Available online at www.sciencedirect.com
ScienceDirect Procedia Engineering 87 (2014) 1243 – 1246
EUROSENSORS 2014, the XXVIII edition of the conference series
A real-time electronic system for automated impact detection on aircraft structures using piezoelectric transducers L. Capineria,*, A. Bullettia, M. Calzolaia, D. Francesconib a
Department of Information Engineering, University of Florence, Via S. Marta 3, 50139, Firenze, Italy b Thales Alenia Space Italia S.p.A, Strada Antica di Collegno, 253 – 10146, Torino, Italy
Abstract Impact damage is one of the major concerns in maintenance of structures built from composite materials. For example damages induced by impact in composite overwrapped pressure vessel (COPV) operating in space conditions required an innovative approach for structural health monitoring (SHM) with piezoelectric transducers. Automated impact detection and characterization on structures has been an elusive goal due to the transitory nature of the detectable signals involved. The work describes the development of a real-time electronic system for automated impact detection on a scale model of COPV using piezoelectric transducers connected to a mixed analog-digital electronics. © 2014 2014 Published The Authors. Published Elsevier Ltd.access article under the CC BY-NC-ND license by Elsevier Ltd. by This is an open Peer-review under responsibility of the scientific committee of Eurosensors 2014. (http://creativecommons.org/licenses/by-nc-nd/3.0/). Peer-review under responsibility of the scientific committee of Eurosensors 2014 Keywords:piezoelectric sensors; impact detection; aircraft; real-time electronic;
1. Introduction Damage detection in composite materials can be divided into active and passive approaches. The active approach is usually based on various non-destructive techniques utilizing actuators and receivers [1][2][3][4][5][6]. In contrast passive approaches do not involve any actuators; receivers are used to “sense and/or hear” any perturbations caused by possible hidden damage. The assumption is that damage occurs above well-defined energy of impacts. The acquisition method does not require any sophisticated instrumentation but relies on advanced signal processing. An
* Corresponding author. Tel.: +39 055 4796376; fax: +39 055 4769516. E-mail address:
[email protected]
1877-7058 © 2014 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/3.0/). Peer-review under responsibility of the scientific committee of Eurosensors 2014 doi:10.1016/j.proeng.2014.11.408
1244
L. Capineri et al. / Procedia Engineering 87 (2014) 1243 – 1246
array of piezoelectric sensors (piezopolymer and piezoceramic) has been used (see Fig. 1) to detect ultrasonic and acoustic waves generatedby an impact applied to the composite structure. In addition to detect an impact, is interesting to locate impact position with adequate precision respect to the smaller dimension of a significant defect. Conventional triangulation techniques fail to correctly predict the acoustic source location in anisotropic plates due to the direction dependent nature of the elastic wave speeds. To overcome this problem, Kundu et al. [7] and TaliehHajzargerbashi [8] proposed an alternative method for acoustic source prediction based on optimizing an objective function. They defined an objective function that uses the time of flight information of the acoustic waves to the passive transducers attached to the plate and the wave propagation direction from the source point to the receiving sensors. This objective function has been applied to our system basing on the minimization of the errors of the difference between time of flight of the ultrasonic signals from the impact point to transducers positions. a
b
Fig. 1. Piezoelectric transducers used for the impact detection on scale model of COPV. (a) piezopolymer interdigital transducer (IDT); (b) piezoceramic transducer [Acellent Technologies Inc].
2. The objective function implemented The general form of the objective function implemented for n transducers is:
n −1
E ( x0 , y 0 ) = ¦
n
¦ ( (t
i =1 j =i +1
i
−tj)−
( xi − x 0 ) 2 + ( y i − y 0 ) 2 v(θ i )
+
( x j − x0 ) 2 + ( yi − y 0 ) 2 v(θ j )
)2
(1)
Where n is the number of the transducers, (xi,yi) are the coordinates of the i-th transducer, (x0,y0) is the coordinates of the impact point, v(ϑ) is the propagation velocity of the ultrasonic wave and ti is the time-of-flight between the impact point and the i-th transducer.
1245
L. Capineri et al. / Procedia Engineering 87 (2014) 1243 – 1246
This function is minimized in the unknown impact point (x0, y0) so the impact point coordinates will be represent the minimum error E. 3. Laboratory Set-up A laboratory set-up for automated impact detection is shown in Fig. 2 and 3. The set-up is composed by a customized handmade impactor for generating low energy impacts on a scale model of COPV and 4 piezoelectric transducers bonded on the surface of the COPV. The real time electronic is composed by a transducer signal conditioning, a high performance ADC TI AFE5851 evaluation board (16 channel VGA, 12 bit, ADC) with TSW1250 FPGA board for data reading and transfer by a serial bus. The bandwidth of the received signals is about 20kHz @ -6dB with the maximum range of 1Vpp.
d
c
b a
Fig. 2. Laboratory set-up for automated impact detection. (a) customized handmade impactor; (b) scale model of COPV; (c) acquisition boards; (d) PC equipped with real-time algorithm.
UT4
UT3 IDT3
IDT4
(a)
UT1
UT2 IDT2
IDT1 500mm
Fig. 3. Transducers displacement on a scale model of COPV. IDT1-IDT4 are piezopolymer interdigital transducers; UT1-UT4 are piezoceramic transducers; (a) impact point.
4. Test results A real-time algorithm that implement the objective function has been developed. The output of the algorithm is a viewgram that shows the impact area and the prediction of the impact position. An example is shown in Fig. 4. In
1246
L. Capineri et al. / Procedia Engineering 87 (2014) 1243 – 1246
this example the monitoring area is 300mm x 200mm, the impact position coordinates are (100,140) mm and the localization error using piezopolymer transducers [1] or piezoceramic transducers connected to the electronics is about 20mm. A video of an experiment with the real time display of the estimated impact on a 3D model COPV will also presented.
;ďͿ ;ĂͿ ;ĐͿ
Fig. 4. Resulting viewgram as output of the real-time algorithm based on Matlab® software. (a) Impact point; (b) Impact area predicted by four piezopolymer transducers (IDT1-IDT4); (c) Impact area predicted by four piezoceramic transducers used (UT1-UT4).
4. Conclusions This paper presented a new real-time system for automated impact detection on aircraft structures using piezoelectric transducers. An objective function has been implemented to our system in order to assess the impact point caused by collision with other bodies or particle debris. The accuracy found with four transducers on a COPV is better than 20 mm. Finally, a multichannel high-speed data acquisition and a real-time algorithm processing was developed in order to display on a 3D model COPVthe estimated impact area. The validation of this technique with a laboratory set-up has pushed the scientific research toward the installation of two arrays of 8 transducers each on a full scale model of a COPV [9] and the integration of the real time impact detection in a new real-time diagnostic system capable to investigate the damages induced by an impact by active ultrasonic Lamb waves. References [1] A. Bulletti, M. Calzolai, L. Capineri, D. Francesconi, PiezopolymerInterdigital Transducers for a Structural Health Monitoring System, Sensors and Microsystems - Proceedings of the 17th National Conference, Brescia, Italy, 5-7 February 2013, 268 (2013) 9-12, DOI:10.1007/978-3-319-00684-0_2. [2] A. Bulletti, M. Calzolai, L. Capineri, D. Francesconi, Lamb Wave Ultrasonic System for Active Mode Damage Detection in Composite Materials, Chemical Engineering Transactions, 33 (2013), DOI:10.3303/CET1333097. [3] W. Ostachowicz, P. Kudela, M. Krawczuk, A. Zak, Guided Waves in Structures for SHM: The Time - domain Spectral Element Methods, Wiley Ed., Feb. 2012, ISBN: 978-0-470-97983-9. [4] D. Adams, Health Monitoring of Structural Materials and Components: Methods with Applications, Wiley Ed., Apr. 2007, ISBN: 978-0-47003313-5. [5] V. Giurgiutiu, Structural Health Monitoring: with Piezoelectric Wafer Active Sensors, Academic press, 2007, ISBN: 9780124186910. [6] W. Staszewski, C. Boller, G. R. Tomlinson, Health Monitoring of Aerospace Structures: Smart Sensor Technologies and Signal Processings, Wiley Ed., 2003, ISBN: 978-0-470-84340-6. [7] T. Kundu, S. Das, K. V. Jata, Point of Impact Prediction in Isotropic and Anisotropic Plates from the Acoustic Emission Data, J. Acoust. Soc. Am. Volume 122, Issue 4, pp. 2057-2066, 2007. [8] T. Hajzargarbashi, T. Kundu, S. Bland, An improved algorithm for detecting point of impact in anisotropic inhomogeneous plates, Ultrasonics, Vol 51,pp. 317 -324, 2011. [9] L.Capineri, A.Bulletti, M.Calzolai, P. Giannelli, D. Francesconi, Arrays of conformable ultrasonic Lamb wave transducers for structural health monitoring with real-time electronics, Eurosensors 2014 the XXVIII edition, Brescia (IT), sept. 7-10 2014.
