phase
°

-180.00 -180.00
5.00 Hz 80.00 5.00 Hz 80.00
FIGURE 8. Burst random results. Comparison between electrical strain gauge FRFs and FBG strain FRFs at (nearly) collocated
points (see FIGURE 5 for locations).

The PolyMAX method [25] has been applied to both acceleration and FBG strain FRFs to estimate the modal
parameters of the blade. The stabilization diagram is represented in FIGURE 9; the estimated modal parameters can
be found in TABLE 1. Some mode shapes are shown in FIGURE 10. The acceleration (displacement) part of a mode
shape is represented as a deformation; the strain part of a mode shape is represented as coloring. Both types of
sensors (acceleration and strain) provide complementary information: e.g. the in-plane bending mode at 14.0 Hz is
not picked up very well by the out-of-plane accelerometers, but nicely captured by the strain sensors. The torsion
mode at 30.4 Hz on the other hand is nicely captured by the accelerometers, but the strain image does not look too
meaningful. Finally, FIGURE 11 indicates the quality of the modal parameter estimation process: the FRFs that are
synthesized based on the identified modal parameters correspond very well to the measured ones; both for
acceleration as for FBG strain sensors.

46 to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

This article is copyrighted as indicated in the article. Reuse of AIP content is subject
151.5.180.15 On: Wed, 25 Jun 2014 11:05:38
 FIGURE 9. PolyMAX applied to the FBG strain FRFs – stabilization diagram.

TABLE 1. Identified modal parameters.

Mode Eigenfrequency Damping ratio Description
[Hz] [%]
1 10.3 0.50 Bending 2
2 14.0 0.84 In-plane bending 1
3 20.3 0.37 Bending 3
4 30.4 0.98 Torsional 1
5 33.8 0.40 Bending 4
6 37.6 0.66 In-plane bending 2
7 49.5 0.52 Bending 5
8 61.2 0.93 Torsional 2
9 67.2 0.43 Bending 6
10 75.9 0.74 In-plane bending 3

1 2

3 4

FIGURE 10. Experimental mode shapes identified by applying PolyMAX to combined acceleration-strain data: the acceleration
(displacement) part of a mode shape is represented as a deformation; the strain part of a mode shape is represented as coloring.

47 to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

This article is copyrighted as indicated in the article. Reuse of AIP content is subject
151.5.180.15 On: Wed, 25 Jun 2014 11:05:38
 0.00 30.00

uE/N
g/N
dB

dB
FRF blade:1:+X/Point1:-X FRF FBG:16:S/Point1:-X
Synthesized FRF blade:1:+X/Point1:-X Synthesized FRF FBG:16:S/Point1:-X
-70.00 -60.00
180.00 180.00
Phase

Phase
°

°
-180.00 -180.00
5.00 Hz 80.00 5.00 Hz 80.00
FIGURE 11. Comparison between measured and synthesized FRFs. (Left) acceleration FRF; (Right) strain FRF.

CONCLUSIONS

This paper discussed the use of optical fiber Bragg grating (FBG) strain sensors for structural dynamics
measurements and experimental modal analysis. The practical application revealed the ease of installation of FBG
sensors when compared to classical electrical strain gauges: 1 fiber contained 10 sensors, there was no need for
soldering, and a large number of sensors can be installed and connected fast. It is also important to consider an
integrated optical-electrical data acquisition system for synchronous heterogeneous sensor data measurement.
The helicopter main rotor blade case study clearly demonstrated that high-quality strain measurements can be
performed and accurate modal analysis results can be obtained by using FBG sensors.

ACKNOWLEDGMENTS

Fábio Luis Marques dos Santos, author of this paper, is an Early Stage Researcher at LMS International, under
the FP7 Marie Curie ITN project “IMESCON” (FP7-PEOPLE-2010-ITN, Grant Agreement No. 264672).

REFERENCES

1. B. Peeters, P. Karkle, M. Pronin, R. Van der Vorst, “Operational Modal Analysis for in-line flutter assessment during wind
tunnel testing” in Proc. of IFASD 2011, the Int. Forum on Aeroelasticity and Structural Dynamics, Paris, France, 2011.
2. O. Bernasconi, D.J. Ewins, “Application of strain modal testing to real structures” in Proc. of IMAC 6, 1988.
3. L.M. Vári, P.S. Heyns, “Using strain modal testing” in Proc. of IMAC 12, 1994.
4. G.W. Reich, K.C. Park, “A theory for strain-based structural system identification” in Journal of Applied Mechanics,
68(4):521–527, 2001.
5. C. Liefooghe, H. Van der Auweraer, P. Janssen, P. Sas, “Validation of modal filtering/editing approach to dynamic fatigue
analysis” in Proc. of ISMA 17, 1992.
6. H. Wentzel, “Fatigue test load identification using weighted modal filtering based on stress” in Mechanical Systems and
Signal Processing, 40:618–627, 2013.
7. D. Vandepitte, P. Sas, “Case study of fracture of a mechanical component due to resonance fatigue” in Mechanical Systems
and Signal Processing, 4(2):131–143, 1990.
8. T. Uhl, W. Lisowski, J. Malecki, “Strain modal testing of a helicopter tail boom” in Proc. of IMAC 16, 1998.
9. G.C.Foss, E.D. Haugse, “Using modal test results to develop strain to displacement transformations” in Proc. of IMAC 13,
1995.
10. A.C. Pisoni, C. Santolini, D.E Hauf, S. Dubowsky, “Displacements in a vibrating body by strain gage measurements” in
Proc. of IMAC 13, 1995.
11. N. Tourjansky, E. Szechenyi, “The measurement of blade deflections-a new implementation of the strain pattern analysis” in
Proc. of the 13th European Rotorcraft Forum, Avignon, France, 1992.
12. L.-H. Kang, D.-K. Kim, J.-H. Han, “Estimation of dynamic structural displacements using fiber Bragg grating strain sensors”
in Journal of Sound and Vibration, 305(3):534–542, 2007.

48 to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

This article is copyrighted as indicated in the article. Reuse of AIP content is subject
151.5.180.15 On: Wed, 25 Jun 2014 11:05:38
 13. H.-Y. Ling, K.-T. Lau, L. Cheng, “Determination of dynamic strain profile and delamination detection of composite
structures using embedded multiplexed fibre-optic sensors” in Composite Structures, 66(1):317–326, 2004.
14. G. Luyckx, E. Voet, N. Lammens, J. Degrieck, “Strain measurements of composite laminates with embedded fibre Bragg
gratings: criticism and opportunities for research” in Sensors, 11:384-408, 2011.
15. FiberSening, http://www.fibersensing.com/, accessed on March 2014.
16. C. Barbosa, N. Costa, L. Ferreira, F. Araújo, F. Paredes, A. Rodriguez, L. Bussion, “Fibre optic based sensing network for
monitoring construction and service life of Deza and Anzo II bridges” in Proc. of SHMII-4, the 4th Int. Conf. on Structural
Health Monitoring on Intelligent Infrastructure, , Zürich, Switzerland, 2009.
17. T.J. Arsenault, A. Achuthan, P. Marzocca, C. Grappasonni, G. Coppotelli, “Development of a FBG based distributed strain
sensor system for wind turbine structural health monitoring” in Smart Materials and Structures, 22(7), 2013.
18. L. Sun, H.-N. Li, L. Ren, Q. Jin, “Dynamic response measurement of offshore platform model by FBG sensors” in Sensors
and Actuators A: Physical, 136(2): 572–579, 2007.
19. A. Grillenbeck, P.J. Hein, O. Reichmann, A. Ihle, “Novel test techniques for lightweight and thermally stable structures” in
Proc. 12th Eur. Conf. on Spacecraft Structures, Materials & Environmental Testing, ESTEC, Noordwijk, The Netherlands,
2012.
20. N. Eaton, “Monitoring Dynamic Response of Composite Sandwich Panels Using New Optical Fibre Sensor Approach” in
Proc. DYNACOMP, the 1st Int. Conf. on Composites Dynamics, Arcachon, France, 2012.
21. A. Panopoulou, S. Fransen, V. Gomez Molinero, V. Kostopoulos, “Experimental modal analysis and dynamic strain fiber
Bragg gratings for structural health monitoring of composite aerospace structures” in Proc. 12th Eur. Conf. on Spacecraft
Structures, Materials & Environmental Testing, ESTEC, Noordwijk, The Netherlands, 2012.
22. F. Santos, B. Peeters, M. Menchicchi, J. Lau, L. Gielen, W. Desmet, L. Goes, “Strain-based dynamic measurements and
modal testing” in Proc. IMAC 32, Orlando (FL), USA, 2014.
23. M. Luczak, K. Dziedziech, M. Vivolo, W. Desmet, B. Peeters, H. Van der Auweraer, “Contact versus non-contact
measurement of a helicopter main rotor composite blade” in Proc. 9th Int. Conf. on Vibration Measurements by Laser and
Noncontact Techniques, Ancona, Italy, 2010.
24. F. Santos, B. Peeters, H. Van der Auweraer, L.C.S. Góes, “Experimental damage detection of a helicopter main rotor blade
based on modal properties” in Proc. ICEDyn 2013, Int. Conf. on Structural Engineering Dynamics, Sesimbra, Portugal, 17-19
June 2013.
25. B. Peeters, H. Van der Auweraer, P. Guillaume, J. Leuridan, “The PolyMAX frequency-domain method: a new standard for
modal parameter estimation?” Shock and Vibration 11:395–409, 2004.

49 to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

This article is copyrighted as indicated in the article. Reuse of AIP content is subject
151.5.180.15 On: Wed, 25 Jun 2014 11:05:38
View publication stats