IEEE SENSORS JOURNAL, VOL. 15, NO. 2, FEBRUARY 2015 659 Structural Damage Identification in an Aluminum Composite Plate by Brillouin Sensing Aldo Minardo, Agnese Coscetta, Romeo Bernini, Ruben Ruiz-Lombera, Jesus Mirapeix Serrano, Jose Miguel Lopez-Higuera, and Luigi Zeni Abstract—We report the results of an experimental modal analysis aimed to estimate the location of mechanical changes in an aluminum composite panel. A distributed Brillouin optical fiber sensor was used to retrieve the mode shapes of the first two bending modes of the plate. We show that the position of the defect is retrieved from the strain mode shapes, with a resolution determined by the disposition of the fiber across the structure, as well as by the spatial resolution of the interrogation unit. Index Terms— Stimulated Brillouin scattering, distributed fiber sensor, Brillouin fiber sensing. I. I NTRODUCTION I N STRUCTURAL health monitoring (SHM), the exper- imental modal analysis (EMA) is the most common approach to retrieve dynamic parameters of a structure. This approach determines modal parameters of the structure (mode shapes, natural frequencies and modal damping) through the responses induced by known and sometimes unknown excitations [1]. Vibration-based modal analysis, typically using acceleration measurements, has the limitation that the resonant frequencies do not allow to detect accurately damage location, while the precision of identified mode shapes is often not sufficient for effective damage identification [2]. In contrast, curvature (strain) is more sensitive to dam- age, but unfortunately fails to work unless the area where strain sensor is fixed covers the damaged region. Therefore, a SHM strategy based on distributed strain sensing is partic- ularly attracting, as it allows detecting arbitrary and unfore- seen damages. Brillouin scattering offers the opportunity to perform fully distributed strain measurements with a single optical fiber. In particular, the Slope-Assisted Brillouin Optical Time-Domain Analysis (SA-BOTDA) allows distributed strain acquisitions at a rate as high as several tens of Hz. In brief, the method relies on the measurement of the interaction between a probe signal and a counter-propagating pump pulse, with the spectral shift between the two lightwaves biased at the midgain Manuscript received October 7, 2014; accepted October 16, 2014. Date of publication October 20, 2014; date of current version November 20, 2014. This work was supported by the European COST action TD1001–OFSESA. The associate editor coordinating the review of this letter and approving it for publication was Prof. Julian C. C. Chan. A. Minardo, A. Coscetta, and L. Zeni are with the Department of Indus- trial and Information Engineering, Seconda Università di Napoli, Aversa 81031, Italy (e-mail: aldo.minardo@unina2.it; agnese.coscetta@unina2.it; zeni@unina.it). R. Bernini is with the International Journal on Engineering Applications, National Research Council, Naples 80131, Italy (e-mail: bernini.r@irea.cnr.it). R. Ruiz-Lombera, J. M. Serrano, and J. M. Lopez-Higuera are with the Grupo de Ingeniería Fotónica, Universidad de Cantabria, Santander 39005, Spain (e-mail: ruben.ruiz@unican.es; jesus.mirapeix@unican.es; miguel.lopezhiguera@unican.es). Color versions of one or more of the figures in this letter are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/JSEN.2014.2364072 Fig. 1. Schematic view of the aluminum composite panel employed for the experiments. The blue line represents the path of the optical fiber glued on the plate, while the yellow circles represent the clamps. point of the Brillouin gain spectrum (BGS) [3], [4]. We have already applied this method to perform experimental modal analysis of a cantilever beam [5] and a rectangular plate [6]. In this letter, we apply the SA-BOTDA method to perform the modal analysis of an aluminum composite plate. Compared to Ref. [6], the analysis presented in this letter is aimed to detect and localize small changes (defects) of the mechanical structure. Experimental tests demonstrate that the technique is capable to localize plate stiffness changes with good resolution. II. EXPERIMENTAL DEMONSTRATION Modal shape measurements were carried out by using the SA-BOTDA apparatus described in Ref. [5]. The pump pulse duration was fixed to 2-ns, resulting in a ∼ 20 cm spatial resolution, while the sampling rate was set to 2 GS/s, corresponding to a sampling step of ∼ 5 cm. Note that the method requirement of small static and dynamic Brillouin frequency shift changes, compared to the BGS bandwidth, is easily verified when using 2-ns pulses [3]–[6]. The mode shapes were measured by acquiring the strain profiles at a maximum rate of 24 profiles/s. The structure chosen for the tests was a 1220 mm×1220 mm×2 mm (L×W×H) composite aluminum plate, made by a polymer layer sandwiched between two 0.3mm-thick aluminum foils. The panel was kept in vertical position and fixed to a wall by use of four clamps distant 2 cm from the plate corners in both x- and y-directions. A single-mode optical fiber was glued on the front surface of the plate, following the path shown in Fig. 1. In detail, the fiber was disposed along the x direction so as to form 24 horizontal strands spaced 5 cm along the y direction. The plate was mechanically excited by use of a voice coil characterized by a maximum continuous force of 28 N and a force constant of 8 N/A, and positioned close to the middle of the plate. The natural frequencies of the plate in the range 1 ÷ 10 Hz were identified by applying a chirped current to the voice coil, while recording the intensity of the detected Brillouin signal [6]. For each identified resonance, the corresponding mode shape was retrieved by exciting the plate at resonance 1530-437X © 2014 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.