Vibration analysis device

Abstract

The present disclosure is a vibration analysis device that compares a frequency characteristic of a vibration distribution in a section in the length direction of an optical fiber with a reference frequency characteristic to determine a section which has a frequency characteristic differing from the reference frequency characteristic.

Description

Vibration analysis device

[0001] The present disclosure relates to a vibration analysis device that analyzes data on vibrations applied to an optical fiber.

[0002] Advances in distributed acoustic sensing (DAS) technology have made it possible to measure vibrations distributed along the length of optical fibers. Vibration measurements using DAS are expected to capture specific vibrations caused by traffic flow, construction work, and various infrastructure defects.

[0003] The amount of measurement data generated by DAS is enormous, ranging from 1 MB to several GB per second. Measurement data from DAS is generally output as a waterfall (WF) diagram, with the x-axis representing the distance along the optical fiber length, the y-axis representing the elapsed time, and the z-axis representing the vibration amplitude. For example, the WF diagram allows the visual confirmation of the trajectory of passing vehicles (Non-Patent Document 1), but it is difficult to visually detect the anomalous vibration information contained in the measurement data.

[0004] It is known that vibration data can be analyzed by deriving and observing the root mean square (RMS) of the vibration amplitude or a frequency spectrum obtained by Fourier transform, thereby detecting specific vibrations. However, because DAS measurement data contains many noise signals due to environmental vibrations, visual observation, simple RMS comparison, and frequency spectrum comparison can lead to instantaneous noise signals being mistaken for specific vibrations.

[0005] "World's first: Utilizing vibration sensing in communication optical fiber as a center to verify communication equipment monitoring technology - Gaining comprehensive understanding of city environmental information and aiming for use in disaster prevention measures," Nippon Telegraph and Telephone, September 27, 2021, news release, https: / / group.ntt / jp / newsrelease / 2021 / 09 / 27 / 210927a.html

[0006] The present disclosure aims to make it possible to detect a specific vibration whose occurrence position does not change over time, even when measurement data contains many noise signals due to environmental vibrations.

[0007] The vibration analysis device of the present disclosure executes the vibration analysis method of the present disclosure, which compares the frequency characteristics of the vibration distribution in a section in the longitudinal direction of the optical fiber with reference frequency characteristics, and determines sections whose frequency characteristics differ from the reference frequency characteristics.

[0008] The section may be a gauge section obtained by dividing a monitoring section in the length direction of the optical fiber. The vibration analysis device may calculate a frequency spectrum for each data sampling point in the length direction of the optical fiber from the vibration distribution in the gauge section, and calculate the frequency characteristics of the vibration distribution in the gauge section by averaging the frequency spectra at each data sampling point over the gauge section.

[0009] The vibration analysis device may calculate a frequency spectrum for each data sampling point in the length direction of the optical fiber from the vibration distribution in the monitoring section, and calculate the reference frequency characteristics by averaging the frequency spectra at each data sampling point over the monitoring section.

[0010] The vibration analysis device may calculate, for each frequency band, a relative band power with respect to a reference frequency characteristic, and determine the presence of a specific vibration frequency based on a frequency distribution of the relative band power.

[0011] The above disclosures can be combined as much as possible.

[0012] According to the present disclosure, an analysis method is provided for detecting, from DAS measurement data, a specific vibration whose occurrence position does not change over time.

[0013] 1 is a system configuration diagram illustrating an embodiment of the present disclosure; FIG. 2 is an explanatory diagram of a section in the length direction of an optical fiber; FIG. 3 is a functional block diagram illustrating an example of an analysis processing unit; FIG. 4 is a flowchart illustrating an example of the operation of the analysis processing unit; FIG. 5 is an example of a graph of a gauge section in which a specific vibration frequency does not exist; and FIG. 6 is an example of a graph of a gauge section in which a specific vibration frequency exists.

[0014] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. Note that the present disclosure is not limited to the embodiments shown below. These implementation examples are merely illustrative, and the present disclosure can be implemented in various forms with various modifications and improvements based on the knowledge of those skilled in the art. Note that components with the same reference numerals in this specification and drawings indicate the same components.

[0015] 1 shows an example of a system configuration illustrating an embodiment of the present disclosure. The vibration detection system according to this embodiment includes an optical fiber 91, an optical fiber distribution type vibration measuring instrument 11 that outputs vibration data representing vibrations applied to each point in the length direction of the optical fiber 91, and an analysis processing unit 12 that analyzes the vibration data.

[0016] The vibration measuring device 11 is a means capable of measuring the change over time in vibration applied to each point in the longitudinal direction of the optical fiber cable 91, and may be, for example, a distributed optical fiber vibration measuring device (DAS: Distributed Acoustic Sensing). Vibration measurement using a DAS can measure vibration at each point in the longitudinal direction of the optical fiber from one end of the optical fiber included in the optical fiber cable 91. Therefore, the analysis processing unit 12 can calculate the vibration distribution in a section in the longitudinal direction of the optical fiber included in the optical fiber cable 91 from the vibration data.

[0017] The analysis processing unit 12 functions as the vibration analysis device of the present disclosure. The specific vibrations that are the subject of the present disclosure are vibrations whose location does not change over time. The analysis processing unit 12 compares the frequency characteristics of the vibration distribution in a section along the length of the optical fiber cable 91 with reference frequency characteristics, and identifies sections whose frequency characteristics differ from the reference frequency characteristics to determine the specific vibrations.

[0018] In this disclosure, the total span length over which vibration can be detected by the DAS is defined as L. In this embodiment, as shown in FIG. 2 , a certain span length X included in L is set and referred to as a monitoring section. Here, span length X is a unit length for monitoring peculiar vibrations, and it is assumed that when no peculiar vibrations are occurring, the vibration pattern is constant within the span length X. When optical fiber in an existing network is used as the sensing optical fiber of the DAS, the appropriate span length X is approximately the span between manholes or utility poles. When laying a new sensing optical fiber, the span length X should be within a range that allows the optical fiber to be laid uniformly.

[0019] In this disclosure, a monitoring section with a span length of X is divided into N equal sections, and a gauge section with a length of X / N is used. X / N is set to approximately the expected transmission distance in the longitudinal direction of the optical fiber when a specific vibration occurs. The spatial resolution of the DAS and the interval between data sampling points in the longitudinal direction are set to be finer than X / N. Hereinafter, each gauge section will be referred to as x ｎ (n is an integer from 1 to N).

[0020] 3 shows an example of functional blocks of the analysis processing unit 12. The analysis processing unit 12 includes a reference spectrum calculation unit 22, a gauge section spectrum calculation unit 23, a frequency power component calculation unit 24, a graph creation unit 26, and a specific vibration point identification unit 27.

[0021] 4 shows an example of a vibration analysis method executed by the analysis processing unit 12. When vibration data is input to the analysis processing unit 12 (S11), the reference spectrum calculation unit 22 executes step S12, the gauge section spectrum calculation unit 23 executes step S13, the frequency power component calculation unit 24 executes steps S14 and S15, the graph creation unit 26 executes step S16, and the peculiar vibration point identification unit 27 executes step S17. Note that steps S12 and S13 may be executed first, or may be executed simultaneously.

[0022] In step S12, the reference spectrum calculation unit 22 calculates the frequency characteristic averaged over the monitoring section as the reference frequency characteristic. In step S13, the gauge section spectrum calculation unit 23 calculates the gauge section x ｎIn step S15, the frequency power component calculation unit 24 calculates the relative band power with respect to the reference frequency characteristic for each frequency band. In steps S16 and S17, the graph creation unit 26 and the peculiar vibration point identification unit 27 determine the presence of a peculiar vibration frequency based on the frequency distribution of the relative band power.

[0023] (Step S12) The reference spectrum calculation unit 22 performs a Fourier transform on the vibration distribution in the monitoring section for each data sampling point in the length direction of the optical fiber 91. This calculates the frequency spectrum at each data sampling point in the monitoring section. The monitoring section spectrum calculation unit 23 averages the calculated frequency spectra at each data sampling point over the monitoring section. This calculates the average frequency spectrum f ＸＳ is calculated.

[0024] (Step S13) The gauge section spectrum calculation unit 23 calculates the gauge section x ｎ The vibration distribution in the gauge section x is subjected to a Fourier transform at each data sampling point in the length direction of the optical fiber 91. ｎ The frequency spectrum at each data sampling point is calculated. The gauge section spectrum calculation unit 23 averages the calculated frequency spectrum at each data sampling point over the gauge section. This calculation is performed for each gauge section. As a result, the frequency spectrum for all gauge sections x ｎ Average frequency spectrum f Ｘｎ is calculated.

[0025] (Step S14) The frequency power component calculation unit 24 calculates the average frequency spectrum f ＸＳ and f Ｘｎ are divided into M parts on the frequency axis, and the power component of the mth frequency band (m is an integer from 1 to M) is expressed as Δf ＸＳ (m) and Δf Ｘｎ Let (m).

[0026] (Step S15) The frequency power component calculation unit 24 calculates Δf ＸＳ (m) and Δf ＸｎFor example, the frequency power component calculation unit 24 calculates the power ratio of ΔRBP (m). ｎ (m) = Δf Ｘｎ (m) / Δf ＸＳ (m) (1) In the present disclosure, ΔRBP ｎ (m) is called relative band power (RBP).

[0027] (Step S16) The graph creation unit 26 creates a graph for each gauge section, with the horizontal axis m and the vertical axis ΔRBPn(m). As shown in FIGS. 5 and 6, ｎ When the graph of (m) is drawn, if there is no specific vibration frequency, as shown in FIG. ｎ When a specific vibration frequency exists, the vertical axis ΔRBP becomes a constant graph as shown in FIG. ｎ (m) is a graph that is convex upward or convex downward.

[0028] (Steps S17 and S18) The specific vibration point identifying unit 27 calculates the ΔRBP ｎ Gauge section x when (m) is not constant ｎ The specific vibration point specifying unit 27 extracts, for example, Δf Ｘｎ (m) / Δf ＸＳ The gauge section x where the change in (m) relative to m is equal to or greater than a predetermined value ｎ Then, the specific vibration point identifying unit 27 extracts the extracted gauge section x ｎ is output as the singular vibration point (S18).

[0029] As described above, the present disclosure provides a method for measuring vibration data obtained by measuring vibration distribution in the length direction of the optical fiber 91 using the vibration measuring device 11, and then dividing the monitoring section into N gauge sections x ｎ and set the average f for the entire monitoring section of the frequency spectrum at each data sampling point. ＸＳ , and the average f for the nth gauge interval Ｘｎ (n=1, 2, ..., N) and ＸＳ and f Ｘｎ The power component Δf of each of the M divided frequency bands Ｘ(m) and Δf Ｘｎ (m) (m=1, 2, ..., M) is calculated, and Δf Ｘｎ (m) / Δf ＸＳ The nth gauge section x where the change in (m) relative to m is equal to or greater than a predetermined value ｎ is detected as the gauge section where the singular vibration occurs.

[0030] When vibration measurement using a DAS involves environmental vibrations or disturbances (noise) that change over time, the vibration measurement should be performed for a time period long enough to accommodate the change in the disturbance, and the average frequency spectrum f Ｘ and f Ｘｎ For example, vehicle traffic may be a disturbance, but vibration measurements are taken over a period of time when the number and types of vehicles are statistically equivalent, and the average frequency spectrum f Ｘ and f Ｘｎ By calculating this, the impact of vehicle traffic can be reduced.

[0031] (Second embodiment) Frequency band f ａ ±Δf ａ Therefore, in this embodiment, it is possible to detect such specific vibrations.

[0032] In this embodiment, in step S14, the frequency power component calculation unit 24 calculates the average frequency spectrum f Ｘ From frequency band f ａ ±Δf ａ is extracted and divided by M on the wave number axis, Ｘ The frequency power component calculation unit 24 calculates the average frequency spectrum f Ｘｎ From frequency band f ａ ±Δf ａ is extracted and divided by M on the wave number axis, Ｘｎ (m) is calculated.

[0033] As a result, in this embodiment, the frequency band f ａ ±Δf ａ Furthermore, since the number of times steps S15 to S17 are performed can be reduced, the calculation cost can also be reduced.

[0034] Third Embodiment In some cases, vibration data before the occurrence of a specific vibration is available. In this embodiment, the specific vibration can be detected using the vibration data before the occurrence of the specific vibration.

[0035] In this embodiment, the reference spectrum calculation unit 22 calculates the average frequency spectrum f′ in the monitoring section using vibration data before the occurrence of the specific vibration. ＸＳ The subsequent processing is the same as in the first embodiment. For example, in step S15, the following equation is calculated: ΔRBP ｎ (m) = Δf Ｘｎ (m) / Δf′ ＸＳ (m) (2)

[0036] In this embodiment, the graph in step S16 can be created using vibration data before the occurrence of specific vibration, which allows the specific vibration frequency to be determined more accurately.

[0037] As described above, the present disclosure compares the power ratio for each frequency component between the frequency spectrum in a monitoring section in which uniform vibrations can be measured under conditions in which no specific vibrations are occurring and the frequency spectrum in a gauge section divided into fixed lengths in the monitoring section, and by extracting divided sections in which the power ratio is not constant, it identifies sections in which specific vibrations are occurring whose occurrence position does not change over time.

[0038] As a result, according to the present disclosure, even if the measurement data contains many noise signals due to environmental vibrations, it is possible to accurately detect peculiar vibrations whose occurrence position does not change over time. Therefore, according to the present disclosure, even if the measurement data contains many noise signals due to environmental vibrations, it is possible to identify the section in which the peculiar vibration frequency is occurring from the DAS measurement data.

[0039] The analysis processing unit 12 of the present disclosure can also be realized by a computer and a program, and the program can be recorded on a recording medium or provided via a network.

[0040] 11: Vibration measuring instrument 12: Analysis processing unit 91: Optical fiber 22: Reference spectrum calculation unit 23: Gauge section spectrum calculation unit 24: Frequency power component calculation unit 26: Graph creation unit 27: Singular vibration point identification unit

Claims

1. A vibration analysis device that compares the frequency characteristics of the vibration distribution in a section along the length of an optical fiber with a reference frequency characteristic, and determines which sections have frequency characteristics that differ from the reference frequency characteristic.

2. The vibration analysis device according to claim 1, wherein the section is a gauge section obtained by dividing a monitoring section in the longitudinal direction of the optical fiber, and the vibration analysis device calculates a frequency spectrum for each data sampling point in the longitudinal direction of the optical fiber from the vibration distribution in the gauge section, and calculates the frequency characteristics of the vibration distribution in the gauge section by averaging the frequency spectra at each data sampling point over the gauge section.

3. The vibration analysis device according to claim 2, wherein a frequency spectrum is calculated for each data sampling point in the length direction of the optical fiber from the vibration distribution in the monitoring section, and the reference frequency characteristics are calculated by averaging the frequency spectra at each data sampling point over the monitoring section.

4. The vibration analysis device according to claim 1, which calculates a relative band power with respect to a reference frequency characteristic for each frequency band, and judges the presence of a specific vibration frequency based on a frequency distribution of the relative band power.

Images