Detection of acoustic emission waves in composite plates by fiber optic sensors

Abstract

The paper provides an analysis of acoustic emission signals recorded with fiber optic sensors during the propagation of ultrasonic waves in a polymer composite material. The fiber optic sensors for acoustic emission were constructed according to the scheme of an adaptive holographic interferometer. Unlike piezoelectric sensors, fiber optic sensors are distributed type sensors. This imposes certain features on the detection of signals in plates in which fiber optic sensors are embedded. It is established that the spectrum of acoustic emission signals differs in different directions of wave propagation. The local maxima of the spectrum are determined by the mode of wave propagation in the plate in different directions and the location of fiber optic sensors.

Full text

Detection of acoustic emission waves in composite plates by fiber optic sensors
O. V. Bashkov, R. V. Romashko, H. Khon, M. N. Bezruk, V. I. Zaikov, and I. O. Bashkov
Citation: AIP Conference Proceedings 1909, 020013 (2017);
View online: https://doi.org/10.1063/1.5013694
View Table of Contents: http://aip.scitation.org/toc/apc/1909/1
Published by the American Institute of Physics

Detection of Acoustic Emission Waves in Composite
Plates by Fiber Optic Sensors
O. V. Bashkov1,a), R. V. Romashko2,b), H. Khon1,c), M. N. Bezruk2,d),
V. I. Zaikov1, and I. O. Bashkov1,e)
1 Komsomolsk-on-Amur State Technical University, Komsomolsk-on-Amur, 681013 Russia
2 Institute of Automation and Control Processes FEB RAS, Vladivostok, 690041 Russia
a) bashkov_ov@mail.ru
b) romashko@iacp.dvo.ru
c) Corresponding author: khonhanhtooaung@gmail.com
d) bezmisha@list.ru
e) bashkovilya@mail.ru
Abstract. The paper provides an analysis of acoustic emission signals recorded with fiber optic sensors during the
propagation of ultrasonic waves in a polymer composite material. The fiber optic sensors for acoustic emission were
constructed according to the scheme of an adaptive holographic interferometer. Unlike piezoelectric sensors, fiber optic
sensors are distributed type sensors. This imposes certain features on the detection of signals in plates in which fiber optic
sensors are embedded. It is established that the spectrum of acoustic emission signals differs in different directions of
wave propagation. The local maxima of the spectrum are determined by the mode of wave propagation in the plate in
different directions and the location of fiber optic sensors.
INTRODUCTION
For safety needs, more attention should be given to the analysis of technical conditions of transport and
engineering devices at hazardous production facilities. For continuous monitoring of critical technical conditions,
structural health monitoring (SHM) systems are used [1, 2]. One of the newest technologies for diagnostics of
materials is acoustic emission. In our country, the use of acoustic emission is restrained due the cost of equipment
and to the lack of data to prove the reliability of such monitoring. The cost of equipment is determined by the
number of applied piezoelectric sensors and modules of acoustic emission detection and can amount up to several
hundred thousand dollars. The reliability is determined by the experience of operators and capabilities of monitoring
software. Now, technologies are developed for monitoring the stress-strain state of objects for safety. One of the
solutions is to use fiber optic sensors [3, 4]. The cost of an optical fiber does not exceed the cost of an electrical
cable for transmission of acoustic emission signals.
Fiber optic sensors with a Bragg diffraction lattice are widely used for strain estimations [5, 6]. However, they
are inapplicable for acoustic emission because of their low sensitivity. Different schemes based on interferometers
are applied to increase the sensitivity of sensors. Adaptive laser interferometers can provide high sensitivity. The
construction of sensors is based on two-wave interaction of laser radiation on a dynamic hologram, which is formed
in a photorefractive crystal. The advantage of fiber optic sensors is the possibility of their integration into the control
of an object. The active use of fibrous composite materials leads to the necessity of monitoring the technical
condition of composite objects. Fiber optic sensors can be embedded in composite materials, so they have a
similarity as such as in the structure of fibers. Here we study the possibility of detecting acoustic waves by fiber
optic sensors embedded into a PCM plate.
Proceedings of the International Conference on Advanced Materials with Hierarchical Structure for New Technologies and Reliable Structures 2017 (AMHS’17)
AIP Conf. Proc. 1909, 020013-1–020013-4; https://doi.org/10.1063/1.5013694
Published by AIP Publishing. 978-0-7354-1601-7/$30.00
020013-1

FIGURE 1. Position of sensors on the PCM plate
MATERIALS AND EXPERIMENT
The PCM sample was made of twelve layers of glass fabric by the vacuum autoclave method. The optical fibers
had a diameter of 0.125 mm and were located between the layers during the laying of the fiberglass [7]. In the PCM
sample, four optical fibers are located: two lengthwise (horizontal) and two crosswise (vertical). They were
connected to the optic scheme of an adaptive interferometer (Fig. 1). The plate size was 290 × 130 mm.
The optical fibers were acoustic emission sensors. They were built into the optical scheme of an adaptive
holographic interferometer. The interferometer is implemented according to the scheme of two-beam interaction of
laser radiation on a dynamic hologram in a photorefractive crystal [8]. The use of a photorefractive crystal allows
one to decrease the damping of an operating point, which can affect the noise immunity and sensitivity of the
adaptive interferometer compared to the use of conventional Mach Zehnder interferometers. For excitation of
acoustic waves, a pencil break from a Hsu–Nielsen source was used (Fig. 1).
RESULTS AND DISCUSSION
The signals generated in the plate can be sensed due to the capability of sensors located, however, in the far field.
This affects the accuracy of changing the recording time for the location of AE sources. The technology of PCM
manufacturing with a certain direction of laying fibers provides anisotropic properties of the composite material.
The sound wave characteristics are also different. As has been noted [8, 9], the velocity of sound differs in vertical
and horizontal directions. The sound wave velocity in a horizontal direction is higher than that in a vertical one.
However, during the propagation of a sound wave, a group of Lamb waves also propagate in this plate, being a
family of symmetric and antisymmetric waves. The difference in acoustic channel characteristics can be seen on
sound wave spectra. Piezoelectric sensors, which are commonly used in acoustic emission, significantly distort the
sound wave spectrum due to the variability of frequency response. Fiber optic sensors with laser interferometers
have a rather stable frequency response. Adaptive interferometers on dynamic holograms also have a stable
frequency response at frequencies above the frequency which determines the time characteristic of a hologram. This
is one of the significant advantages of an adaptive interferometer. Another advantage of an adaptive interferometer
is that it works as a high-pass filter and can depress low-frequency attenuations. The filter frequency is determined
by the type of a photorefractive crystal in an adaptive interferometer. Due to the stable frequency response, fiber
optic sensors can identify the frequency characteristics of acoustic wave propagating in a monitored object.
Figure 2 shows AE signals and Fourier spectra recorded by fiber optic sensors 3 and 4 (Fig. 1). The Fourier
spectra display qualitative similarity, determining the characteristic of wave propagation in a given direction. The
spectra of both waves have a local maximum at a frequency of about 5 kHz. Its value varies with decreasing and
increasing the frequency with respect to the local maximum frequency 5 kHz.
Figure 3 shows AE signals and Fourier spectra recorded by fiber optic sensors 1 and 2 (Fig. 1). The Fourier
spectra, like the signals, differ from those detected by fiber optical sensors 3 and 4. The Fourier spectra of signals
recorded by fiber optic sensors 1 and 2 have several local maxima.
sensor no. 4 sensor no. 3
sensor no. 1
sensor no. 2
source
020013-2

(a) (b)
(c) (d)
FIGURE 2. AE signals (a, b) and Fourier spectra (c, d) recorded by fiber optic sensors 3 and 4, respectively
The highest maximum is at a frequency of 330 Hz. The next local maxima are located at frequencies of 1550,
3100, and 4950 Hz. Increasing the frequency decreases the amplitude of the local maxima according to the
exponential law. As the frequency reaches 15 kHz, the spectrum magnitude is actually reduced to zero. It is easy to
see that the maximum frequency of each following is a multiple of the frequency of the second harmonic 1550 Hz.
A careful analysis of the AE spectra recorded by fiber optic sensors 3 and 4 suggests the presence of a harmonic
increase. Local maxima are found at 890, 1780, 2620, 3450, 4170, and 4950 Hz. The maximum magnitude rises
from the lower to higher frequencies. The frequency of the first harmonic having the smallest magnitude is observed.
The spectral features are associated not only with the features of wave propagation in the plates but also with
their dimensions and sensor length. The acoustic wave propagating radially in all directions excites the waves of
various harmonics, which are determined by the longitudinal and transverse dimensions of the PCM plate.
(a) (b)
(c) (d)
FIGURE 3. AE signals (a, b) and Fourier spectra (c, d) recorded by fiber optic sensors 1 and 2, respectively
020013-3

For the vertical position of sensors 3 and 4, the length of the optical fibers built into the plate is smaller than that
for sensors 1 and 2. The smallest wavelength in the vertical direction (they are also harmonic) can determine the
maximum amplitude of the attenuations.
The maximum deformation of the optical fibers of the sensors, which determines the maximum signal level,
occurs when they are longitudinally stretched or compressed. For vertical wave propagation and location of the
sensors on the plate, the maximum magnitude is observed for waves with a regular number of the largest harmonic.
For horizontal wave propagation, the maximum magnitude is observed at the lowest harmonic.
CONCLUSION
Fiber optic adaptive holographic sensors are known as widely distributed sensors. The paper has demonstrated
the possibility of recording acoustic emission waves by fiber optic sensors embedded in a PCM plate. These
distributed sensors can detect signals associated with wave propagation in plates of finite dimensions. The spectra of
acoustic waves in longitudinal and transverse directions differ for the plate having different dimensions and different
directions. This is because the group Lamb wave excites various oscillations modes in the plate. The distribution of
oscillation amplitudes at different harmonics is determined in this case by the direction of wave propagation and by
the dimensions of a fiber optic sensor that senses an acoustic wave in a given direction.
ACKNOWLEDGMENTS
This project was support by the Russian Science Foundation (project No. 16-19-10149).
REFERENCES
1. F. G. Yuan, Structural Health Monitoring (SHM) in Aerospace Structures (Woodhead Publishing, 2016).
2. A. Kesavan, S. John, and I. Herszberg, J. Intel. Mat. Sys. Str. 19(1), 63–72 (2007).
3. R. Di Sante, Sensors 15(8), 18666–18713 (2015). doi 10.3390/s150818666
4. R. V. Romashko, Yu. N. Kulchin, and A. A. Kamshilin, J. Russ. Laser Res. 31, 55–60 (2010). doi
10.1007/s10946-010-9123-9
5. G. Meltz, W. W. Morey, and W. H. Glenn, Opt. Lett. 14, 823–825 (1989). doi 10.1364/OL.14.000823
6. A. Othonos, “Optical fiber sensor technology,” in Bragg Gratings in Optical Fibers: Fundamentals and
Applications (Springer, 2000), pp. 79–187.
7. O. V. Bashkov, A. E. Protsenko, A. A. Bryanskii, and R. V. Romashko, Mech. Compos. Mater. 1–8 (2017). doi
10.1007/s11029-017-9683-7
8. O. V. Bashkov, R. V. Romashko, V. I. Zaikov, S. V. Panin, M. N. Bezruk, K. Khun, and I. O. Bashkov, Russ.
J. Nondestruct. Test 53, 415–421 (2017). doi 10.1134/S1061830917060031
9. O. V. Bashkov, R. V. Romashko, V. I. Zaykov, A. E. Protsenko, M. N. Bezruk, and H. Htoo, “Detection of
acoustic emission signals in the polymer composite material by adaptive fiber-optic sensors,” in Proc. SPIE
10176, Asia-Pacific Conference on Fundamental Problems of Opto- and Microelectronics, 1017613
(2016/12/14). doi org/10.1117/12.2268226
020013-4