Proceedings of the ASME 2011 Conference on Smart Materials, Adaptive Structures and Intelligent Systems

Size: px

Start display at page:

Download "Proceedings of the ASME 2011 Conference on Smart Materials, Adaptive Structures and Intelligent Systems"

Basil Dorsey
5 years ago
Views:

1 Proceedings of the ASME 2011 Conference on Smart Materials, Adaptive Structures and Intelligent Systems SMASIS2011 September 18-21, 2011, Scottsdale, Arizona, USA Proceedings of the ASME 2011 Conference on Smart Materials, Adaptive Structures and Intelligent Systems SMASIS2011 September 18 21, 2011, Scottsdale, Arizona, USA SMASIS2011- SMASIS SIMULANEOUS MEASUREMEN OF DEFORMAION AND FRACURE OF COMPOSIE SRUCURES USING FIBER BRAGG GRAING SENSORS Hong-Il Kim Jae-Hung Han* Department of Aerospace Engineering, KAIS, Daejeon, Republic of Korea * jaehunghan@kaist.ac.kr Hyung-Joon Bang Soo-Hyun Kim Korea Institute of Energy Research, Daejeon, Republic of Korea Bongwan Lee Fiberpro Inc., Daejeon, Republic of Korea ABSRAC Because of high specific strength and many other benefits, the use of composites for the large lightweight structures such as modern aircrafts and wind turbines are increasing. However, one of the serious drawbacks of composites is that the structural failure occurs in complex patterns without yielding. herefore, structural health monitoring has been intensively investigated for the early detection of any problems in structural integrity. One of the promising sensors for this purpose is fiber Bragg grating (FBG) sensor. hey can be easily inserted into the layered-structure of the composite materials due to their small size. he excellent multiplexing capability enables measurement to be taken at multiple points along a single sensor line. As well as damage detection, the structural shape measurement also draws attention. Particularly for structures experiencing aerodynamic forces such as wind turbines or helicopter blades, the structural shape itself is important because the applied aerodynamic forces are affected by structural shape deflections. herefore, the authors have conducted a series of studies on the structural shape estimation of various structures. We have also developed a wavelength division multiplexing (WDM) Bragg grating sensing system for high speed strain sensing as well as low frequency dynamic strains. In the case of high-speed sensing, the interrogator allows a sampling ratio of over 40 khz for six linearly arrayed FBG sensors per channel. Utilizing the developed interrogator, this paper presents some experimental results for simultaneous measurement of deformation and fracture signals of composite structures. An array of FBG sensors were installed onto composite beam specimens and the acoustic emission (AE) signals due to structural failure was continuously monitored while the overall structural deflection shape was monitored in real time. he reconstructed shapes of the specimens were in good agreement with the shapes captured from photographs taken with a high-speed camera. In summary, it was demonstrated that both fracture signals and the overall deformation shape of composite structures could be simultaneously monitored. INRODUCION As the structures are upsizing with light-weight materials such as composite materials, the smart structure technologies including monitoring of loads, impacts, fractures and delamination are increasingly required. In particular, it might be very helpful to acquire the deformations of the structures such as aircraft wings, helicopter rotors and reflectors of the satellites where the structural shape changes directly influence the performances of the whole system. hus, there have been several studies for structure status monitoring [1-4] and shape estimations [5-12] on the basis of various sensor systems and algorithms. Meanwhile, not only the application of the newly developed technologies, but also the economic efficiency of the structure monitoring throughout the operation is important; this results in the integration and simplification of the monitoring systems using a variety of sensors. he fiber Bragg grating (FBG) sensor could be an ideal solution for both applications because of its inherent small size, light-weight, electromagnetic immunity, excellent strain accuracy as well as multiplexing capability which enables measurement to be taken at multiple points along a single sensor line based on the wavelength-division multiplexing (WDM) technologies [13-15]. From the previous studies, we have experienced in the fields of the structure status monitoring and the shape estimation of the various shaped structures using FBG sensors. Based on the previous experiences, in this study, we developed an integrated sensor system for the simultaneous measurements of deformation and fracture of the composite structures using FBG sensors. he concept of the simultaneous measurements is depicted in Fig Copyright 2011 by ASME

2 shape was compared with the measured shape by the optical 3D point tracking system. FBG SENSOR SYSEMS Working principles of FBG sensor he working principle of the FBGs is based on a periodic change of the refractive index in the optical fiber as depicted in Fig. 2. In case of FBGs, the gratings of period are engraved at the core by using ultraviolet laser and the refractive index of grating region is changed to the effective refractive index n due to the continuous gratings. hus, when light wave travels along the fiber core, the constructive interferences at the grating region become the reflection of a specific wavelength, called Bragg wavelength B. he Bragg wavelength is expressed using the effective refractive index n and grating period (Eq. 1). 2 n (1) B Fig. 1 Schematic diagram of simultaneous measurements of deformation and fracture of the composite structure using multiplexed FBG sensors he strain data from the FBG sensors distributed on the large composite structure are gathered through the FBG interrogator. From the adequate signal processing, strain data are divided into two data sets of different frequency ranges. he low frequency strain responses (< 100 Hz) are applied to estimate the full-field deformation of the target structure while high frequency strain responses (~ 40 khz) are used to detect the impact locations and fracture/delamination. In this concept, it is most important to have a high-speed FBG interrogator which allows the sampling ratio over 40 khz while maintaining multiplexing capability for acquiring the high frequency strain responses of the multi-points on the structure. hus, we developed a high-speed interrogation system which was fitted with a spectrometer-type demodulator based on a linear photo detector, implementing a 40 khz or higher sampling ratio per channel. Utilizing the developed interrogator, in this paper, we conducted a demonstration test for simultaneous measurement of deformation and fracture signals of composite structures. An array of FBG sensors were mounted onto composite specimens and the acoustic emission (AE) signals resulting from the impacts were monitored. At the same time, the full-field shape of the composite specimen was estimated and the resulting Fig. 2 Periodic change of refractive index in Bragg grating he reflected wavelengths of FBGs are changed by strain and temperature in fiber core. hus, the structure strain can be measured on the basis of shift of the wavelength B as Eq. (2) [16]: 1 B f 1 p e (2) B where p e is the photo-elastic constant of an optical fiber and f are the thermo-optic coefficient. High-speed FBG interrogator A wavelength division multiplexing (WDM) Bragg grating sensing system was developed with the help of FiberPro, Inc. Fig. 3 shows the schematic diagram and the appearance of the high-speed FBG interrogator (KHFI-140) developed. he FBG interrogator was fitted with a spectrometer-type demodulator based on a linear photo detector, implementing a 40 khz or 2 Copyright 2011 by ASME

higher sampling ratio per channel. For highly efficient signal processing, the demodulation of the FBG wavelength signal was carried out using an embedded FPGA.

3 higher sampling ratio per channel. For highly efficient signal processing, the demodulation of the FBG wavelength signal was carried out using an embedded FPGA. he system was designed with a main control/communication board which saved the demodulated data and exchanged them with the DSP of demodulator and external devices. In the high-speed demodulation mode, up to six arrayed FBGs can conduct 40 khz sampling per each channel simultaneously, and the system is suitable for structural damage detection by high-frequency vibration sensing such as AE (acoustic emission) detection. In the low-speed demodulation mode, the system can be effectively used for the dynamic strain monitoring of large structures which require many sensors, by expansion of the sensing channels. (a) BBS Control Board of the longitudinal waves. For sensor i, the wave arrival time t i is expressed in Eq. (3): r i ti (3) vi where v i is the propagation speed of the longitudinal wave along the impact location to sensor and r i is the distance between impact location to sensor. If we assume that the propagation speed, v i, is constant along the longitudinal direction of the composite specimen, the impact location can be expressed as Eq. (4) [1,4] i i i o i o j o j o tv x x y y x x y y ( i j, i, j 1, 2,..., M) (4) AMP Broad Band Source Linear PD Spectrometer module Channel Switching Controller FBG1-1 FBG1-2 FBG2-1 FBG2-2 FBG6-1 FBG6-2 where (x i, y i ) and (x j, y j ) are the positions of the i-th and j-th FBG sensors and (x o, y o ) means the location of the impact. A/D Converter FPGA Digital Signal Processor Control Data Main Control / Communication Board (b) Fig. 4 Impact strain signal measured by KHFI-140 Fig. 3 (a) Schematic diagram of high-speed FBG interrogator, (b) High-speed FBG interrogator (KHFI-140) PRINCIPLES OF MEASUREMENS Impact detection Generally, the strain wave signal produced by the impact can be divided into two waves by their propagation speeds, longitudinal waves and bulk waves including transverse waves and reflected waves as shown in Fig. 4. Longitudinal waves travel relatively faster than transverse waves and have regular reproducibility. But the bulk waves have lower frequency components with irregularity which are dominantly below 20 khz. hus, the faster sampling frequency was demanded to pick up the impact waves whose frequency are about a few tens of khz in order to detect the fracture or impact location. Also, the strains should be measured simultaneously to obtain the time differences between sensors. Based on this, the impact location can be derived from the differences of the arrival times Strain based shape estimation [9-12] Strain based displacement estimation method can be divided into two categories depending on how displacementstrain transformation matrix is obtained. In general, displacement field can be calculated for a given load vector from finite element or analytical models. Using the obtained displacement field, strains of specified positions can be obtained according to the definition of strain. From these strain and displacement results at a given loading condition, it is possible to formulate transformation matrix between displacement and strain. On the other hand, the second category uses structural bases such as mode shapes to construct displacement strain transformation relation. Mode shapes and corresponding strain mode shapes are easily obtained even for a complex structure from the analytic or FEM results without consideration of detail loading conditions such as acting point, load direction and magnitude of load. In this study, this modal approach is used for the construction of the displacement-strain transformation relation because modal approach does not need loading information. In addition, the modal approach is easily applied to dynamic 3 Copyright 2011 by ASME

1 2 n N n 1 2 n N n y N 1 N n n 1 (5) (6) (7) M 1 n 1 (8) q M n shown in Fig. 5, total 24 markers were attached on the surface of the composite plate.

4 excitation cases. In general, the structural displacements y and strains ε can be expressed by the linear combination of mode shapes φ i and strain mode shapes ψ i, respectively, at a certain time as Eq. (7) and (8). 1 2 n N n 1 2 n N n y N 1 N n n 1 (5) (6) (7) M 1 n 1 (8) q M n shown in Fig. 5, total 24 markers were attached on the surface of the composite plate. For this experiment, the static loadings from 50gf to 350gf was applied at the tip of the plate by hanging the mass while the strain responses of the FBG sensors and three dimensional positions of the markers using SPR system were simultaneously captured at the sampling rate of 100Hz. Actually, displacements y and strains ε are reduced by N- dof displacement vector {y N 1 and M-dof strain vector {ε} M 1 expressed by multiplying modal coordinate {q, #141} n 1 and n mode shape matrix [Φ] N n and n strain mode shape matrix [Ψ] M n respectively. If we suppose the modal coordinate in Eq. (7) is equal to that in Eq. (8), the modal coordinate can be expressed as in Eq. (9) by substituting Eqs. (5) and (6) for mode shapes in Eqs. (7) and (8). 1 M n M n M n q (9) n 1 M 1 As a result, it is possible to construct a displacement-strain transformation (DS) matrix as in Eq. (10). y 1 (10) e N 1 N n M n M n M n M 1 DS where {ye, #118} N 1 is the estimated displacements. EXPERIMENS est description In order to demonstrate the simultaneous measurements of the structure shape and the impact location using FBG sensors, the graphite/epoxy composite plate layered as [0/45/-45/90] 2s was prepared. his specimen had the dimension of 500 mm 300 mm 2 mm and one end was clamped to the jig as shown in Fig. 5. Four FBG sensors whose Bragg wavelengths were 1535nm, 1540nm, 1545nm and 1553nm were attached on the specimen; the locations of the each sensor were 100mm, 200mm, 300mm and 400mm from the fixed root, respectively. For the impact detection tests, 0.1 J of impact energy was inflicted while the impact location was varied as shown in Fig. 6. he impact strain responses from the FBG sensors were captured at the sampling rate of 40 khz; the response of the FBG1 was used as the trigger signal and the time differences of the longitudinal waves were calculated. In addition, in order to measure the shape of the structure, the stereo pattern recognition (SPR) system composed of two Eagle cameras (Motion analysis corp.) which tracked the three dimensional positions of the reflecting markers was used. As Fig. 5 est setup; the composite specimen is under the static loading. he shape of the composite specimen is obtained from reflecting markers image by SPR camera Fig. 6 Experimental setup for impact location monitoring 4 Copyright 2011 by ASME

5 Results Impact detection In this experiment, the impact signals were successfully obtained by using the developed interrogator (KHFI-140); the propagation speed, v, was measured to be m/s for the [0/45/-45/90] 2s graphite/epoxy composite specimen; the time differences of the peaks of longitudinal waves were about 0.2 ms as shown in Fig. 7. Results Shape estimation From the finite element model of the plate, the mode shape matrix and corresponding strain mode shape matrix were obtained; four bending modes shown in Fig. 9 were selected as the relevant modes and location of the FBG sensors were taken into account. hen, the DS matrix in Eq. (10) was obtained. Fig. 7 Strain wave signals detected by the FBG sensors Based on the measured wave propagation speed, we conducted a series of tests for detecting the impact locations. Fig. 8 shows the strain signals detected by the FBG sensors for the two impact locations (150 mm, 350 mm from the clamped end). From the results, time differences were derived from the locations of the peaks of longitudinal waves. (a) Fig. 9 Selection of the relevant modes and mode shapes For each static loading case, the wavelength changes of each FBG sensor were measured while SPR system captured the markers; Fig. 10 shows the measured wavelength values of the FBG 4 (1552 nm) while static loading was applied. hese wavelength changes were transformed to the strain data; the plate shapes were calculated by multiplying the DS matrix and strains. Fig. 11 shows the estimated shapes of the composite plate using strain data and DS matrix. he tip maximum deflections were estimated as -9.7mm, -18.8mm and -27.7mm for the static loadings of 100gf, 200gf and 300gf, respectively. he estimated shapes were compared with measured shapes by SPR system. Fig. 12 shows both estimated shapes and measured shapes for 100gf, 250gf and 350gf cases; the estimated shapes were shown in contoured shapes and measured shapes were presented in edge lines. From the results, it was clear that the composite shapes were successfully estimated using strain data from the four FBG sensors. (b) Fig. 8 Strain wave signals detected by the four FBG sensors for the two impact locations; (a) impact at 150mm (mid-point of FBG1 and FBG2) (b) impact at 350 mm (mid-point of FBG3 and FBG4) Fig. 10 Wavelength results of the FBG 4 (1552nm) under the static loading from 0gf to 350gf 5 Copyright 2011 by ASME

Fig. 11 Estimated shapes of the composite plate using strain data and DS matrix; the maximum deflections were shown for each loading case CONCLUSION For the large composite structures, such as wind

$he low/high speed impact, fracture and delaminations should be reported to the operator to avoid system failure.$

6 Fig. 11 Estimated shapes of the composite plate using strain data and DS matrix; the maximum deflections were shown for each loading case CONCLUSION For the large composite structures, such as wind turbine blades and towers, continuous status monitoring is required during the operation. he low/high speed impact, fracture and delaminations should be reported to the operator to avoid system failure. In addition, the information on the structural deformation would be very helpful for the further improvement in the control of the structures which are affected by aerodynamic forces. he FBG sensors would be adopted for both structural status monitoring and shape estimation on the basis of the many advantages; they can be easily inserted into the layered-structure of the composite materials due to their small size. he excellent multiplexing capability enables measurement to be taken at multiple points along a single sensor line. In this study, therefore, we developed the wavelength division multiplexing (WDM) Bragg grating interrogation system which allowed a sampling ratio of over 40 khz for six linearly arrayed FBG sensors per channel in order to apply FBG sensors to the both structural status monitoring and shape estimation. Utilizing the developed interrogator, simultaneous measurement of deformation and impact signals on the composite structures was demonstrated. An array of FBG sensors were installed onto the composite specimens and the acoustic emission (AE) signals due to impacts was monitored while the structural shape was monitored. he estimated shapes of the composite plate were in good agreement with the shapes measured from SPR system. In summary, it was demonstrated that both impact signals and the overall deformation shape of composite structures could be simultaneously monitored. ACKNOWLEDGMENS his work was supported by the New & Renewable Energy of the Korea Institute of Energy echnology Evaluation and Planning (KEEP) grant funded by the Korea government Ministry of Knowledge Economy (Grant No. 2008NWD08J ). Fig. 12 Shape comparisons; contoured shapes are estimated shapes by DS matrix, Line edge shapes are measured shapes using SPR system. For each loading case, the estimated shapes using strain data are quite similar to the measured shapes using SPR system. REFERENCES [1] Greene, J.A., ran,.a., Bhatia V., Gunther, M.F., Wang, A., Murphy, K.A., and Claus, R.O., 1995, Optical-Fiber Sensing echnique for Impact Detection and Location in Composites and Metal Specimens, Smart Mat. Struct., 4(2), pp.93-9 [2] Sung, D.U., Oh, C.G., and Hong C.S., 2000, Impact Monitoring of Smart composite Laminates using Neural Networks and Wavelet Analysis, J. Int. Material Systems and Struct., 11(3), pp [3] Kang, H.K., Bang, H.J., Hong C.S., Kim, C.G, 2002, Simultaneous Measurement of Strain, emperature, and Vibration Using Fiber Optic Sensor, Meas. Sci. echnol., 13(8), pp Copyright 2011 by ASME

7 [4] Bang, H.J., Park, S.W., Kim, D.H., Hong C.S., Kim, C.G, 2004, Impact Monitoring in Smart Composites Using Stabilization Controlled FBG Sensor System, Proc. of SPIE, 5384, pp [5] Kirby, G.C., Lindner, D.K., Davis, M.A., Kersey, A.D, 1995, Optimal sensor layout for shape estimation from strain sensors, Proc. of SPIE, 2444, pp [6] Jones, R.., Bellemore, D.G., Berko,.A., Sirkis, J.S., Davis, M.A., Putnam, M.A., Friebele, E.J., and Kersey, A. D., 1998 Determination of cantilever plate shapes using wavelength division multiplexed fiber Bragg grating sensors and least-squares strain-fitting algorithm, Smart Mater. Struct., 7(2), pp [7] Davis, M.A., Kersey, A.D., Sirkis, J.S., and Friebele, E.J., 1996 Shape and vibration mode sensing using a fiber optic Bragg grating array, Smart Mater. Struct., 5(6), pp [8] Dyllong, E., and Kreuder, A., 1999 Optimal reconstruction of mode shapes using non-unitorm strain sensor spacing, Proc IEEE/ASME Int. Conf. Advanced Intelligent Mechatronics, pp [9] Kang, L,-H,, Kim, D.-K., and Han, J.-H., 2007 Estimation of dynamic structural displacements using fiber Bragg grating strain sensors, J. Sound Vib., 305(3), pp [10] Rapp, S., Kang, L.-H., Han, J.-H., Mueller, U.C., and Baier, H Displacement field estimation for a twodimensional structure using fiber Bragg grating sensors, Smart Mater. Struct., 18(2),(025006) [11] reiber, J., Mueller, U.C., Han, J.-H., Baier, H., 2008 Filtering techniques in the dynamic deformation estimation using multiple strains measured by FBGs, Proc. of SPIE, 6932(69322A) [12] Kim, H.-I., Kang, L.-H., and Han, J.-H., 2011, Shape estimation with distributed fiber Bragg grating sensors for rotating structures, Smart Mater. Struct., 20(3),( ) [13] Hill, K.O., Fujii, Y., Johnson, D.C., and Kawasaki, B.S., 1978, Photosensitivity in optical fiber waveguides: Application to reflection filter fabrication, Appl. Phys. Lett., 32(10), pp [14] Kersey, A.D., Davis, M.A., Patrick, H.J., LeBlanc, M., Koo, K.P., Askins, C.G., Putnam, M.A., and Friebele, E.J., 1997, Fiber Grating Sensors, J. Lightwave echnol., 15(8), pp [15] Davis, M.A., Bellemore, D.G., and Kersey A.D., 1994, Structural Strain Mapping using a Wavelength/ime Division Addressed Fiber Bragg Grating Array, Proc. of SPIE, 2361, pp [16] rutzel, M.N., Wauer, K., Betz, D., Staudigel, L., Krumpholz, O., Muehlmann, H.-C., Muellert,., and Gleine, W., 2000, Smart Sensing of Aviation Structures with Fiber-optic Bragg Grating Sensors, Proc. of SPIE, 3986, pp Copyright 2011 by ASME

Impact Monitoring in Smart Composites Using Stabilization Controlled FBG Sensor System

Impact Monitoring in Smart Composites Using Stabilization Controlled FBG Sensor System H. J. Bang* a, S. W. Park a, D. H. Kim a, C. S. Hong a, C. G. Kim a a Div. of Aerospace Engineering, Korea Advanced