An optical frequency domain reflectometry demodulation method and device based on adaptive local spectral matching and a sensing system
By using an adaptive local spectral matching method to demodulate the optical frequency domain reflection signal, the problem of insufficient demodulation accuracy and reliability of the traditional OFDR method at high resolution is solved, and high-resolution temperature/strain sensing with a small sweep frequency range is realized.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-11
- Publication Date
- 2026-04-10
AI Technical Summary
Traditional optical frequency domain reflectance (OFDR) demodulation methods are susceptible to system noise and environmental interference when high sensing resolution is required, which leads to a decrease in demodulation accuracy and reliability. Furthermore, the fixed cut-off length and wavelength range of local spectroscopy limit high-resolution sensing of a wide range of temperature/strain.
An adaptive local spectral matching method is adopted. By extracting local spectra with different truncation lengths from the global reference and measurement spectra, the offset error value is calculated, and the local reference spectrum with the optimal wavelength range is selected as the wavelength offset, thereby achieving high spatial resolution sensing within a small frequency sweep range.
Temperature/strain sensing with a large measurement range and high spatial resolution was achieved under a small sweep frequency range, improving demodulation accuracy and reliability.
Smart Images
Figure CN121026197B_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to optical frequency domain sensing technology, and in particular to an optical frequency domain reflectometry demodulation method, device and sensing system based on adaptive local spectrum matching method. BACKGROUND
[0002] As a high-performance distributed optical fiber sensing technology, optical frequency domain reflectometry (OFDR) technology has attracted extensive attention and in-depth research from the academic and engineering communities at home and abroad in recent years due to its ultra-high spatial resolution and excellent sensing accuracy. This technology has great potential in large-scale structure health detection, large-range strain measurement, real-time monitoring in complex environments, and has been applied and verified in many fields such as aerospace, civil engineering, energy and power.
[0003] Traditional OFDR demodulation methods usually use cross-correlation operation to calculate the wavelength shift of the reference signal and the measurement signal in the wavelength domain spectrum. This method is widely used because of its simplicity and efficiency. However, with the demand for higher sensing resolution, the traditional cross-correlation algorithm is easily affected by system noise, environmental interference, and noise of the scattered signal itself during demodulation, which reduces the similarity between the spectra and causes problems such as multiple peaks and pseudo-peak in the cross-correlation results, which seriously affects the demodulation accuracy and sensing reliability.
[0004] Local spectrum method is a demodulation method that calculates the spectral similarity between local reference spectrum and multiple local measurement spectra, and takes the wavelength shift between the local reference spectrum with the largest similarity and the local measurement spectrum as the wavelength shift between the reference signal and the measurement signal. It can improve the sensing accuracy of the OFDR system under large-range temperature / strain changes.
[0005] However, the spatial resolution of the OFDR system is directly limited by the sweep range in physics. In order to achieve higher sensing spatial resolution, a sweep range of no less than 20N(M) is usually required, which increases the hardware requirements for tunable lasers of the OFDR system, thereby increasing the cost. In addition, the length and wavelength range of the local reference spectrum in the local spectrum method are fixed, which also has certain limitations for high-resolution sensing of large-range temperature / strain. SUMMARY
[0006] In order to solve the above problems of the prior art, the present application provides an optical frequency domain reflectometry demodulation method, device and sensing system based on adaptive local spectrum matching method, which can realize temperature / strain sensing with large measurement range and high spatial resolution under small sweep range conditions.
[0007] The technical problems to be solved by the present application are solved by the following technical solutions:
[0008] A light frequency domain reflection demodulation method based on adaptive local spectrum matching method, comprising the following steps:
[0009] Step 100: obtaining a reference signal and a measurement signal;
[0010] Step 200: converting the reference signal and the measurement signal into wavelength domain respectively to obtain a global reference spectrum and a global measurement spectrum;
[0011] Step 300: using different intercept lengths to intercept the global reference spectrum respectively to obtain a plurality of local reference spectra with different wavelength ranges;
[0012] Step 400: using the same plurality of intercept lengths as the global reference spectrum to intercept the global measurement spectrum respectively to obtain a plurality of local measurement spectra with different wavelength ranges;
[0013] Step 500: according to each local reference spectrum and each local measurement spectrum with the same intercept length, calculating the offset error value of each local reference spectrum, and selecting the local reference spectrum with the minimum offset error value, so as to take the best offset amount of the local reference spectrum as the wavelength offset amount between the reference signal and the measurement signal.
[0014] Further, in step 200, the steps of converting the reference signal and the measurement signal into wavelength domain respectively to obtain a global reference spectrum and a global measurement spectrum are as follows:
[0015] Step 210: converting the reference signal and the measurement signal into distance domain respectively by fast Fourier transform to obtain a reference distance domain signal and a measurement distance domain signal;
[0016] Step 220: performing windowing and zero padding operation on the reference distance domain signal and the measurement distance domain signal respectively;
[0017] Step 230: converting the reference distance domain signal and the measurement distance domain signal after windowing and zero padding into wavelength domain by inverse fast Fourier transform to obtain the global reference spectrum and the global measurement spectrum.
[0018] Further, in step 300, when the global reference spectrum is intercepted, the starting position of the interception is the center position of the global reference spectrum, and the termination position of the interception is the position on both sides of the global reference spectrum.
[0019] Further, in each local reference spectrum with the same intercept length, the center position between one of the local reference spectra and the global reference spectrum is the same.
[0020] Further, in each of the partial reference spectra with the same length, the two adjacent partial reference spectra do not overlap or partially overlap with an overlap range of 25%-95%.
[0021] Further, in step 400, when the global measurement spectrum is intercepted, the starting position of the interception is one side position of the global measurement spectrum, and the ending position of the interception is the other side position of the global measurement spectrum.
[0022] Further, in each of the partial measurement spectra with the same length, the two adjacent partial measurement spectra partially overlap with an overlap range of 25%-95%.
[0023] Further, in step 500, according to each of the partial reference spectra and each of the partial measurement spectra with the same length, the offset error value of each of the partial reference spectra is calculated, and the partial reference spectrum with the minimum offset error value is selected, so that the optimal offset of the partial reference spectrum is taken as the wavelength offset between the reference signal and the measurement signal. The steps are as follows:
[0024] Step 510: the spectral similarity between each of the partial reference spectra and each of the partial measurement spectra with the same length is calculated;
[0025] Step 520: each of the spectral similarities of each of the partial reference spectra is individually compared, and the partial measurement spectrum with the maximum spectral similarity is matched for each of the partial reference spectra;
[0026] Step 530: the wavelength offset between each of the partial reference spectra and each of the partial measurement spectra with the same length is calculated;
[0027] Step 540: the wavelength offset between each of the partial reference spectra and the matched partial measurement spectrum is taken as the optimal offset of each of the partial reference spectra;
[0028] Step 550: the offset error value between the optimal offset of each of the partial reference spectra and the wavelength offset of each of the partial measurement spectra with the same length is calculated;
[0029] Step 560: each of the offset error values is compared, the partial reference spectrum with the minimum offset error value is selected, and the optimal offset of the partial reference spectrum is taken as the wavelength offset between the reference signal and the measurement signal.
[0030] An optical frequency domain reflectance demodulation device based on an adaptive local spectrum matching method, comprising a processor and a memory, the memory stores a computer program for the processor to execute, and the processor performs the optical frequency domain reflectance demodulation method when executing the computer program.
[0031] An optical frequency domain reflectance sensing system based on an adaptive local spectrum matching method, comprising the optical frequency domain reflectance demodulation device, a data acquisition device, an optical frequency domain reflectometer, a first-in-first-out module and a sensing optical fiber, the optical frequency domain reflectometer is connected to the sensing optical fiber through the first-in-first-out module, the data acquisition device is connected to the optical frequency domain reflectometer, and the data acquisition device is connected to the optical frequency domain reflectance demodulation device.
[0032] The optical frequency domain reflectance demodulation method has the following beneficial effects: by setting different interception lengths, the global reference spectrum is intercepted to obtain multiple local reference spectra with different wavelength ranges, and according to the offset error value of each local reference spectrum, the local reference spectrum with the optimal wavelength range is selected, so that the optimal offset of the local reference spectrum is used as the wavelength offset between the reference signal and the measurement signal, and finally, the temperature / strain sensing with a large measurement range and high spatial resolution is demodulated under the condition of a small sweep frequency range. BRIEF DESCRIPTION OF DRAWINGS
[0033] Figure 1 A structural principle diagram of the optical frequency domain reflectance sensing system provided by the present application.
[0034] Figure 2 A step block diagram of the optical frequency domain reflectance demodulation method provided by the present application.
[0035] Figure 3 A sub-step block diagram of step 200 in the optical frequency domain reflectance demodulation method provided by the present application.
[0036] Figure 4 A sub-step block diagram of step 500 in the optical frequency domain reflectance demodulation method provided by the present application.
[0037] Figure 5 Strain curve diagrams obtained by respectively demodulating the same set of reference signals and measurement signals by the optical frequency domain reflectance demodulation method and the conventional cross-correlation algorithm. DETAILED DESCRIPTION
[0038] The present application will be described in detail below with reference to the accompanying drawings and examples, wherein the same or similar reference numerals represent the same or similar elements or elements having the same or similar functions throughout the drawings. The examples described below by reference to the drawings are exemplary and are intended to explain the present application, and cannot be understood as limiting the present application.
[0039] In the description of the present application, it needs to be understood that the terms "cut length", "width", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, which is only for the convenience of describing the present application and simplifying the description, and does not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application.
[0040] In addition, the terms "first", "second", "third" are only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features referred to. Therefore, the features defined with "first", "second", "third" can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "multiple" is two or more, unless otherwise specifically limited.
[0041] In the present application, unless otherwise specifically defined and limited, the terms "mounting", "connection", "connection", "fixing", "setting" and the like should be broadly understood, for example, it can be fixedly connected, or it can be detachably connected, or it can be integrated; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two elements or the interaction relationship between two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.
[0042] Embodiment one
[0043] As Figure 1 shown, a light frequency domain reflectance sensing system based on adaptive local spectral matching method, comprising light frequency domain reflectance demodulation device 1 and data acquisition device 2, light frequency domain reflectometer 3, first-in first-out module 4, sensing optical fiber 5, the light frequency domain reflectometer 3 is connected to the sensing optical fiber 5 through the first-in first-out module 4, the data acquisition device 2 is connected to the light frequency domain reflectometer 3, the data acquisition device 2 is connected to the light frequency domain reflectance demodulation device 1.
[0044] In operation, the measurement light signal emitted by the optical frequency domain reflectometer 3 enters the sensing optical fiber 5 through the first-in-first-out module 4, the Rayleigh scattered light generated by the sensing optical fiber 5 is input into the optical frequency domain reflectometer 3 according to the principle of first-in-first-out after being collected by the first-in-first-out module 4, the optical frequency domain reflectometer 3 converts the Rayleigh scattered light into beat signals and outputs the beat signals to the data acquisition device 2, the data acquisition device 2 collects the beat signals, and outputs the reference signal and the measurement signal to the optical frequency domain reflectometer demodulation device 1, the optical frequency domain reflectometer demodulation device 1 demodulates the reference signal and the measurement signal based on the adaptive local spectrum matching method, and finally obtains the wavelength shift of the sensing optical fiber 5.
[0045] The optical frequency domain reflectometer 3 comprises a tunable laser 31, a first optical fiber coupler 32, an auxiliary interference arm, a main interference arm, a first photoelectric balance detector 314, a second photoelectric balance detector 315 and a third photoelectric balance detector 316, the tunable laser 31 is connected to the auxiliary interference arm and the main interference arm through the first optical fiber coupler 32, the auxiliary interference arm is connected to the first photoelectric balance detector 314, the main interference arm is connected to the second photoelectric balance detector 315 and the third photoelectric balance detector 316 respectively, and the first photoelectric balance detector 314, the second photoelectric balance detector 315 and the third photoelectric balance detector 316 are all connected to the data acquisition device 2.
[0046] The auxiliary interference arm comprises a first optical fiber circulator 33, a second optical fiber coupler 34, a delay optical fiber 35, a first Faraday optical mirror 36 and a second Faraday optical mirror 37, the first port of the first optical fiber circulator 33 is connected to the first optical fiber coupler 32, the second port of the first optical fiber circulator 33 is connected to the first port of the second optical fiber coupler 34, and the third port of the first optical fiber circulator 33 is connected to the first photoelectric balance detector 314; the second port of the second optical fiber coupler 34 is connected to the first photoelectric balance detector 314, the third port of the second optical fiber coupler 34 is connected to the first Faraday optical mirror 36 through the delay optical fiber 35, and the fourth port of the second optical fiber coupler 34 is connected to the second Faraday optical mirror 37.
[0047] The main interference arm comprises a third fiber coupler 38, a polarization controller 39, a second fiber circulator 310, a fourth fiber coupler 311, a first polarization beam splitter 312 and a second polarization beam splitter 313, the first port of the third fiber coupler 38 is connected to the first fiber coupler 32, the second port of the third fiber coupler 38 is connected to the first port of the fourth fiber coupler 311 through the polarization controller 39, and the third port of the third fiber coupler 38 is connected to the first port of the second fiber circulator 310; the second port of the second fiber circulator 310 is connected to the first-in-first-out module 4, and the third port of the second fiber circulator 310 is connected to the second port of the fourth fiber coupler 311; the third port of the fourth fiber coupler 311 is connected to the second photoelectric balance detector 315 and the third photoelectric balance detector 316 through the first polarization beam splitter 312 respectively, and the fourth port of the fourth fiber coupler 311 is connected to the second photoelectric balance detector 315 and the third photoelectric balance detector 316 through the second polarization beam splitter 313 respectively.
[0048] In use, the swept frequency light emitted by the tunable laser 31 is split into 1:99 after passing through the first fiber coupler 32, of which 1% enters the auxiliary interference arm and the other 99% enters the main interference arm; the swept frequency light in the auxiliary interference arm is split into two beams after passing through the first fiber circulator 33 and the second fiber coupler 34, one of which reaches the first Faraday rotatory mirror 36 through the delay fiber 35, and the other directly reaches the second Faraday rotatory mirror 37, and the two beams return to the second fiber coupler 34 to generate first beat frequency interference light after being reflected by the first Faraday rotatory mirror 36 and the second Faraday rotatory mirror 37 respectively, and the first beat frequency interference light is finally input into the first photoelectric balance detector 314 to convert the optical signal into a first beat frequency signal; the swept frequency light in the main interference arm is split into two beams after passing through the third fiber coupler 38, one of which enters the fourth fiber coupler 311 through the polarization controller 39, and the other enters the first-in-first-out module 4 after passing through the second fiber circulator 310, and then enters the fourth fiber coupler 311 after forming Rayleigh scattering light in the sensing fiber 5, the Rayleigh scattering light and the swept frequency light generate second beat frequency interference light in the fourth fiber coupler 311, the second beat frequency interference light is split into two beams after passing through the fourth fiber coupler 311, and then is polarized and split by the first polarization beam splitter 312 and the second polarization beam splitter 313 respectively, and finally is input into the second photoelectric balance detector 315 and the third photoelectric balance detector 316 to convert the optical signal into a second beat frequency signal.
[0049] The data acquisition device 2 collects the second beat frequency signal as a sampling clock signal, and when the measurement parameters such as temperature and strain in the target environment do not change, the data collected by the data acquisition device 2 is the reference signal, and when the measurement parameters such as temperature and strain in the target environment change, the data collected by the data acquisition device 2 is the measurement signal.
[0050] Embodiment two
[0051] As shown in Figure 2 A light frequency domain reflectometry demodulation method based on an adaptive local spectral matching method is used in the light frequency domain reflectometry sensing system of embodiment one to demodulate the reference signal and the measurement signal collected by the data acquisition device.
[0052] The light frequency domain reflectometry demodulation method includes the following steps:
[0053] Step 100: Obtain the reference signal and the measurement signal.
[0054] In this step 100, the data acquisition device collects the second beat frequency signal output by the light frequency domain reflectometer as a sampling clock signal, and when the measurement parameters such as temperature and strain in the target environment do not change, the data collected by the data acquisition device is the reference signal, and when the measurement parameters such as temperature and strain in the target environment change, the data collected by the data acquisition device is the measurement signal. The light frequency domain reflectometry demodulation device is connected with the data acquisition device, so as to obtain the reference signal and the measurement signal collected by the data acquisition device.
[0055] Step 200: Convert the reference signal and the measurement signal into the wavelength domain respectively to obtain a global reference spectrum and a global measurement spectrum.
[0056] In this step 200, the global reference spectrum and the global measurement spectrum show the data distribution of the reference signal and the measurement signal in the wavelength domain respectively, with the wavelength as the horizontal axis and the optical power or the optical intensity as the vertical axis.
[0057] Specifically, as shown in Figure 3 In step 200, the steps of converting the reference signal and the measurement signal into the wavelength domain respectively to obtain a global reference spectrum and a global measurement spectrum are as follows:
[0058] Step 210: Convert the reference signal and the measurement signal into the distance domain respectively by fast Fourier transform (FFT) to obtain a reference distance domain signal and a measurement distance domain signal.
[0059] Step 220: windowing and zero padding operation is performed on the reference range domain signal and the measured range domain signal respectively;
[0060] Step 230: inverse fast Fourier transform (IFFT) is performed on the windowed and zero padded reference range domain signal and the measured range domain signal respectively to convert them into wavelength domain, obtaining the global reference spectrum and the global measured spectrum.
[0061] The fast Fourier transform (FFT), windowing and zero padding operation, and inverse fast Fourier transform (IFFT) performed on the reference signal and the measured signal are all prior art in the field, so the individual steps in step 200 will not be described in detail.
[0062] Step 300: the global reference spectrum is cut at different cutting lengths to obtain multiple local reference spectra with different wavelength ranges.
[0063] In this step 300, the cutting length refers to the length of the local reference spectrum in the horizontal axis direction or the wavelength direction of the global reference spectrum. All local reference spectra are divided into multiple groups of local reference data, one group of local reference data corresponds to one cutting length, and the number of spectra in each group of local reference data is at least two. Since the spectral length of the global reference spectrum is fixed and the cutting length is variable, the number of local reference spectra cut each time is not necessarily the same, that is, the number of spectra in each group of local reference data can be the same or different.
[0064] Suppose there are M cutting lengths, M ≥ 2, which are the first cutting length, the second cutting length, …, and the Mth cutting length. In step 300, the global reference spectrum is first cut at the first cutting length to obtain a first group of local reference data containing A1 local reference spectra; then the global reference spectrum is cut at the second cutting length to obtain a second group of local reference data containing A2 local reference spectra; …; finally, the global reference spectrum is cut at the Mth cutting length to obtain an Mth group of local reference data containing Am local reference spectra; A1, A2, …, Am ≥ 2.
[0065] Among the same group of local reference data, the cutting lengths of the local reference spectra are the same, but the wavelength ranges are different; among different groups of local reference data, the cutting lengths and wavelength ranges of the local reference spectra are all different.
[0066] In addition to the length of the cut, the number of spectra in each set of local reference data also depends on the overlapping range between two adjacent local reference spectra. In the same set of local reference data, i.e. in each local reference spectrum with the same length of cut, there is no overlapping between two adjacent local reference spectra, or there is partial overlapping between two adjacent local reference spectra, and the overlapping range is between 25% and 95%.
[0067] Assuming that the sweep range of the tunable laser is 1548 nm-1552 nm, the length of the global reference spectrum is 4 nm, the lengths of the cut for cutting the global reference spectrum are 2 nm, 2.5 nm and 3 nm respectively, the center interval between two adjacent local reference spectra is 0.5 nm (the overlapping ranges are 25%, 20% and 17% respectively), and finally three sets of local reference data are obtained. Among them, the length of the cut for the first set of local reference data is 2 nm, the number of spectra is 5, and the wavelength ranges are 1548.0 nm-1550.0 nm, 1548.5 nm-1550.5 nm, 1549.0 nm-1551.0 nm, 1549.5 nm-1551.5 nm and 1550.0 nm-1552.0 nm respectively; the length of the cut for the second set of local reference data is 2.5 nm, the number of spectra is 4, and the wavelength ranges are 1548.0 nm-1550.5 nm, 1548.5 nm-1551.0 nm, 1549.0 nm-1551.5 nm and 1549.5 nm-1552.0 nm respectively; the length of the cut for the third set of local reference data is 3 nm, the number of spectra is 3, and the wavelength ranges are 1548.0 nm-1551.0 nm, 1548.5 nm-1551.5 nm and 1549.0-1552.0 nm respectively.
[0068] When cutting the global reference spectrum, the starting position of the cut is preferably the center position of the global reference spectrum, and the ending position of the cut is preferably the position on both sides of the global reference spectrum, i.e. a cut frame is set according to the length of the cut at the center position of the global reference spectrum, and then the cut frame is stepwise moved from the center position to the minimum wavelength position on the left side, and then the cut frame is stepwise moved from the center position to the maximum wavelength position on the right side, or vice versa, i.e. the cut frame is stepwise moved from the center position to the maximum wavelength position on the right side, and then the cut frame is stepwise moved from the center position to the minimum wavelength position on the left side. The amount of stepwise movement each time is the center interval between two adjacent local reference spectra.
[0069] Since the spectral features of the global reference spectrum are mostly located on the central position, the global reference spectrum is cut from the central position to both sides, so that in the same set of local reference data, that is, in each local reference spectrum with the same cutting length, the central positions of one of the local reference spectrums and the global reference spectrum are the same, that is, at least one local reference spectrum contains most of the spectral features of the global reference spectrum.
[0070] Step 400: The global measurement spectrum is cut with the same cutting length as the global reference spectrum to obtain a plurality of local measurement spectrums with different wavelength ranges.
[0071] In this step 400, the cutting length refers to the length of the local measurement spectrum in the horizontal or wavelength direction of the global measurement spectrum. All local measurement spectrums are divided into a plurality of sets of local measurement data, one set of local measurement data corresponds to one cutting length, and the number of spectrums in each set of local measurement data is at least two. Since the spectral length of the global measurement spectrum is fixed and the cutting length is variable, the number of local measurement spectrums cut each time is not necessarily the same, that is, the number of spectrums in each set of local measurement data can be the same or different.
[0072] Suppose the cutting length has M, M≥2, respectively, the first cutting length, the second cutting length, …, the Mth cutting length, in step 400, the global measurement spectrum is cut with the first cutting length to obtain the first set of local measurement data containing B1 local measurement spectrums; the global measurement spectrum is cut with the second cutting length to obtain the second set of local measurement data containing B2 local measurement spectrums; …; finally, the global measurement spectrum is cut with the Mth cutting length to obtain the Mth set of local measurement data containing Bm local measurement spectrums; B1, B2, …, Bm≥2.
[0073] Among the same set of local measurement data, the cutting lengths of each local measurement spectrum are the same, but the wavelength ranges are different; among different sets of local measurement data, the cutting lengths and wavelength ranges of each local measurement spectrum are different.
[0074] In addition to the cutting length, the number of spectrums in each set of local measurement data also depends on the overlap range between adjacent local measurement spectrums. In the same set of local measurement data, that is, in each local measurement spectrum with the same cutting length, the adjacent local measurement spectrums are locally overlapped, and the overlap range is between 25%-95%.
[0075] It is particularly pointed out that, when the global reference spectrum and the global measurement spectrum are respectively intercepted with the same intercept length, the overlapping range between the two adjacent local reference spectra can be the same as or different from the overlapping range between the two adjacent local measurement spectra. That is, A1 can be equal to B1 or not, A2 can be equal to B2 or not, …, and Am can be equal to Bm or not.
[0076] Suppose the sweep range of the tunable laser is 1548 nm-1552 nm, the spectral length of the global measurement spectrum is 4 nm, the intercept lengths used for intercepting the global measurement spectrum are 2 nm, 2.5 nm and 3 nm respectively, the center interval between the two adjacent local measurement spectra is 0.2 nm (the overlapping ranges are 90%, 92% and 93% respectively), and finally three groups of local measurement data are obtained. Among them, the intercept length of the first group of local measurement data is 2 nm, the number of spectra is 11, and the wavelength ranges are 1548.0 nm-1550.0 nm, 1548.2 nm-1550.2 nm, 1548.4 nm-1550.4 nm, 1548.6 nm-1550.6 nm, 1548.8 nm-1550.8 nm, 1549.0 nm-1551.0 nm, 1549.2 nm-1551.2 nm, 1549.4 nm-1551.4 nm, 1549.6 nm-1551.6 nm, 1549.8 nm-1551.8 nm and 1550.0 nm-1552.0 nm; the intercept length of the second group of local measurement data is 2.5 nm, the number of spectra is 8, and the wavelength ranges are 1548.0 nm-1550.5 nm, 1548.2 nm-1550.7 nm, 1548.4 nm-1550.9 nm, 1548.6 nm-1551.1 nm, 1548.8 nm-1551.3 nm, 1549.0 nm-1551.5 nm, 1549.2 nm-1551.7 nm and 1549.4 nm-1551.9 nm; the intercept length of the third group of local measurement data is 3 nm, the number of spectra is 6, and the wavelength ranges are 1548.0 nm-1551.0 nm, 1548.2 nm-1551.2 nm, 1548.4 nm-1551.4 nm, 1548.6 nm-1551.6 nm, 1548.8 nm-1551.8 nm and 1549.0 nm-1552.0 nm.
[0077] When truncating the global measurement spectrum, the starting position of the truncation is preferably on one side of the global measurement spectrum, and the ending position is preferably on the other side of the global measurement spectrum. That is, a truncating frame is set at the minimum wavelength position on the left side of the global measurement spectrum according to the truncating length, and then the truncating frame is moved step by step to the maximum wavelength position on the right side. Alternatively, the reverse is true: a truncating frame is set at the maximum wavelength position on the right side of the global measurement spectrum according to the truncating length, and then the truncating frame is moved step by step to the minimum wavelength position on the left side. The amount of each step is the center-to-center distance between two adjacent local measurement spectra.
[0078] Step 500: Based on each local reference spectrum and each local measurement spectrum with the same cut length, calculate the offset error value of each local reference spectrum, and select the local reference spectrum with the smallest offset error value, so as to use the optimal offset of the local reference spectrum as the wavelength offset between the reference signal and the measurement signal.
[0079] In step 500, the offset error value is used to perform self-matching on the local reference spectrum and the local measurement spectrum with the same truncation length, thereby achieving demodulation of temperature / strain sensing with a large measurement range and high spatial resolution under small sweep frequency range conditions.
[0080] Specifically, such as Figure 4 As shown, in step 500, the offset error value of each local reference spectrum is calculated based on each local reference spectrum and each local measurement spectrum with the same truncation length, and the local reference spectrum with the smallest offset error value is selected. The optimal offset of this local reference spectrum is then used as the wavelength offset between the reference signal and the measurement signal. The steps are as follows:
[0081] Step 510: Calculate the spectral similarity between each local reference spectrum and each local measurement spectrum with the same truncation length.
[0082] In step 510, the calculation of the spectral similarity needs to be performed between a local reference spectrum and a local measurement spectrum with the same truncation length.
[0083] For example, A1 local reference spectra in the first group of local reference data are respectively calculated with B1 local measurement spectra in the first group of local measurement data to obtain A1*B1 spectral similarities; A2 local reference spectra in the second group of local reference data are respectively calculated with B2 local measurement spectra in the second group of local measurement data to obtain A2*B2 spectral similarities; and so on; and Am local reference spectra in the Mth group of local reference data are respectively calculated with Bm local measurement spectra in the Mth group of local measurement data to obtain Am*Bm spectral similarities; a total of [A1*B1+A2*B2+…+Am*Bm] spectral similarities are obtained.
[0084] The method for calculating the spectral similarity can be, but is not limited to, a cosine similarity method, a spectral feature comparison method, a neural network method, or the like. In this embodiment, the cosine similarity method is used.
[0085] Step 520: individually comparing the spectral similarities of each local reference spectrum, and respectively matching the local measurement spectrum with the maximum spectral similarity for each local reference spectrum.
[0086] In step 520, the similarity comparison needs to be performed between the spectral similarities of the same local reference spectrum.
[0087] For example, for A1 local reference spectra in the first group of local reference data, each local reference spectrum corresponds to B1 spectral similarities, which are consistent with the number of spectra in the first group of local measurement data. By comparing the B1 spectral similarities, the local measurement spectrum with the maximum spectral similarity is matched for each local reference spectrum in the first group of local reference data in the first group of local measurement data.
[0088] For A2 local reference spectra in the second group of local reference data, each local reference spectrum corresponds to B2 spectral similarities, which are consistent with the number of spectra in the second group of local measurement data. By comparing the B2 spectral similarities, the local measurement spectrum with the maximum spectral similarity is matched for each local reference spectrum in the second group of local reference data in the second group of local measurement data.
[0089] The same is true for other groups of local reference data. By repeating the above calculation, the local measurement spectrum with the maximum spectral similarity is matched for each local reference spectrum in the Mth group of local reference data in the Mth group of local measurement data.
[0090] Step 530: calculating the wavelength shift between each local reference spectrum and each local measurement spectrum with the same length as the local reference spectrum.
[0091] In this step 530, the wavelength shift needs to be calculated between the local reference spectrum and the local measurement spectrum with the same length of interception.
[0092] For example, A1 local reference spectra in the first group of local reference data are respectively matched with B1 local measurement spectra in the first group of local measurement data to calculate the wavelength shift, obtaining A1*B1 wavelength shifts; A2 local reference spectra in the second group of local reference data are respectively matched with B2 local measurement spectra in the second group of local measurement data to calculate the wavelength shift, obtaining A2*B2 wavelength shifts; …; Am local reference spectra in the Mth group of local reference data are respectively matched with Bm local measurement spectra in the Mth group of local measurement data to calculate the wavelength shift, obtaining Am*Bm wavelength shifts; a total of [A1*B1+A2*B2+…+Am*Bm] wavelength shifts are obtained.
[0093] The method for calculating the wavelength shift can be, but is not limited to, a wavelength peak searching algorithm, a cross-correlation algorithm or a spectral fitting algorithm. In this embodiment, the wavelength peak searching algorithm is used.
[0094] Step 540: The wavelength shift between each local reference spectrum and the matched local measurement spectrum is taken as the best shift of each local reference spectrum.
[0095] For example, in the first group of local reference data, the wavelength shift corresponding to the first local reference spectrum and the local measurement spectrum matched in step 520 in the calculation result of step 530 is determined, and the wavelength shift is taken as the best shift of the first local reference spectrum; the wavelength shift corresponding to the second local reference spectrum and the local measurement spectrum matched in step 520 in the calculation result of step 530 is determined, and the wavelength shift is taken as the best shift of the second local reference spectrum; …; the wavelength shift corresponding to the A1th local reference spectrum and the local measurement spectrum matched in step 520 in the calculation result of step 530 is determined, and the wavelength shift is taken as the best shift of the A1th local reference spectrum.
[0096] In the second group of local reference data, according to the local measurement spectrum matched by the first local reference spectrum in step 520, the wavelength offset corresponding to both is determined in the calculation result of step 530, and the wavelength offset is taken as the best offset of the first local reference spectrum, according to the local measurement spectrum matched by the second local reference spectrum in step 520, the wavelength offset corresponding to both is determined in the calculation result of step 530, and the wavelength offset is taken as the best offset of the second local reference spectrum, …, according to the local measurement spectrum matched by the A1th local reference spectrum in step 520, the wavelength offset corresponding to both is determined in the calculation result of step 530, and the wavelength offset is taken as the best offset of the A1th local reference spectrum.
[0097] The same is true for other groups of local reference data. By repeating the above process, the best offset of each local reference spectrum in the Mth group of local reference data is obtained.
[0098] Step 550: Calculate the offset error value between the best offset of each local reference spectrum and the wavelength offset of each local measurement spectrum with the same length.
[0099] In this step 550, the calculation of the offset error value needs to be performed between the local reference spectrum and the local measurement spectrum with the same length.
[0100] For example, in the first group of local reference data, according to the best offset of the first local reference spectrum, the wavelength offset of each local measurement spectrum in the first group of measurement reference data is calculated to obtain the offset error value RMSE(1_1), according to the best offset of the second local reference spectrum, the wavelength offset of each local measurement spectrum in the first group of measurement reference data is calculated to obtain the offset error value RMSE(1_2), …, according to the best offset of the A1th local reference spectrum, the wavelength offset of each local measurement spectrum in the first group of measurement reference data is calculated to obtain the offset error value RMSE(1_A1).
[0101] In the second group of local reference data, according to the best offset of the first local reference spectrum, the wavelength offset of each local measurement spectrum in the second group of measurement reference data is calculated to obtain the offset error value RMSE(2_1), according to the best offset of the second local reference spectrum, the wavelength offset of each local measurement spectrum in the second group of measurement reference data is calculated to obtain the offset error value RMSE(2_2), …, according to the best offset of the A2th local reference spectrum, the wavelength offset of each local measurement spectrum in the second group of measurement reference data is calculated to obtain the offset error value RMSE(2_A2).
[0102] The same is true for other groups of local reference data. By repeating the above calculation, the offset error values RMSE(M_1), RMSE(M_2), …, RMSE(M_Am) of each local reference spectrum in the Mth group of local reference data are obtained.
[0103] In this embodiment, the root mean square error (RMSE) is used as the offset error value to evaluate the wavelength offset difference between each local reference spectrum of the same length. The formula is as follows:
[0104]
[0105] The offset error value of the i-th local reference spectrum in the m-th (1≤m≤M) group of local reference data is The optimal offset of the i-th local reference spectrum in the m-th group of local reference data is The average wavelength offset of each local measurement spectrum in the m-th group of local measurement data is n, which is the number of spectra in the m-th group of local reference data, i.e. n=A1, A2, …, Am.
[0106] Of course, in addition to the root mean square error (RMSE), the mean absolute error (MAE), the mean absolute percentage error (MAPE), or the median absolute error (MedAE) can also be used as the offset error value to evaluate the wavelength offset difference between each local reference spectrum of the same length.
[0107] Step 560: Compare each offset error value, select the local reference spectrum with the smallest offset error value, and take the optimal offset of the local reference spectrum as the wavelength offset between the reference signal and the measurement signal.
[0108] In this step 560, the error value comparison needs to be performed not only between each local reference spectrum of the same length, but also between each local reference spectrum of different lengths.
[0109] The optical frequency domain reflection demodulation method of the present application sets different lengths to intercept the global reference spectrum, thereby obtaining multiple local reference spectra with different wavelength ranges, and selects the local reference spectrum with the optimal wavelength range according to the offset error value of each local reference spectrum, and takes the optimal offset of the local reference spectrum as the wavelength offset between the reference signal and the measurement signal, thereby finally realizing the demodulation of the temperature / strain sensor with large measurement range and high spatial resolution under the condition of small sweep frequency range.
[0110] Figure 5The strain curve diagrams obtained by using the conventional cross-correlation algorithm (a) and the demodulation method of the present application (b) to demodulate the same set of reference signals and measurement signals with 2.5 nm spatial resolution in the 8 nm sweep range are shown respectively. As can be seen from the diagrams, as the strain value increases, the conventional cross-correlation algorithm is difficult to accurately demodulate the axial strain distribution of the sensing optical fiber, resulting in more false points in the strain curve diagram; while the demodulation method of the present application can correctly recover the axial strain distribution of the sensing optical fiber, and the false points are significantly reduced, and the demodulation effect is significantly improved.
[0111] Embodiment Three
[0112] An optical frequency domain reflectometry demodulation device based on an adaptive local spectral matching method, comprising a processor and a memory, the memory stores a computer program for the processor to execute, and the processor, when executing the computer program, performs the optical frequency domain reflectometry demodulation method of embodiment two.
[0113] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the embodiments of the present application rather than limit them, and although the embodiments of the present application have been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the embodiments of the present application can still be modified or replaced equivalently, and these modifications or equivalent replacements should not make the modified technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.
Claims
1. An optical frequency domain reflectometry demodulation method based on adaptive local spectral matching method, characterized in that, The method comprises the following steps: Step 100: obtaining a reference signal and a measurement signal; Step 200: converting the reference signal and the measurement signal into the wavelength domain respectively to obtain a global reference spectrum and a global measurement spectrum; Step 300: using different intercept lengths to intercept the global reference spectrum respectively to obtain a plurality of local reference spectra with different wavelength ranges; Step 400: using the same plurality of intercept lengths as the global reference spectrum to intercept the global measurement spectrum respectively to obtain a plurality of local measurement spectra with different wavelength ranges; Step 500: calculating the offset error value of each local reference spectrum according to each local reference spectrum and each local measurement spectrum with the same intercept length, and selecting the local reference spectrum with the minimum offset error value, so as to take the optimal offset amount of the local reference spectrum as the wavelength offset amount between the reference signal and the measurement signal; In step 500, the offset error value of each local reference spectrum is calculated according to each local reference spectrum and each local measurement spectrum with the same intercept length, and the local reference spectrum with the minimum offset error value is selected, so as to take the optimal offset amount of the local reference spectrum as the wavelength offset amount between the reference signal and the measurement signal. The steps are as follows: Step 510: calculating the spectral similarity between each local reference spectrum and each local measurement spectrum with the same intercept length; Step 520: individually comparing each spectral similarity of each local reference spectrum to match the local measurement spectrum with the maximum spectral similarity for each local reference spectrum respectively; Step 530: calculating the wavelength offset amount between each local reference spectrum and each local measurement spectrum with the same intercept length; Step 540: taking the wavelength offset amount between each local reference spectrum and the matched local measurement spectrum as the optimal offset amount of each local reference spectrum respectively; Step 550: calculating the offset error value between the optimal offset amount of each local reference spectrum and the wavelength offset amount of each local measurement spectrum with the same intercept length; Step 560: comparing each offset error value, selecting the local reference spectrum with the minimum offset error value, and taking the optimal offset amount of the local reference spectrum as the wavelength offset amount between the reference signal and the measurement signal.
2. The optical frequency domain reflectometry method of claim 1, wherein, In step 200, the reference signal and the measurement signal are converted into the wavelength domain respectively to obtain the global reference spectrum and the global measurement spectrum. The steps are as follows: Step 210: converting the reference signal and the measurement signal into the distance domain respectively by fast Fourier transform to obtain a reference distance domain signal and a measurement distance domain signal; Step 220: performing windowing and zero padding on the reference distance domain signal and the measurement distance domain signal respectively; Step 230: converting the reference distance domain signal and the measurement distance domain signal after windowing and zero padding into the wavelength domain by inverse fast Fourier transform to obtain the global reference spectrum and the global measurement spectrum.
3. The optical frequency domain reflectometry demodulation method of claim 1, wherein, In step 300, when the global reference spectrum is intercepted, the starting position of the interception is the center position of the global reference spectrum, and the ending position of the interception is the position on the two sides of the global reference spectrum.
4. The optical frequency domain reflectometry demodulation method of claim 1, wherein, In each local reference spectrum with the same interception length, the center position of one of the local reference spectra is the same as that of the global reference spectrum.
5. The optical frequency domain reflectometry demodulation method of claim 1, wherein, In each local reference spectrum with the same interception length, the two adjacent local reference spectra are not overlapped, or are partially overlapped, and the overlapping range is between 25% and 95%.
6. The optical frequency domain reflectometry method of claim 1, wherein, In step 400, when the global measurement spectrum is intercepted, the starting position of the interception is the position on one side of the global measurement spectrum, and the ending position of the interception is the position on the other side of the global measurement spectrum.
7. The optical frequency domain reflectometry method of claim 1, wherein, In each local measurement spectrum with the same interception length, the two adjacent local measurement spectra are partially overlapped, and the overlapping range is between 25% and 95%.
8. An optical frequency domain reflectometry demodulation device based on adaptive local spectral matching method, comprising a processor and a memory, wherein the memory stores a computer program for the processor to execute, characterized in that, The processor, when executing the computer program, performs the optical frequency domain reflectometry method of claim 1.
9. An optical frequency domain reflectometry sensing system based on adaptive local spectral matching method, characterized in that, The optical frequency domain reflectometry device of claim 8, and a data acquisition device, an optical frequency domain reflectometer, a first-in-first-out module, and a sensing optical fiber, wherein the optical frequency domain reflectometer is connected to the sensing optical fiber through the first-in-first-out module, the data acquisition device is connected to the optical frequency domain reflectometer, and the data acquisition device is connected to the optical frequency domain reflectometry device.