Optical fiber temperature measurement device and optical fiber temperature measurement method

Abstract

An optical fiber temperature measurement device (10) comprises: a COTDR measurement unit (1) that measures Rayleigh backscattered light from an optical fiber to be measured, the COTDR measurement unit (1) causing light emitted from a first light source to be incident on the optical fiber to be measured, the first light source performing temperature control by changing the frequency by hopping and being capable of changing the line width of the emitted light; a BOTDR measurement unit (2) that measures Brillouin backscattered light from the optical fiber to be measured, the BOTDR measurement unit (2) causing light emitted from a second light source to be incident on the optical fiber to be measured, the second light source being capable of changing the line width of the emitted light and sweeping the emitted light by changing the frequency by triangular chirp pulses; a calibration optical fiber (8) that calibrates temperature and the like; and an LPR analysis unit (5) that calculates the intensity of the Rayleigh backscattered light and the intensity of the Brillouin backscattered light on the basis of the outputs of the COTDR measurement unit (1) and the BOTDR measurement unit (2), and measures the temperature of the optical fiber to be measured at a prescribed position using a Landau-Placzek ratio obtained on the basis of the outputs of the COTDR measurement unit (1) and the BOTDR measurement unit (2).

Description

Optical Fiber Temperature Measurement Device and Optical Fiber Temperature Measurement Method

[0001] This application relates to an optical fiber temperature measurement device and an optical fiber temperature measurement method.

[0002] Distributed fiber optic sensor (DFOS (Distributed Fiber Optic Sensing)) technology has made great technological progress in recent years. In various fields such as distributed temperature sensing (DTS) in temperature measurement, distributed strain sensing (DSS) in strain measurement, and distributed acoustic sensing (DAS) in acoustic wave measurement, various advancements have been achieved.

[0003] In any of the above fields, the sensing signal is affected by both temperature change and strain change characteristics. However, various measures are required to separate these two and obtain individual changes (see, for example, Non-Patent Documents 1 and 2).

[0004] In DTS (Distributed Temperature Sensing), the measurement method based on Raman scattering is the mainstream. In this measurement method, since the intensity of the backward scattered light is weaker than other scattered lights (Rayleigh scattered light and Brillouin scattered light), a sufficient signal must be incident during measurement. For an MMF (abbreviation for multi-mode fiber, about 50 μm in diameter) with a large core diameter, an accuracy of ±0.1 °C can be obtained, but for an SMF (abbreviation for single-mode fiber, about 8 μm in diameter), only an accuracy of ±3 °C can be expected. Therefore, commercially, only those applying MMF have been developed.

[0005] On the other hand, MMF has the following problems. That is, 1) the possible measurement range (measurement distance) is about up to 10 km, 2) the communication performance is inferior to that of SMF. It is easy to attenuate (large transmission loss) and has a large non-linearity. Also, the difference in fiber characteristics between individual optical fibers is large, and calibration of the output performance is required in advance, 3) due to the large non-linearity, usually, calibration by a separate sensor is required at both ends of the optical fiber, 4) since the mainstream communication network fiber is calibrated with SMF, it is difficult to deploy over long distances and in existing communication networks, etc. (see, for example, Non-Patent Document 3).

[0006] As one method for solving the above problems, a temperature measurement method using the Landau Placzek Ratio (LPR) technique (hereinafter simply referred to as LPR) has been proposed. In this proposal, it has been experimentally shown that the intensity ratio of Rayleigh backscattered light and Brillouin backscattered light in an optical fiber (Landau Placzek Ratio) has a temperature dependence, and the technique using LPR may be used as the basis for a distributed temperature sensor. According to the results, when combined with the known frequency dependence of Brillouin backscattering on temperature and strain, it has been shown that Brillouin backscattering can be used to uniquely determine either temperature or strain in a distributed optical fiber sensing system (see, for example, Non-Patent Document 4).

[0007] T. R. Parker, et. al. , “Temperature and strain dependency of the power level and frequency of spontaneous Brillouin scattering in optical fibers”, OPTICS LETTERS, Vol. 22, No. 11, June 1, 1997, pp. 787-789. Y. Sakairi, et. al. , “A system for measuring temperature and strand separation by BOTDR and OTDR”, Advanced Sensor Systems and Applications, Proceeding of SPIE, Vol. 4920, 2002, pp. 274-284. Fumio Wada, “Temperature measurement of optical fibers using Raman scattering”, Applied Physics, Vol. 60, No. 1, 1991, pp. 68-69. P. C. Wait, et al. , "Landau Placzek ratio applied to distributed fiber sensing", Optics Communications 122, 1996, pp. 141-146. Yosuke Mizuno et al., "Brillouin scattering in optical fibers and its sensor application", Ultrasonic TECHNO, 2014.5-6. pp. 84-89. Hideki Nagatani, "On the development of monitoring technology using optical fiber sensors", Technical presentation materials of the Chugoku Regional Construction Technology Development Exchange Meeting, November 15, 2023.H. Izumita, et. al. , “Stochastic Amplitude Fluctuation in Coherent OTDRand a New Technique for Its Reduction by Stimulating Synchronous Optical Frequency Hopping”, Journal of Lightwave Technology, Vol. 15, No. 2, February 1997. H. Takahashi, et. al. , “Brillouin-besed fiber loss measurement for PON using ERA-BOTDA with differential width pairs of frequency-swept pump pulses", Optics Express Vol. 32, No. 15, 15 Jul 2024, pp. 26232-26244. .

[0008] International Publication No. 2023 / 058160

[0009] Therefore, we will first provide an overview of the temperature measurement method using the LPR method described above. The basic principle of this temperature measurement method is related to the absorption and radiation of photons associated with the free energy of particles. For the aforementioned particles, Thomson scattering for electrons and Raman scattering for molecules have been put into practical use. Brillouin scattering, which is related to phonons, also has a temperature effect, and in terms of signal sensitivity, it is about one-third that of Raman scattering. For temperature measurement, the intensity of Brillouin scattering is sufficiently large compared to Raman scattering, so there is a lot of experience with it even in single-mode fibers (SMF). However, in actual optical fibers, the measurement accuracy is greatly affected by the eccentricity of the core and the variation in diameter, so it is necessary to correct the measurement results of Brillouin scattering by Rayleigh scattering at each measurement point.

[0010] More specifically, one of the problems of the intensity-based sensor as described above is that it is difficult to determine whether the change in intensity is due to a change in the measurement object or due to optical fiber loss caused by splice loss of the optical fiber or micro bending. Here, since the Rayleigh backscattering signal is sensitive to temperature change, it can be used as a means to measure the optical fiber loss caused by the latter factor. Due to the relatively small frequency difference between the Brillouin frequency shift signal and the Rayleigh backscattered light signal, the optical fiber loss can be regarded as the same between the two. Therefore, the LPR defined by the ratio of the intensities of the above Rayleigh backscattering and Brillouin backscattering becomes a temperature measurement means by DFS (Distributed Fiber Sensing). However, since the frequency shift, which is the difference in frequency between the measured incident light and the Brillouin backscattered light, is a function of both temperature and strain parameters, it is necessary to separate the influence of strain in order to obtain the temperature.

[0011] Next, the measurement principle of the temperature measurement method using the LPR method will be described (see Non-Patent Document 4, etc.). LPR was first derived by Landau and Placzek as the ratio of the intensity I R of the Rayleigh backscattered light signal and 2I B which is twice the intensity I B of the Brillouin backscattered light signal. They showed that in a liquid, this ratio (LPR) is expressed by Equation (1) using the specific heat at constant pressure C p and the specific heat at constant volume C v . Therefore, when LPR is denoted as R LP ,

[0012] Subsequent to them, Schroeder et al. showed that in the case of a single-component glass, LPR is expressed as in the following Equation (2). Here, T T is the virtual temperature, T is the temperature, ρ0 is the density, v is the velocity of the elastic wave, and β T is the isothermal compressibility of the melt at the virtual temperature T T . The virtual temperature T T is the temperature at which the thermodynamic density fluctuation of the melt is frozen in the glass. Note that PR = I R , P B = 2I B We set it as follows. Equation (2) shows that LPR has a reciprocal temperature dependence (inversely proportional to temperature).

[0013] In the above, the point to note is R LP The fact is that it is reciprocally dependent with respect to temperature T. That is, in equation (2), the change in density ρ0 with respect to temperature is very small and can be ignored. Also, the velocity v of the elastic wave is expressed by the following equation (3). In equation (3) above, E is Young's modulus and μ is Poisson's ratio. Both E and μ are temperature-dependent, and therefore, although the velocity of elastic waves is small, they have a positive temperature coefficient. Thus, in equation (2), the term with respect to the reciprocal of temperature T (1 / T) is the dominant term. This has also been demonstrated experimentally (see, for example, Non-Patent Document 4).

[0014] Therefore, if we apply the above results to an optical fiber, the temperature (absolute temperature) T(z) at a predetermined point z of the optical fiber is determined by the following equation (4). Here is the first K T I in optical fiber B Temperature coefficient (= 0.0036 / °C), LPR T(z) is ratio P R (z) / P B (z), LPR Tref The ratio P in the calibrated optical fiber is R (ref) / P B (ref) and T ref is the temperature of the calibrated optical fiber. Note that P in the above-mentioned calibrated optical fiber R (ref) and P B (ref), and the temperature T of the calibrated optical fiber. ref These points will be explained in more detail later using Figure 4.

[0015] Next, we will explain recent advances in measurement techniques using Brillouin scattering and Rayleigh scattering, which are fundamental technologies for determining the LPR mentioned above, and clarify the challenges in these technologies. In distributed optical fiber measurement technology, by integrating optical fiber sensors with structures, it is possible to accurately measure the aging deterioration and earthquake damage of target infrastructure over long periods of time at distances of 10 km or more, making it useful as an infrastructure monitoring system.

[0016] First, we will provide an overview of recent advances in distributed optical fiber sensor technology based on Brillouin scattering (see, for example, Non-Patent Document 5). Brillouin scattering is a technique that utilizes the frequency information of scattered light due to acoustic phonons (density fluctuations due to thermal vibrations in a material), and is superior in measurement stability and accuracy compared to Rayleigh scattering or Raman scattering, which utilize the intensity information of scattered light. In particular, among the characteristics of this Brillouin scattering, the Brillouin frequency shift with respect to the frequency of the incident light (hereinafter abbreviated as BFS) changes almost linearly with respect to strain or temperature changes applied to the optical fiber, so the strain or temperature of the optical fiber can be detected from this BFS. Three types of measurement methods are known that utilize this Brillouin scattering: the time-domain method, the frequency-domain method, and the correlation-domain method.

[0017] Of these methods, the time-domain method utilizes time information from the time a light pulse is incident until the scattered light returns, converting it into distance information. In particular, in Brillouin optical time-domain measurement (BOTDR), the spatial resolution is proportional to the width of the incident light pulse, so shortening the pulse width leads to obtaining high spatial resolution. However, if it is too short, the spectral linewidth (the FWHM bandwidth, which will be explained in detail later; the same applies below) widens, so there is a limit (a limit due to the relaxation time of acoustic phonons, about 10 ns). Real-time and high-precision measurement of temperature distribution or strain distribution can be achieved, and recent published examples have achieved measurement times in seconds, strain measurement accuracy of ±20 με, temperature accuracy of about ±1 °C, and spatial resolution of 1 m. Note that spatial resolution also varies depending on the measurement method, and better values ​​than the above figures can be achieved with the frequency-domain method.

[0018] Next, we will provide an overview of recent advances in distributed optical fiber sensor technology based on Rayleigh scattering (see, for example, Non-Patent Document 6). The method utilizing Brillouin scattering light relies in principle on the vibration of molecules constituting the optical fiber, and therefore its light intensity is extremely low. Consequently, the strain measurement accuracy is only about ± several tens of με, and measurement is time-consuming because it requires repeated measurements and averaging. In contrast, the method utilizing Rayleigh scattering generates scattered light due to density inhomogeneities in the optical fiber, and therefore its light intensity is very strong. Consequently, in recent years, strain measurement accuracy has been achievable at 1 με or less. Furthermore, measurements can now be performed in a short time.

[0019] Next, we will discuss the problems with existing OTDR methods, including the BOTDR method described above. First, since the frequency shift of Brillouin scattering is proportional not only to the strain applied to the optical fiber but also to the temperature, there is a problem in that strain and temperature cannot be separated by measuring only the Brillouin frequency shift. Therefore, until now, methods have been employed to separate the temperature and strain components by combining the measurement of the Brillouin frequency shift with the intensity of the Brillouin scattered light. For example, a measurement system combining the use of BOTDR and OTDR has been proposed (a method of solving simultaneous equations relating the Brillouin frequency shift and the Brillouin scattered light power), and in tests with a spatial resolution of 1 m, a strain measurement accuracy of ±50 με and a temperature measurement accuracy of ±5 °C were achieved (see, for example, Non-Patent Document 2). In this measurement system, several challenges arise during measurement, including the need to measure the intensity of Brillouin scattered light with an accuracy of ±0.05 dB using microwave heterodyne detection, and the need to accurately correct for optical fiber loss in Brillouin backscattered light using the Brillouin spectral width.

[0020] In addition, in the construction and maintenance of submarine optical systems, coherent optical time-domain reflectometry (COTDR), which applies Rayleigh backscattered light, is used. COTDR is a technique that evaluates optical fiber loss or failure by injecting pulsed light from one end of the test object and receiving the backscattered light (Rayleigh backscattered light and Fresnel reflected light) generated in the optical fiber. However, when evaluating the entire system, it is necessary to perform a great deal of averaging to improve the signal-to-noise ratio.

[0021] Furthermore, Rayleigh scattering is inherently highly coherent, and suppressing this requires the use of a special laser beam with a linewidth of 100 nm. When converted to frequency, a 100 nm linewidth reaches a bandwidth on the order of T (tera) Hz, exceeding the coherent reception range, thus limiting the use of direct measurement reception techniques. For this reason, current reception sensitivity is low, and practical applications are limited to short distances. Moreover, only low spatial resolution can be achieved through averaging by many sources. Consequently, the precision required for LPR, as explained earlier, cannot be realized. Specifically, commercially available systems have an optical pulse width of 100 μs and a spatial resolution on the order of 10 km.

[0022] Furthermore, there is a problem with the amplitude of the signal obtained by COTDR (see Figure 1). Figure 1 shows an example of a COTDR signal at a measurement distance of 30 km. As shown in this figure, the COTDR signal has a problem in that it has a large amplitude called CRN (Coherent Rayleigh Noise) which is caused by light interference due to multi-point reflection. This problem usually occurs when a narrow linewidth LD is used. It can also be caused by manufacturing variations (shape errors) during the production of optical fibers, i.e., errors in diameter and eccentricity.

[0023] One method for reducing the CNR mentioned above utilizes the fact that the optical frequency of a DFB-LD (Distributed Feedback-LD) hops to different frequencies depending on the LD temperature (see, for example, Non-Patent Document 7). Non-Patent Document 7 describes a method for reducing amplitude fluctuations by changing the LD temperature during the integration of the Rayleigh scattering signal.

[0024] As a conclusion of Non-Patent Document 7 mentioned above, it was theoretically shown that amplitude fluctuations in COTDR can be explained by the probability density function of each factor causing the fluctuations. Experiments were also conducted using a 10 km long optical fiber. In the experiment, an RF (Radio Frequency) current pulse was induced in the drive current of the LD during a temperature change of the LD (a 13°C change from 27°C to 14°C), thereby causing optical frequency hopping synchronized with the emission timing of the signal pulse. The CNR reduction effect (also simply called the CNR effect) in this case will be explained using Figure 2. Figure 2 is a reference to Figure 12 of Non-Patent Document 7. Under the condition of an RF pulse width of 100 ns, the value was 0.24 dB in Case 1 (when the LD temperature is stable and an RF current pulse is induced) and 0.04 dB (best value) in Case 2 (when the LD temperature is changed from 27°C to 14°C and an RF current pulse is induced) (see Figure 2 and Non-Patent Document 7).

[0025] Next, a challenge with using Brillouin scattering signals is the issue of Brillouin intensity. That is, while Brillouin time-domain reflectance measurement (BOTDR) is an effective means of structural monitoring because it can measure the longitudinal strain distribution of an optical fiber with high accuracy and stability, the frequency shift of the Brillouin scattered light changes in proportion to the temperature of the optical fiber and the strain applied. Therefore, the measured Brillouin frequency shift contains information on both strain and temperature, which presents a problem (see Non-Patent Literature 2; see Figure 10 in that document).

[0026] Specifically, it is known that when different strain or temperature distributions exist within a 1m interval, the width of the Brillouin spectrum widens, resulting in two peaks. In other words, the Brillouin intensity P BThere is a problem with this (Figure 3). Figure 3 is a reference to Figure 10 of Non-Patent Document 2. In Figure 3, curve (1) represents a state in which no strain is applied to the 1m section, curve (2) represents a state in which strain is applied to a part of the 1m section, curve (3) represents a state in which strain is applied to a part of the 1m section, and curve (4) represents a state in which uniform strain is applied to almost the entire 1m section.

[0027] Comparing curves (1) and (2) above, it can be seen that the peak level decreases due to the broadening of the spectrum. Furthermore, in curve (3), the FWHM (Full Width at Half Maximum) bandwidth increases by approximately 50 MHz, and the peak level decreases by approximately 28.8%. This change in peak level will result in a large error when separating temperature and strain. In other words, if different strains exist within a fiber of a length comparable to the spatial resolution, two peaks will occur, resulting in the energy of the Brillouin scattered light being split into two, the peak level decreasing, and the temperature and strain separation effect being diminished. Here, Figure 3 shows the peak power of the spectrum of Brillouin scattered light power obtained by BOTDR, and the numerical values ​​of each frequency shown at the top of the figure are the FWHM bandwidths when approximation curves are drawn for each BOTDR spectrum. In other words, the Brillouin scattered light power spectrum obtained by BOTDR has the problem of being affected by the receiving bandwidth and the strain distribution.

[0028] Another problem is that in optical fiber loss monitoring using Brillouin scattered light, variations in the Brillouin gain spectrum (hereinafter abbreviated as BGS) along the optical fiber occur due to factors such as the mixing of different types of optical fibers, manufacturing tolerances, strain along the cable, or temperature distribution.

[0029] To address this problem, an ERA (end-reflection-assisted)-BOTDA, or end-reflection-assisted BOTDA, equipped with a frequency-swept pump pulse that can efficiently compensate for fluctuations in the gain profile along the cable, has been proposed (see, for example, Non-Patent Document 8). This ERA-BOTDA employs a frequency-swept pump pulse, and experimental improvements in sensitivity have been confirmed after correcting BGS fluctuations. However, it is known that using pump pulses with long durations can cause dead zones due to crosstalk before and after large loss events such as 8-way splitters.

[0030] This disclosure discloses a technology to solve the above-mentioned problems, employing a temperature hopping method using an RF modulated input signal that does not require optical frequency hopping synchronized with the emission timing of the RF pulse to achieve high-precision COTDR measurement with suppressed CRN, and by using broad-frequency probe light to achieve high-precision Brillouin intensity measurement by directly acquiring the intensity of the BOTDR signal without using a method that combines Brillouin frequency shift measurement and Brillouin scattered light intensity measurement, thereby obtaining the LPR value, which is the ratio of signal intensities, with high precision from these two types of measurement results, and based on this LPR value, realizing an optical fiber temperature measurement device that can determine the temperature at a predetermined position in the optical fiber with high precision.

[0031] The optical fiber temperature measuring device of this disclosure includes: a COTDR measuring unit that has a first light source whose temperature is controlled by changing the frequency by hopping and whose emitted light linewidth can be changed, and which incidents the emitted light from the first light source onto the optical fiber to be measured and measures the Rayleigh backscattered light from the optical fiber to be measured; a BOTDR measuring unit that has a second light source whose emitted light linewidth can be changed and whose emitted light is swept by changing the frequency with a triangular wave chirp pulse, and which incidents the emitted light from the second light source onto the optical fiber to be measured and measures the Brillouin backscattered light from the optical fiber to be measured; and a calibration optical fiber that is arranged in a constant temperature and distortion-free state on the path of the optical fiber to which the COTDR measuring unit, the BOTDR measuring unit and the optical fiber to be measured are connected. The present invention comprises an LPR analysis unit which determines the intensity of the Rayleigh backscattered light and the Brillouin backscattered light based on the output from the COTDR measurement unit and the output from the BOTDR measurement unit, and measures the temperature of the optical fiber under test at a predetermined position from the Landau-Placek ratio calculated based on the intensity of the Rayleigh backscattered light and the Brillouin backscattered light, and is characterized by measuring the temperature of the optical fiber under test at a predetermined position based on the intensity of the Rayleigh backscattered light, the intensity of the Brillouin backscattered light, and the temperature of the calibration optical fiber measured with the calibration optical fiber.

[0032] The optical fiber temperature measuring device of this disclosure employs a temperature hopping method using an RF modulated input signal that does not require optical frequency hopping synchronized with the emission timing of the RF pulse, thereby achieving high-precision COTDR measurement with suppressed CRN. Furthermore, by using broad-frequency probe light, it is possible to achieve high-precision Brillouin intensity measurement by directly acquiring the intensity of the BOTDR signal without using a method that combines Brillouin frequency shift measurement and Brillouin scattered light intensity measurement. From these two types of measurement results, the LPR value, which is the ratio of signal intensities, can be obtained with high precision, and based on this LPR value, an optical fiber temperature measuring device can be realized that can determine the temperature at a predetermined position of the optical fiber with high precision.

[0033] This figure shows an example of the amplitude of coherent noise in COTDR to explain the problems of the optical fiber temperature measuring device according to Embodiment 1. This figure shows an example of reducing CNR by data summing to explain the problems of the optical fiber temperature measuring device according to Embodiment 1. This figure explains the problem of Brillouin intensity in BOTDR, which is a problem of the optical fiber temperature measuring device according to Embodiment 1. This figure shows an example of the overall configuration of the optical fiber temperature measuring device according to Embodiment 1. This figure shows the detailed configuration of the light source unit input to the COTDR measurement unit of the optical fiber temperature measuring device according to Embodiment 1. This figure shows the detailed configuration of the light source unit input to the BOTDR measurement unit of the optical fiber temperature measuring device according to Embodiment 1. This figure shows the details of the sawtooth wave pulse generation circuit input to the BOTDR measurement unit of the optical fiber temperature measuring device according to Embodiment 1. This table compares the characteristics of the input signal input to the COTDR measurement unit of the optical fiber temperature measuring device according to Embodiment 1 with a conventional example. This figure explains the method for reducing CRN in COTDR measurement of the optical fiber temperature measuring device according to Embodiment 1. This figure shows an example of the simulation results regarding the CRN effect on the output from the BOTDR measurement unit of the optical fiber temperature measuring device according to Embodiment 1. This figure shows another example of the simulation results regarding the CRN effect on the output from the BOTDR measurement unit of the optical fiber temperature measuring device according to Embodiment 1. This figure shows an example of the simulation results regarding the CRN effect on the output from the COTDR measurement unit of the optical fiber temperature measuring device according to Embodiment 1. This figure shows another example of the simulation results regarding the CRN effect on the output from the COTDR measurement unit of the optical fiber temperature measuring device according to Embodiment 1.

[0034] The optical fiber temperature measurement device disclosed herein employs a temperature hopping method for controlling the frequency of the light source in order to suppress the CRN width, which is the amplitude of fluctuations in the COTDR signal, which is a problem with signals obtained by the conventional COTDR method. Furthermore, in temperature measurement using optical fibers that utilize the frequency shift of Brillouin scattering, it was conventionally necessary to measure both the Brillouin frequency shift and the Brillouin scattered light power in order to separate the effect of strain occurring in the optical fiber. However, by using broad frequency probe light, the intensity of the BOTDR signal can be directly acquired, thereby obtaining LPR with high accuracy and realizing highly accurate temperature measurement. The specific details of this device will be described in detail below with reference to the figures.

[0035] Embodiment 1. First, the overall configuration of the optical fiber temperature measuring device 10 according to Embodiment 1 will be described below with reference to Figure 4.

[0036] As mentioned above, LPR(R LP When calculating R, LP = I R / 2I B Therefore, the Rayleigh backscatter light signal intensity I R and Brillouin backscattered light signal intensity I B Two values ​​are needed. Therefore, as shown in Figure 4, the optical fiber temperature measuring device 10 has I R and I B The system is equipped with a COTDR measurement unit 1 and a BOTDR measurement unit 2 for determining the following:

[0037] Furthermore, these two types of measurement units are connected to a calibration optical fiber 8 (hereinafter also referred to as the calibration optical fiber 8) via an optical switch 3 (also called optical SW3) or a WDM unit 4 (also called a wavelength division multiplexer 4), and this calibration optical fiber 8 is further connected to the optical fiber under measurement 7 (also referred to as the measurement FO7). In addition, the COTDR measurement unit 1 and the BOTDR measurement unit 2 and the measurement optics 7 are connected by a shared optical fiber. This shared optical fiber is placed between the optical switch 3 or WDM unit 4 and the measurement optics 7, and the calibration optical fiber 8 is provided on the path of this shared optical fiber, installed inside the constant temperature chamber 9.

[0038] At that time, the calibration optical fiber 8 (approximately 50 m in length) is stored inside the constant temperature bath 9 in a strain-free state (for example, unbundled) and at a constant temperature. The temperature of the constant temperature bath 9 is controlled using a thermistor and a Peltier element (not shown), and the temperature T of the constant temperature bath 9 is controlled. ref The temperature is kept constant. Also, the thermistor R th From the value of this T ref The value of is determined. Here, the temperature measurement accuracy of the thermistor used is set to approximately ±0.01°C. Also, since this calibration optical fiber 8 is placed on the path of the shared optical fiber mentioned above, the calibration optical fiber 8 is measured each time the optical fiber under test 7 is measured. At this time, as described above, the calibration optical fiber 8 is in a constant temperature and unstrained state (for example, in a bundled state), and position z ref The Rayleigh backscatter light intensity P R (z ref ), Brillouin backscatter light intensity P B (z ref Each of these values ​​is measured each time, and the average value of each measurement is P R (ref), P B Find (ref). Also, find position z ref The temperature of the calibration optical fiber 8 was also measured, T ref It is stored as P above. R (ref), P B (ref) and T refHowever, it is always used as a reference value when you want to determine the temperature of the optical fiber 7 under test at a specific location (see equation (4)).

[0039] Furthermore, the Rayleigh backscattered light signal and the Brillouin backscattered light signal, which are outputs from the COTDR measurement unit 1 and the BOTDR measurement unit 2, are sent separately to the LPR analysis unit 5 (also called the Landau-Placek ratio analysis unit 5). Based on these two signals, calculations are performed to determine the temperature at a predetermined position of the optical fiber 7 under test (FO7 under test), and the resulting data, such as the temperature distribution with respect to the change in position of the optical fiber 7 under test, is sent to the output unit 6 and stored. At this time, the LPR analysis unit 5 uses the calibration optical fiber 8, which is in a constant temperature and unstrained state (for example, in a bundled state), to determine the temperature at a predetermined position z ref The Rayleigh backscatter light intensity P R (z ref ), Brillouin backscatter light intensity P B (z ref Each of these values ​​is measured each time, and the average value of each measurement is P R (ref), P B (ref) is required. Also, the predetermined position z of the calibration optical fiber 8 is required. ref The temperature of the calibration optical fiber 8 was also measured, T ref It is stored as P above. R (ref), P B (ref) and T ref However, it is always used as a reference value when you want to determine the temperature of the optical fiber 7 under measurement at a predetermined location.

[0040] Next, the light sources that generate the laser light used in the COTDR measurement unit 1 and the BOTDR measurement unit 2 described above will be explained in detail below, using Figures 5 and 6, respectively.

[0041] Figure 5 shows the configuration of the laser light source used in the COTDR measurement unit 1. In Figure 5, the LD 11 (Laser Diode 11) generates laser light as pump light. At this time, the LD 11 receives a signal from a linewidth control circuit 14 for controlling the linewidth of the laser light via a current drive circuit 15. Simultaneously, a hopping generation circuit 12 is provided that generates a hopping signal for temperature hopping control via a temperature control circuit 13. The laser light generated in the LD 11 is incident on the optical fiber temperature measuring device 10, which is the main body, or on the FO 7 to be measured, through a polarization-dependent isolator 16 and an optical coupler 17.

[0042] Figure 6 shows the configuration of the laser light source used in the BOTDR measurement unit 2. In Figure 6, the LD (local light) 21 (Laser Diode 21; hereinafter also referred to as LD21) generates laser light as local light. At this time, the LD21 receives signals from a linewidth control circuit 24 for controlling the linewidth of the laser light, and signals from a sawtooth pulse generation circuit 22 for generating sawtooth pulses, via a current drive circuit 25. Simultaneously, a temperature control circuit 23 is provided for controlling the temperature of the LD. The laser light generated in the LD21 is incident on the optical fiber temperature measuring device 10, which is the main body, or on the FO 7 to be measured, via an optical coupler 27 from a polarization-dependent isolator 26.

[0043] Next, an example of a sawtooth wave generated by the sawtooth pulse generation circuit 22, which constitutes the BOTDR measurement unit 2, will be explained using Figure 7. A DFB-LD is used to generate sawtooth pulses. First, a triangular wave is input to this DFB-LD by current control. At this time, a control signal is input to this DFB-LD using a control circuit or the like so that the temperature of the LD is kept constant.

[0044] From the LD configured in this way, a signal with a mountain-shaped waveform, as shown just to the upper right of the LD, is output. That is, a sawtooth wave-like signal is formed and repeatedly output, consisting of a waveform in which the frequency increases linearly from f0 (start frequency) to f0 + 15 GHz, and then the frequency decreases linearly from f0 + 15 GHz back to f0. Subsequently, this signal is pulsed with a pulse width of about 100 ns, forming a pulsed signal (also called a pulse signal) as shown on the far right. The rising frequency during this process is pulsed using a pulsation circuit or the like so that it changes, for example, from f0 + 11.5 GHz to f0 + 10.5 GHz, i.e., changes of about 1 GHz. Then, the pulsed wave configured in this way is repeatedly supplied to the current drive circuit 25, and its output is input to the LD 21, driving the LD and generating laser light in the LD 21. In both of the graphed figures, the horizontal axis is the time axis and the vertical axis is the frequency axis.

[0045] Furthermore, by forming numerous chirp-like laser beams of the shape described above with small pulse widths, that is, by scanning multiple chirp pulses in a step-like manner, a probe beam with a broad frequency range can be formed (for details on how to form such a probe beam, see, for example, Patent Document 1), and a BOTDR measurement unit that can utilize such a probe beam can be constructed. Therefore, with such a BOTDR measurement unit, it is possible to directly acquire the intensity of the BOTDR signal (Brillouin backscattered light signal intensity). Note that the Brillouin backscattered light described above is spontaneous Brillouin scattered light. Since this spontaneous Brillouin scattered light originates from the acoustic vibrations of glass molecules, its level fluctuates with temperature changes. On the other hand, stimulated Brillouin scattered light is Brillouin scattered light due to counter-light, so its level does not fluctuate with temperature changes, and measuring its level does not reflect the temperature.

[0046] By the way, LPR is Rayleigh backscatter light signal intensity I R and Brillouin backscattered light signal intensity I B It is expressed as a ratio of and the temperature of the optical fiber can be determined based on this value. Therefore, in order to accurately detect the temperature value, the Rayleigh backscattered light signal intensity I Rand Brillouin backscattered light signal intensity I B Each of these values ​​needs to be determined with high precision. Of these, the Brillouin backscatter signal can be obtained from the BOTDR measurement unit as described above.

[0047] On the other hand, in order to obtain a Rayleigh backscatter light signal with greater accuracy than before using the COTDR measurement unit, it is necessary to suppress the above-mentioned CRN as much as possible. Therefore, this method will be explained in detail below using Figures 8 and 9 and Equation (5). Specifically, one way to improve the above-mentioned CRN is to reduce it using TW-COTDR (tunable wavelength-COTDR, COTDR using a tunable laser). That is, this method involves widening the linewidth of the LD and widening the frequency sweep range. To explain with a specific example, if the data within the relevant frequency range is averaged and the spatial resolution Δz is set to 1 m and the frequency sweep range to 1000 GHz, the width of the CRN is expected to be less than 0.03 dB (see Equation (5) and Figure 9).

[0048] First, Figure 8 shows a comparison of the laser beam used in the COTDR measurement section of this device with the laser beam used in a conventional method (see Non-Patent Document 7). Two items that differ particularly between the conventional method and the method disclosed in this application—the input signal and the temperature change method—were selected, and these two items were compared between the two methods.

[0049] Regarding the input signal, the conventional method uses an RF pulse, whereas the present invention uses an RF-modulated signal. Furthermore, the method for changing the LD temperature differs from the conventional method, which scans and changes the temperature by 13°C from 27°C to 14°C. In contrast, the present invention uses a hopping generation circuit to suppress CRN by changing the temperature through temperature hopping equivalent to 50°C. Due to these differences in the method of suppressing CRN, the CRN suppression effect is expected to be significantly improved from 0.04 dB in the conventional method to 0.002 dB in the present invention.

[0050] Next, we will explain the expected value of the CRN range (see COTDR range explained in Figure 1), which is the CRN suppression effect, using Figure 9 and equation (5).

[0051] Figure 9 is a graph showing the results of calculating the CRN width using equation (5) below, under the conditions that the laser beam used as the input signal has a linewidth of 300 kHz and the frequency sweep width when the input signal is RF modulated to 1000 GHz. Figure 9 shows the CRN value (unit: dB) calculated using equation (5) for six different spatial resolutions, from a minimum of 20 cm to a maximum of 10 m, with the spatial resolution value shown at the top of the figure as a parameter. From this graph, it can be seen that the CRN width (CRN suppression effect) increases as the spatial resolution value and the frequency sweep range increase.

[0052] Here, c represents the speed of light in a vacuum (299,790 km / s), n represents the group refractive index of the optical fiber (typical value: 1.46), and Δz represents the spatial resolution. As a specific example, when the spatial resolution Δz is 1 m and the frequency sweep range is 1000 GHz, the width of the CRN is 0.03 dB (see Figure 9).

[0053] Therefore, from equation (5) and Figure 9, it can be predicted that increasing the linewidth of the laser beam, or in other words, the frequency sweep range, and thereby increasing the spatial resolution will be effective in reducing the CRN. However, there is a constraint that the spatial resolution cannot be increased arbitrarily, as it is necessary to identify the measurement position as accurately as possible.

[0054] Here, regarding the linewidth, as mentioned above, it is usually the linewidth of the laser beam, i.e., the full width at half maximum (FWHM). However, in the case of COTDR, because of CRN, the variability in loss measurement cannot be compressed. In other words, in order to reduce CRN, the linewidth of the laser beam (emitted light) must be increased. From equation (5), it can be seen that in order to reduce CRN, the laser linewidth must be increased, and the methods for doing so are: 1) measure with a different measurement frequency each time, and then perform averaging and summation, and 2) hop the laser temperature and change the laser frequency with asynchronous temperature changes. By implementing this method, the linewidth in equation (5) above can be increased. Note that in BOTDR, there is no effect of CRN, so it is not necessary to increase the laser linewidth. However, the laser frequency is chirpened in order to measure the bandwidth of the natural Brillouin scattered light all at once.

[0055] Under these conditions, the expected CRN width (CRN suppression effect) is determined using equation (5) with a specific example. For example, in equation (5), if n = 1.46, Δz = 10m, and the frequency sweep range (linewidth) is 1000 GHz, then CRN = 0.0098 (dB). Since the accuracy of the loss can be improved by averaging the data within this range, the number of averaging iterations is set to 2. N If we consider it as a number of times, the accuracy of the loss is CRN width / N 0.5 Therefore, for example, N = 24 (number of additions: 2 24 If so, the loss accuracy is 0.098 / 24 0.5 This equals 0.002 (dB).

[0056] Finally, we will explain the results of our investigation into the relationship between the suppression effect of the above-mentioned CRN and the number of additions, using Figures 10 and 11. Figure 10 shows the relationship when the average number of additions is 2 20 Figure 11 shows a graph of the loss (dB) values ​​in BOTDR at a maximum measurement distance of 50 km for each cycle (simulation calculation result), with an average number of additions of 2. 24The graphs (simulation calculation results) show the loss (dB) values ​​in BOTDR at a maximum measurement distance of 50 km for each measurement. To clearly show the amount of loss, in both figures, the lower figure shows an enlarged view of the loss in the 40-45 km distance region of the upper figure. In both Figures 10 and 11, the linewidth (frequency bandwidth) is 500 MHz, the sampling interval is 20 cm, and the spatial resolution is 20 m.

[0057] From Figure 10, the average number of additions is 2 20 The loss (CRN width) in the case of 2 is approximately 0.05 dB, as shown in Figure 11, when the average number of additions is 2 24 It can be seen that the loss amount (CRN width) in the case of 1 is approximately 0.02 to 0.03 dB. By comparing these values, the accuracy of the loss is N 0.5 It was shown that the improvement was in the reciprocal of the result. Furthermore, the relationship between the suppression effect and the number of additions was explained above using a graph of the loss (dB) value in BOTDR.

[0058] Next, the measurement results of the loss (dB) in COTDR when the LD temperature and the number of average summations are varied are confirmed below using Figures 12 and 13. Specifically, the effect of significantly reducing the CRN width by changing the LD temperature and frequency hopping, and the effect of the number of average summations were confirmed. In both Figures 12 and 13, the temperature change of the LD due to hopping is 2°C, and the spatial resolution is 1m.

[0059] Figure 12 shows that the average number of additions is 2 20 Figure 13 shows a graph of the loss (dB) values ​​in COTDR at a maximum measurement distance of 2 km for each cycle (simulation calculation result), where the average number of additions was 2. 24 This graph shows the loss (dB) values ​​in COTDR at a maximum measurement distance of 2 km for each measurement (simulation calculation result). The average number of additions was 2. 20 In the first instance, the loss fluctuation range was ±0.10 dB (see Figure 12), and the average number of additions was 2. 24 In this test, the loss fluctuation range was reduced to ±0.05 dB (see Figure 13).

[0060] In the above example, the temperature change was set to 2°C, but in the actual implementation (hardware), it is possible to change the temperature to about 20°C. In this case, the average number of additions shown in Figure 12 is 2 20 Further improvement is possible compared to the previous result, and the degree of improvement is {(20°C × 12 GHz / °C) / (2°C × 12 GHz / °C)} 0.5 It doubles (= 3.16 times). That is, for a spatial resolution of 1m, the CRN width changes from ±0.1dB to ±0.032dB. Furthermore, if the LD linewidth is changed from 300kHz (see Figure 9) to 10MHz, the improvement is (10MHz / 0.3MHz). 0.5 The value doubles (to 5.8 times), and the CRN width changes from ±0.032 dB to ±0.005 dB.

[0061] As explained above, by widening the linewidth of the pump light to 1000 GHz and generating frequency hopping through temperature control, the width of the CRN in Rayleigh scattered light intensity is improved, R It is expected that the variation in (z) can be reduced to ±0.001 dB or less. This value and the Brillouin backscattered light intensity P B Considering that the coefficient of the relationship between (z) and temperature is 0.36% / °C, it can be seen that by using COTDR and BOTDR measurements with the CRN width improved to ±0.001 dB or less, an optical fiber temperature measurement device with high accuracy of ±0.1°C can be realized. In other words, because the width of the CRN can be improved in this measurement device, measurement accuracy can be ensured regardless of bending or structural loss of the optical fiber, and it is possible to measure temperatures over long distances using SMF and with high accuracy of 0.1°C.

[0062] While this disclosure describes exemplary embodiments, the various features, aspects, and functions described in the embodiments are not limited to the application of any particular embodiment, but can be applied individually or in various combinations to the embodiments. Accordingly, countless variations not illustrated are conceivable within the scope of the art disclosed in this specification. These include, for example, modifications, additions, or omissions of at least one component.

[0063] 1. COTDR measurement unit, 2. BOTDR measurement unit, 3. Optical SW (optical switch), 4. WDM unit, 5. LPR analysis unit, 6. Output unit, 7. Optical fiber under test (FO), 8. Calibration optical fiber, 9. Constant temperature chamber, 10. Optical fiber temperature measurement device, 11. LD (pump light), 12. Hopping generation circuit, 13, 23. Temperature control circuit, 14, 24. Line width control circuit, 15, 25. Current drive circuit, 16, 26. Polarization-dependent isolator, 17, 27. Optical coupler, 21. LD (local light)

Claims

1. A COTDR measurement unit having a first light source whose temperature is controlled by changing the frequency by hopping and whose emitted light linewidth can be changed, which incidents the emitted light from the first light source onto the optical fiber under test and measures the Rayleigh backscattered light from the optical fiber under test; a BOTDR measurement unit having a second light source whose emitted light linewidth can be changed and whose emitted light is swept by changing the frequency with a triangular wave chirp pulse, which incidents the emitted light from the second light source onto the optical fiber under test and measures the Brillouin backscattered light from the optical fiber under test; a calibration optical fiber arranged in a constant temperature and distortion-free state on the path of the optical fiber to which the COTDR measurement unit, the BOTDR measurement unit and the optical fiber under test are connected; An optical fiber temperature measuring device comprising: an LPR analysis unit that determines the intensity of the Rayleigh backscattered light and the Brillouin backscattered light based on the output from the COTDR measurement unit and the output from the BOTDR measurement unit, and measures the temperature of the optical fiber under test at a predetermined position from the Landau-Placek ratio calculated based on the intensity of the Rayleigh backscattered light and the Brillouin backscattered light; and an optical fiber temperature measuring device that measures the temperature of the optical fiber under test at a predetermined position based on the intensity of the Rayleigh backscattered light, the intensity of the Brillouin backscattered light, and the temperature of the calibration optical fiber measured with the calibration optical fiber.

2. The optical fiber temperature measuring device according to claim 1, characterized in that the first light source comprises an LD, a temperature control circuit for controlling the temperature of the LD, a hopping generation circuit for inputting a hopping signal to the temperature control circuit, a current drive circuit for driving the LD, and a line width control circuit for inputting a signal to change the line width of the laser to the current drive circuit.

3. The optical fiber temperature measuring device according to claim 1, characterized in that the second light source comprises an LD, a temperature control circuit for controlling the temperature of the LD, a current drive circuit for driving the LD, and a line width control circuit for inputting a signal to change the line width of the laser to the current drive circuit and a sawtooth pulse generation circuit for generating a sawtooth pulse signal.

4. The optical fiber temperature measuring device according to any one of claims 1 to 3, characterized in that the temperature of the optical fiber under test at a predetermined position is determined from the temperature coefficient of the intensity of the Brillouin backscattered light, the Landau-Placek ratio at the predetermined position of the optical fiber under test, the Landau-Placek ratio of the calibrated optical fiber, and the temperature of the calibrated optical fiber.

5. The amplitude of the signal of the Rayleigh backscattered light obtained by the COTDR measurement unit is in dB units using the speed of light c in vacuum, the group refractive index n of the optical fiber, the spatial resolution Δz, and the line width of the emitted light It is predicted by, and the optical fiber thermometer according to claim 1 or 3 is characterized in that 6. The optical fiber temperature measuring device according to claim 5, characterized in that the amplitude of the Rayleigh backscattered light signal from the COTDR measuring unit and the amplitude of the Brillouin backscattered light signal from the BOTDR measuring unit are obtained by integrating them 2 to the power of 24 or more times.

7. A method for measuring the temperature of an optical fiber, characterized by:

1. Including light emitted from a first light source that controls temperature by changing the frequency by hopping and can change the line width of the emitted light into the optical fiber to be measured, and measuring the Rayleigh backscattered light from the optical fiber to be measured and the calibration optical fiber; 2. Including light emitted from a second light source that can change the line width of the emitted light and sweeps the emitted light by changing the frequency by a triangular wave chirp pulse into the optical fiber to be measured, and measuring the Brillouin backscattered light from the optical fiber to be measured and the calibration optical fiber; 3. Measuring the temperature of the optical fiber to be measured at a predetermined position using the intensities of two types of Rayleigh backscattered light calculated from the measured Rayleigh backscattered light, two types of Landau-Placek ratios calculated based on the intensities of the two types of Brillouin backscattered light calculated from the measured Brillouin backscattered light, and the measured temperature of the calibration optical fiber.

Images