Measurement Range Enlargement in Brillouin Optical Correlation Domain Analysis Using Multiple Correlation Peaks

Size: px

Start display at page:

Download "Measurement Range Enlargement in Brillouin Optical Correlation Domain Analysis Using Multiple Correlation Peaks"

Cory Conley
5 years ago
Views:

1 Journal of the Optical Society of Korea Vol. 16, No. 3, Septeber 2012, pp DOI: Measureent Range Enlargeent in rillouin Optical Correlation Doain Analysis Using Multiple Correlation Peaks Ji Ho Jeong 1,2, Kwanil Lee 1 *, Je-Myung Jeong 2, and Sang ae Lee 1 1 Center for Opto-Electronic Convergence Systes, Korea Institute of Science and Technology (KIST), Seoul , Korea 2 Departent of Electrical and Coputer Engineering, Hanyang University, Seoul , Korea (Received July 13, 2012 : revised August 14, 2012 : accepted August 14, 2012) We propose and experientally deonstrate a ethod for extension of easureent range in a rillouin optical correlation doain analysis sensor syste without resolution deterioration. In the experient, a 300 easureent range with about 8 c spatial resolution was successfully obtained by cascading three different kinds of fibers as a sensing eleent. Keywords : Fiber optics sensors, Stiulated rillouin scattering, rillouin optical correlation doain analysis OCIS codes : ( ) Fiber optics; ( ) Fiber optics sensors; ( ) Modulation; ( ) Scattering easureents; ( ) Scattering, stiulated rillouin I. INTRODUCTION So far, rillouin scattering based sensors have been studied as proising technology for a distributed teperature or strain easureent in structural health onitoring [1-10]. Intrinsically, rillouin scattering is an effect caused by acoustic vibrations in the ediu when laser light travels in a ediu such as an optical fiber. Since the sound wave is traveling, light is also subjected to a Doppler shift, so its frequency changes. The rillouin frequency shift ν can be calculated fro the effective refractive index of the propagating edia n, the acoustic velocity V a, and the wavelength of the incident light λ : 2nV a ν = (1) λ On the other hand, in optical fibers, the acoustic velocity depends on the aterial density ρ [11]: K V a = (2) ρ where K is the bulk odulus. Since any teperature variation or deforation experienced by the fiber will also have an ipact on its density, the rillouin frequency will be shifted with the teperature or strain as expressed in Eq. (1) and (2). Also it is known that rillouin frequency shift (FS) is linearly proportional to these two quantities [9]: ν ( ε ) = ν (0)[1 + C ε ] (3) s ν t) = ν ( t )[1 + C ( t t )] (4) ( r t r where ε is strain, t is teperature, t r is reference teperature, and C s and C t are the proportional coefficients of strain and teperature, respectively. (C s = 4.6, C t = K). Aong any rillouin scattering based sensors, the ost extensively studied techniques are rillouin optical tie doain reflectoetry (OTDR), rillouin optical tie doain analysis (OTDA), and rillouin optical correlation doain analysis (OCDA). OTDR and OTDA have an advantage of long easureent range of several tens of *Corresponding author: klee21@kist.re.kr Color versions of one or ore of the figures in this paper are available online

2 Measureent Range Enlargeent in rillouin Optical Correlation Doain Analysis - Ji Ho Jeong et al. 211 k. However, their long easureent tie (~ several inutes) and liited spatial resolution (> 50 c) ay prevent soe specific applications for health onitoring of structures such as aircraft wings, rockets, and wind turbine blades. Meanwhile, OCDA has a high spatial resolution (~ c order) and a high sapling rate (~ khz) with rando access of sensing position. In this syste, odulation paraeters (aplitude and frequency) of a light source ust be chosen such that only a single correlation peak lies within the sensing fiber. Thus, there is a trade-off proble between the spatial resolution and the easureent range [12-14]. Several techniques have been introduced to extend the easureent range while aintaining the spatial resolution such as optical tie gating [14] (the easureent range: 250, the spatial resolution: 8 c), double frequency odulation [15] (the easureent range: 34, the spatial resolution: 10 c), and using different kinds of fibers [16]. Particularly, in [16], they deonstrated two ties enlargeent of strain distribution easureent range by using two different kinds of fibers and a typical electro-optic odulator (EOM) with the easureent range of 8 and the spatial resolution of 4 c. In this schee, it is undesirable for the odulation aplitude Δf to be higher than the rillouin frequency ν, since it suffers the reduction of signal due to counterbalancing of rillouin gain and loss when Δf is larger than ν, where spectru overlap occurs between the two sidebands fro the EOM [17]. Since the spatial resolution is inversely proportional to Δf, there is soe liit of spatial resolution in that schee. Moreover, it requires a proper optical filter to distinguish the probe signal fro the backward reflected pup waves at the overlapped frequency. To solve these probles, we use a single sideband (SS) odulator instead of an EOM. In this schee, Δf can be higher than ν, capable of achieving high spatial resolution. Additionally, to circuvent a beat noise between the lock-in frequency and the odulation frequency, we used a variation-frequency lock-in detection ethod [18]. Also we adopted a polarization diversity schee to avoid the signal fluctuation due to polarization change of probe and pup light. II. PRINCIPLE In the typical OCDA syste, a sinusoidal frequency odulation is applied to the pup and probe waves, producing periodical correlation peaks along the test fiber [3]. The periodic correlation position z q is given by [19]: z q 1 1 c = ( l + ld ) q (5) 2 2 f n where f is the odulation frequency of the light source, c is the light speed in vacuu, n is the refractive index, q is (a) (b) FIG. 1. Scheatic diagra of OCDA syste; (a) Conventional ethod. (b) Proposed ethod. an integer, l d and l are the length of delay fiber and test fiber, respectively. y adjusting the length of delay fiber, we can locate only one non-zero correlation peak z q along the fiber under test (FUT) sandwiched between the circulator and the optical isolator as shown in Fig. 1(a). And the stiulated rillouin scattering is generated only at this correlation peak. Therefore, we can selectively obtain the inforation of the stiulated rillouin gain spectru (GS) at the correlation peak position. This correlation peak, i.e., easuring position, is scanned by sweeping the odulation frequency f. Therefore, spatial range of easureent is liited to the interval of the correlation peaks d : d Vg = (6) 2 f where V g is the group velocity of light. Here, the spatial resolution Δz is given by: Vg Δν Δz = 2πf Δf where Δν is the rillouin gain bandwidth and Δf is the aplitude of frequency odulation of the light source. Fro above two equations, the nuber of sensing points (7)

3 212 Journal of the Optical Society of Korea, Vol. 16, No. 3, Septeber 2012 N R can be calculated as: N R d πδf = = (8) Δz Δν Accordingly, to increase the easureent range, odulation frequency f ust be lowered or odulation aplitude Δf ust be increased. When f is lowered, the spatial resolution is deteriorated fro Eq. (7). Therefore large Δf is required to extend the easureent range while aintaining the spatial resolution. However, it is not easy to increase Δf. Another ethod for enhancing the easureent range is to utilize ultiple correlation peaks. In this technique, different kinds of fibers with different rillouin frequency are connected to copose of FUT [16]. Furtherore, the length of each fiber is set in order that only one correlation peak is located in each fiber. Thus, strain distribution in each fiber can be selectively obtained by adjusting the sweep region of the icrowave frequency ν since each rillouin gain signal is well separated due to different rillouin frequency shift aong the fibers. Thus, the easureent range can be ultiplied. For exaple, as shown in Fig. 1(b), assue that FUT is coposed of three sections of fibers with different rillouin frequencies which are connected by the correlation peak interval such that only one correlation peak is located on each fiber. y sweeping the icrowave frequency wide enough to cover three different rillouin frequencies, we can obtain overall strain distribution of the FUT. Accordingly, the easureent range can be expanded three ties while aintaining the sae spatial resolution. III. EXPERIMENTS AND DISCUSSION The experiental setup is shown in Fig. 2. A FUT was constructed by splicing 100 -DCF, 100 -DSF and 100 -SMF, which have different rillouin resonance frequencies. As a light source, 1548 n-distributed feedback laser diode (DF-LD) was used, and a sinusoidal frequency odulation was applied to generate two correlation peaks within FUT. The odulation frequency f was set to MHz depending on the easureent position (i.e., the correlation peak) in FUT and odulation aplitude Δf was about 12 GHz. Fro these paraeters the spatial resolution and easureent range was estiated to be about 8 c fro Eq. (7) and 100 fro Eq. (6), respectively. The output fro the DF-LD was divided into two beas by a bea splitter. One of the beas was used as the rillouin pup after passing through 10-k delay fiber to control the order of correlation peaks and a high power erbiu doped fiber aplifier (EDFA). The other bea was injected into the SSM which was driven by a icrowave signal generator, so that the first lower sideband, serving as the probe light, was generated and propagating against the pup in the FUT. efore entering the FUT, the pup and probe powers were about 30 d and 23 d, respectively. Additionally, a polarization switch (PSW) was inserted after the SSM for suppressing the polarization dependence of the rillouin signal [3]. Moreover, the pup light was chopped by an intensity odulator for lock-in detection and the lock-in frequency was chosen to get a sharp rillouin peak and noise suppression. During the position sweep easureent, f l was varied according to the relation f l = f /4 (f l is lock-in frequency and f is odulation frequency) for iniizing the beat noise [18]. A 125-MHz photo receiver was used as a detector and the GS was obtained through a lock-in aplifier. As a proof of concept experient, we perfored easureent of GS along the FUT, in which the strain of 2.2 ε was applied to three 8-c sections and a teperature was aintained by an environent control syste at 25 as shown in Fig 3. There are three correlation positions within the FUT: One is located in the DCF (0-100 ), another one in the DSF ( ) and the third in the SMF ( ). Since the ν of each fiber is different, we can selectively easure GS distribution in each fiber by adjusting the sweep frequency. The GS was easured every 4 c along the FUT, sweeping Δν fro 9.8 to 11.1 GHz. There are three rillouin gain peaks in the spectru and each peak coes fro a different position (z q+1, z q and z q-1 in Fig. 1(b)) along the FUT. The easured distributed GS along the FUT is shown in Fig. 4. Clearly different rillouin frequencies FIG. 2. Experiental setup for expanding the easureent range. FIG. 3. Structure of the fiber under test (FUT).

Measureent Range Enlargeent in rillouin Optical Correlation Doain Analysis - Ji Ho Jeong et al. 213 FIG. 4. 3D distribution of GS along the FUT. FIG. 6. Distributed FS along the FUT.

5 shows an expanded view of the distributed GS of the circled segents shown in Fig. 4. On the other hand, Fig. 6 shows the distribution of the FS derived fro the data of Fig. 5(a)-(c).

The change of the FS was about 100 MHz, 106 MHz, and 110 MHz for DCF, DSF, and SMF region, respectively, which agree well with the applied strain of 0.22%.

4 Measureent Range Enlargeent in rillouin Optical Correlation Doain Analysis - Ji Ho Jeong et al. 213 FIG. 4. 3D distribution of GS along the FUT. FIG. 6. Distributed FS along the FUT. (a) (b) are observed in each fiber. Note that the ν of each fiber was around 10.0 GHz, GHz, and GHz for DCF, DSF, and SMF, respectively. Fig. 5 shows an expanded view of the distributed GS of the circled segents shown in Fig. 4. On the other hand, Fig. 6 shows the distribution of the FS derived fro the data of Fig. 5(a)-(c). The spatial resolution was calculated to be about 8.0 c by Eq. (7). We can observe clear rillouin frequency shifts (FS) at the strain applied sections. The change of the FS was about 100 MHz, 106 MHz, and 110 MHz for DCF, DSF, and SMF region, respectively, which agree well with the applied strain of 0.22%. Also the FSs have different slopes of FS which are MHz / 1 µε, MHz / 1 µε and 0.05 MHz / 1 µε for DCF, DSF, and SMF, respectively by Eq. (3). The easureent error was about +/- 3.5 MHz, corresponding to the strain accuracy of +/0.007% (+/-70 µε ). Thus, we successfully deonstrated strain distributed easureent over 300- easureent range with 8.0-c spatial resolution by using the proposed schee, which has three ties larger easureent range than that of the conventional OCDA syste. IV. CONCLUSION (c) FIG. 5. Expanded view of easureent of GS distribution; (a) DCF region, (b) DSF region, (c) SMF region. In conclusion, we achieved a factor of 3 expansion of easureent range in OCDA syste by using different kinds of fibers while aintaining the spatial resolution. In this technique, ultiple correlation peaks within the sensing section can be utilized since any correlation peak point can be arbitrarily selected due to different rillouin frequency shifts. In the experient, by adopting the SSM and variation -frequency lock-in detection technique, we could get a better spatial resolution. We expect that the proposed OCDA syste could be useful in the strain onitoring of civil structures.

5 214 Journal of the Optical Society of Korea, Vol. 16, No. 3, Septeber 2012 ACKNOWLEDGMENT This work was partially supported by asic Science Research Progra through the National Research Foundation of Korea (NRF) funded by the Ministry of Education Science and Technology ( ) and the KIST R&D progra (2E23220). And authors are grateful to Prof. Kwang Yong Song at Chung-Ang University for helpful discussions and coents. REFERENCES 1. H.-N. Li, Recent applications of fiber optic sensors to health onitoring in civil engineering, Eng. Structures 26, (2004). 2. A. Rogers, Distributed optical-fibre sensing, in Handbook of Fibre Optic Sensing Technology, J. M. Lopez-Higuera, ed. (John Wiley & Sons, Chichester, England, 2002), pp K. Hotate, K. Abe, and K. Y. Song, Suppression of signal fluctuation in rillouin optical correlation doain analysis syste using polarization diversity schee, IEEE Photon. Technol. Lett. 18, (2006). 4. X. ao, Optical fiber sensors based on rillouin scattering, Optics & Photonics News 9, (2009). 5. K. Y. Song, Z. He, and K. Hotate, Distributed strain easureent with illieter-order spatial resolution based on rillouin optical correlation doain analysis, Opt. Lett. 31, (2006). 6. Y. Mizuno, Z. He, and K. Hotate, Measureent range enlargeent in rillouin optical correlation-doain reflectoetry based on teporal gating schee, Opt. Express 17, (2009). 7. K. Hotate and T. Yaauchi, Fiber-optic distributed strain sensing syste by rillouin optical correlation doain analysis with a siple and accurate tie-division pup-probe generation schee, Jpn. J. Appl. Phys. 44, L1030-L1033 (2005). 8. J. Dakin and. Culshaw, Optical Fiber Sensors Applications, Analysis and Future Trends IV (Artech House, oston, USA, 1997). 9. T. Horiguchi, K. Shiizu, T. Kurashia, M. Tateda, and Y. Koyaada, Developent of distributed sensing technique using rillouin scattering, J. Lightwave Technol. 13, (1995). 10. C.-S. Park, K.-I. Joo, S.-W. Kang, and H.-R. Ki, A PDMS-coated optical fiber ragg grating sensor for enhancing teperature sensitivity, J. Opt. Soc. Korea 15, (2011). 11. T. Horiguchi, T. Kurashia, and M. Tateda, Tensile strain dependence of rillouin frequency shift in silica optical fibers, IEEE Photon. Technol. Lett. 1, (1989). 12. K. Hotate and T. Hasegawa, Measureent of rillouin gain spectru distribution along an optical fiber using a correlation-based technique-proposal, experient and siulation, IEICE Trans. Electron. E83-C, (2000). 13. K. Y. Song, Z. He, and K. Hotate, rillouin optical correlation doain analysis syste with kiloeter easureent range based on intensity odulation schee, in Proc. OSA/CLEO 2006 (California, USA, 2006), CD, paper CThL K. Hotate and H. Arai, Enlargeent of easureent range of siplified OCDA fiber-optic distributed strain sensing syste using a teporal gating schee, Proc. SPIE 5855, 184 (2005). 15. W. Zou, Z. He, and K. Hotate, Enlargeent of easureent range by double frequency odulations in one-laser rillouin correlation-doain distributed discriination syste, in Proc. CLEO/QELS 2011 (altiore, USA, May 2011), CD, paper CThL K. Hotate and M. Tanaka, Enlargeent of easureent range of optical-fiber rillouin distributed strain sensor using correlation-based continuous-wave technique, in Proc. CLEO/QELS 2001 (altiore, USA, May 2001), CD, paper CtuD6, pp K. Y. Song and K. Hotate, Enlargeent of easureent range in a rillouin optical correlation doain analysis syste using double lock-in aplifiers and a single-sideband odulator, IEEE Photon. Technol. Lett. 18, (2006). 18. J. H. Jeong, K. Lee, K. Y. Song, J.-M. Jeong, and S.. Lee, Variable-frequency lock-in detection for the suppression of beat noise in rillouin optical correlation doain analysis, Opt. Express 19, (2011). 19. M. S. Seo, S. C. Yun, J. Y. Hyun, and H. G. Park, Experient of distributed optical fiber sensor using spatiallyselective rillouin scattering, Korean J. Opt. Photon. 17, (2006).

Efficient dynamic events discrimination technique for fiber distributed Brillouin sensors

Efficient dynamic events discrimination technique for fiber distributed Brillouin sensors Efficient dynaic events discriination technique for fiber distributed rillouin sensors Carlos A. Galindez,* Francisco J. Madruga, and Jose M. Lopez-Higuera Photonics Engineering Group, Universidad de Cantabria,Edif.