Dynamic strain sensing signal demodulation method, system and device of OFDR system, medium and product

Abstract

The invention discloses a dynamic strain sensing signal demodulation method, system and device of an OFDR system, a medium and a product, and relates to the technical field of signal demodulation, and the method comprises the steps: obtaining a reference Rayleigh scattering signal intensity array and a sensing Rayleigh scattering signal intensity array through the OFDR system; performing cross-correlation demodulation on the reference Rayleigh scattering signal intensity array and the sensing Rayleigh scattering signal intensity array according to preset demodulation parameters to obtain under-sampling demodulation two-dimensional strain distribution of the sensing optical fiber; inputting the under-sampling demodulated two-dimensional strain distribution into a two-dimensional strain distribution reduction model to obtain reduced two-dimensional strain distribution of the sensing optical fiber; and performing two-dimensional deconvolution calculation on the original two-dimensional strain distribution to obtain the demodulated two-dimensional strain distribution of the sensing optical fiber. According to the invention, on the basis of not changing the hardware of the OFDR system, the high-speed demodulation of the dynamic strain sensing signal is realized.

Description

Technical Field

[0001] The present application relates to the field of signal demodulation technology, and in particular to a method, system, device, medium and product for demodulating dynamic strain sensing signals of an OFDR system. Background Art

[0002] With the increasing demand for sensing measurements in engineering applications, distributed fiber optic sensing technology is increasingly required to detect dynamic disturbances along the sensing fiber. Furthermore, in the design and manufacture of certain precision equipment structures, it is not only necessary to accurately measure detailed vibration information, including amplitude, frequency, and phase, but also to achieve a spatial resolution of centimeters or even millimeters, enabling real-time monitoring of the deformation state of key parts and providing feedback for further optimization of design and manufacturing solutions. Therefore, Optical Frequency Domain Reflectometry (OFDR), a distributed fiber optic sensing technology, has been widely used in various precision monitoring scenarios due to its high spatial resolution (millimeter level). OFDR measures the Rayleigh scattering signal in the optical fiber and demodulates the changes in the signal before and after sensing to measure the change in external factors, obtain the location information of the change, and the specific spatial length of the change, thereby realizing dynamic strain sensing.

[0003] Traditional OFDR-based dynamic strain sensing technology improves OFDR dynamic strain sensing performance by optimizing system hardware, rather than optimizing software-based demodulation algorithms. This technology fails to fundamentally overcome the mutual limitations between spatiotemporal resolution and measurement accuracy in time-resolved schemes. Furthermore, OFDR-based time-resolved schemes require distributed cross-correlation demodulation of multiple spectral data segments, exponentially increasing the amount of data computation required and making it difficult to rapidly demodulate dynamic signals. To increase data processing speed, hardware such as GPUs and FPGAs are typically used to accelerate static measurement schemes, rather than utilizing software-based demodulation algorithms to achieve high-speed demodulation of dynamic strain sensing signals.

[0004] In summary, it can be seen that the traditional high-speed demodulation technology of dynamic strain sensing signals requires hardware changes to the OFDR system. Summary of the Invention

[0005] The purpose of this application is to provide a dynamic strain sensing signal demodulation method, system, device, medium and product for an OFDR system, so as to achieve high-speed demodulation of dynamic strain sensing signals without changing the hardware of the OFDR system.

[0006] To achieve the above objectives, this application provides the following solutions:

[0007] In a first aspect, the present application provides a method for demodulating dynamic strain sensing signals of an OFDR system, comprising:

[0008] Obtaining a reference Rayleigh scattering signal intensity array and a sensing Rayleigh scattering signal intensity array using an OFDR system;

[0009] According to the preset demodulation parameters, the reference Rayleigh scattering signal intensity array and the sensing Rayleigh scattering signal intensity array are cross-correlatedly demodulated to obtain the under-sampling demodulated two-dimensional strain distribution of the sensing optical fiber;

[0010] Inputting the undersampled demodulated two-dimensional strain distribution into a two-dimensional strain distribution restoration model to obtain a restored two-dimensional strain distribution of the sensing optical fiber; the two-dimensional strain distribution restoration model is obtained by training a compressed sensing deep neural network;

[0011] A two-dimensional deconvolution calculation is performed on the restored two-dimensional strain distribution using the sliding average matrix corresponding to the preset demodulation parameters to obtain the demodulated two-dimensional strain distribution of the sensing optical fiber.

[0012] In one embodiment, obtaining a reference Rayleigh scattering signal intensity array and a sensing Rayleigh scattering signal intensity array using an OFDR system includes:

[0013] Before the sensing fiber is strained, the OFDR system injects a frequency-linearly tuned laser into the sensing fiber and simultaneously collects Rayleigh backscattered signals at propagation constants corresponding to different laser frequencies, obtaining Rayleigh backscattered frequency domain signals at multiple propagation constants before the sensing fiber is strained.

[0014] Based on the intensities of the backward Rayleigh scattering radio frequency signals under all propagation constants before the strain of the sensing optical fiber, a reference Rayleigh scattering signal intensity array is constructed;

[0015] After the sensing fiber is strained, the OFDR system injects a frequency-linearly tuned laser into the sensing fiber and simultaneously collects backscattered Rayleigh signals at propagation constants corresponding to different laser frequencies, obtaining backscattered Rayleigh frequency domain signals at multiple propagation constants after the strain of the sensing fiber.

[0016] Based on the intensity of the back Rayleigh scattering frequency domain signal under all propagation constants after the strain of the sensing optical fiber, a sensing Rayleigh scattering signal intensity array is constructed.

[0017] In one embodiment, the preset demodulation parameters include: preset demodulation parameters, including: preset frequency domain data point step, preset frequency domain data point window length, preset time step, preset time window length, preset distance domain data point step, preset distance domain data point window length, preset distance step and preset distance window length; the time step corresponding to the preset frequency domain data point step is the preset time step, the time step corresponding to the preset frequency domain data point window length is the preset time window length, the distance step corresponding to the preset distance domain data point step is the preset distance step, and the distance step corresponding to the preset distance domain data point window length is the preset distance window length.

[0018] In one embodiment, cross-correlation demodulation is performed on the reference Rayleigh scattering signal intensity array and the sensing Rayleigh scattering signal intensity array according to preset demodulation parameters to obtain an undersampled demodulated two-dimensional strain distribution of the sensing optical fiber, including:

[0019] Dividing the reference Rayleigh scattering signal intensity array according to the preset frequency domain data point step size and the preset frequency domain data point window length to obtain a plurality of reference sub-segment spectrum signals; the reference sub-segment spectrum signals are arranged according to the preset time step size and the preset time window length;

[0020] Performing discrete Fourier transform on each reference sub-segment spectrum signal to obtain multiple reference Rayleigh scattering range domain signals;

[0021] sampling each reference Rayleigh scattering distance domain signal according to the preset distance domain data point step size and the preset distance domain data point window length, to obtain a plurality of reference Rayleigh scattering signals at random times and random positions, thereby forming an undersampled reference Rayleigh scattering signal; and arranging the reference Rayleigh scattering signals according to the preset distance step size and the preset distance window length;

[0022] The sensor Rayleigh scattering signal intensity array is divided according to the preset frequency domain data point step size and the preset frequency domain data point window length to obtain a plurality of sensor sub-segment spectrum signals; the spectrum signals of each sensor sub-segment are arranged according to the preset time step size and the preset time window length;

[0023] Performing discrete Fourier transform on the spectrum signal of each sensor segment respectively to obtain multiple sensor CorelDRAW range domain signals;

[0024] sampling each sensor Rayleigh scattering distance domain signal according to the preset distance domain data point step size and the preset distance domain data point window length, to obtain a plurality of sensor Rayleigh scattering signals at random times and random positions, thereby forming an undersampled sensor Rayleigh scattering signal; and arranging each sensor Rayleigh scattering signal according to the preset distance step size and the preset distance window length;

[0025] The undersampled reference Rayleigh scattering signal and the undersampled sensing Rayleigh scattering signal are strain demodulated by using a cross-correlation algorithm to obtain an undersampled demodulated two-dimensional strain distribution of the sensing optical fiber; the undersampled demodulated two-dimensional strain distribution is a demodulated two-dimensional strain distribution of the sensing optical fiber in a random distribution with a preset time step and a preset time window length and a demodulation window with a preset distance step and a preset distance window length.

[0026] In one embodiment, the process of determining the two-dimensional strain distribution restoration model includes:

[0027] Initializing the compressed sensing deep neural network; the compressed sensing deep neural network includes multiple reconstruction layers, multiple auxiliary variable update blocks, and multiple multiplier update layers; wherein the current reconstruction layer is connected to the current auxiliary variable update block and the current multiplier update layer, respectively, the current auxiliary variable update block is connected to the current multiplier update layer, and the current auxiliary variable update block and the current multiplier update layer are both connected to the next reconstruction layer;

[0028] Acquire a training set; the training set includes: a plurality of samples of two-dimensional strain distributions to be restored and corresponding two-dimensional strain distributions after restoration;

[0029] The training set is used to train a compressed sensing deep neural network to obtain the two-dimensional strain distribution restoration model.

[0030] In one embodiment, a two-dimensional deconvolution calculation is performed on the restored two-dimensional strain distribution using a sliding average matrix corresponding to the preset demodulation parameters to obtain a demodulated two-dimensional strain distribution of the sensing optical fiber, including:

[0031] Determining the sliding average matrix based on a time ratio and a distance ratio; the time ratio is a ratio of the preset time window length to the preset time step length, and the distance ratio is a ratio of the preset distance window length to the preset distance step length;

[0032] The sliding average matrix is used to perform a two-dimensional deconvolution calculation on the restored two-dimensional strain distribution to obtain a demodulated two-dimensional strain distribution of the sensing optical fiber.

[0033] In a third aspect, the present application provides a dynamic strain sensing signal demodulation system for an OFDR system to implement any of the above-mentioned dynamic strain sensing signal demodulation methods for an OFDR system, wherein the dynamic strain sensing signal demodulation system for the OFDR system comprises:

[0034] an array acquisition module, for acquiring a reference Rayleigh scattering signal intensity array and a sensing Rayleigh scattering signal intensity array using an OFDR system;

[0035] A cross-correlation demodulation module is used to perform cross-correlation demodulation on the reference Rayleigh scattering signal intensity array and the sensing Rayleigh scattering signal intensity array according to preset demodulation parameters to obtain an undersampled demodulated two-dimensional strain distribution of the sensing optical fiber;

[0036] a restoration module, configured to input the undersampled demodulated two-dimensional strain distribution into a two-dimensional strain distribution restoration model to obtain a restored two-dimensional strain distribution of the sensing optical fiber; the two-dimensional strain distribution restoration model is obtained by training a compressed sensing deep neural network;

[0037] The deconvolution module is used to perform a two-dimensional deconvolution calculation on the restored two-dimensional strain distribution using the sliding average matrix corresponding to the preset demodulation parameters to obtain the demodulated two-dimensional strain distribution of the sensing optical fiber.

[0038] In a second aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the dynamic strain sensing signal demodulation method of the OFDR system described in any one of the above items.

[0039] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the dynamic strain sensing signal demodulation method of the OFDR system described in any one of the above items.

[0040] In a fifth aspect, the present application provides a computer program product, comprising a computer program, which, when executed by a processor, implements the dynamic strain sensing signal demodulation method of the OFDR system described in any one of the above items.

[0041] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0042] This application discloses a dynamic strain sensing signal demodulation method, system, device, medium, and product for an OFDR system. First, an OFDR system is used to acquire a reference Rayleigh scattering signal intensity array and a sensing Rayleigh scattering signal intensity array. Then, the reference Rayleigh scattering signal intensity array and the sensing Rayleigh scattering signal intensity array are cross-correlated and demodulated according to preset demodulation parameters to obtain an undersampled demodulated two-dimensional strain distribution of the sensing fiber. Subsequently, the undersampled demodulated two-dimensional strain distribution is input into a two-dimensional strain distribution restoration model to obtain a restored two-dimensional strain distribution of the sensing fiber. The two-dimensional strain distribution restoration model is obtained by training a compressed sensing deep neural network. Finally, a two-dimensional deconvolution calculation is performed on the restored two-dimensional strain distribution using a sliding average matrix corresponding to the preset demodulation parameters to obtain a demodulated two-dimensional strain distribution of the sensing fiber. The method of the present application sequentially processes the acquired reference Rayleigh scattering signal intensity array and the sensing Rayleigh scattering signal intensity array to obtain a demodulated two-dimensional strain distribution of the sensing fiber. High-speed demodulation of dynamic strain sensing signals is achieved without changing the hardware of the OFDR system. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0044] Figure 1 This is a diagram showing an application environment of a dynamic strain sensing signal demodulation method of an OFDR system in one embodiment of the present application;

[0045] Figure 2 A schematic flow chart of a dynamic strain sensing signal demodulation method for an OFDR system according to an embodiment of the present application;

[0046] Figure 3 Schematic diagram of the cross-correlation demodulation process;

[0047] Figure 4 Schematic diagram of the compressed sensing deep neural network structure;

[0048] Figure 5 Schematic diagram of the compressed sensing deep neural network restoration results;

[0049] Figure 6 Schematic diagram of two-dimensional deconvolution calculation results;

[0050] Figure 7 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0052] The purpose of this application is to provide a dynamic strain sensing signal demodulation method, system, device, medium and product for an OFDR system, aiming to achieve high-speed demodulation of dynamic strain sensing signals without changing the hardware of the OFDR system.

[0053] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0054] The dynamic strain sensing signal demodulation method of the OFDR system provided in the embodiment of the present application can be applied to Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be set up separately, integrated on the server 104, or placed on the cloud or other servers. The terminal 102 can send the reference Rayleigh scattering signal intensity array and the sensing Rayleigh scattering signal intensity array to the server 104. After the server 104 receives the reference Rayleigh scattering signal intensity array and the sensing Rayleigh scattering signal intensity array, the server 104 performs cross-correlation demodulation on the reference Rayleigh scattering signal intensity array and the sensing Rayleigh scattering signal intensity array according to the preset demodulation parameters to obtain the under-sampled demodulated two-dimensional strain distribution of the sensing optical fiber; the under-sampled demodulated two-dimensional strain distribution is input into the two-dimensional strain distribution restoration model to obtain the restored two-dimensional strain distribution of the sensing optical fiber; the sliding average matrix corresponding to the preset demodulation parameters is used to perform two-dimensional deconvolution calculation on the restored two-dimensional strain distribution to obtain the demodulated two-dimensional strain distribution of the sensing optical fiber. The server 104 can feed back the obtained demodulated two-dimensional strain distribution to the terminal 102. In addition, in some embodiments, the dynamic strain sensing signal demodulation method of the OFDR system can also be implemented independently by the server 104 or the terminal 102. For example, the terminal 102 can directly perform dynamic strain sensing signal demodulation of the OFDR system on the reference Rayleigh scattering signal intensity array and the sensing Rayleigh scattering signal intensity array. Alternatively, the server 104 can obtain the reference Rayleigh scattering signal intensity array and the sensing Rayleigh scattering signal intensity array from a data storage system and perform dynamic strain sensing signal demodulation of the OFDR system on the reference Rayleigh scattering signal intensity array and the sensing Rayleigh scattering signal intensity array.

[0055] In an exemplary embodiment, Figure 2 As shown, a dynamic strain sensing signal demodulation method of an OFDR system is provided, comprising:

[0056] Step 1: Use the OFDR system to obtain the reference Rayleigh scattering signal intensity array and the sensing Rayleigh scattering signal intensity array.

[0057] As an optional implementation, step 1 includes:

[0058] Step 11: Before the sensing fiber is strained, the OFDR system injects a frequency-linearly tuned laser into the sensing fiber and simultaneously collects backscattered Rayleigh signals at propagation constants corresponding to different laser frequencies to obtain backscattered Rayleigh frequency domain signals at multiple propagation constants before the sensing fiber is strained.

[0059] Step 12: Based on the intensities of the backward Rayleigh scattering signals in the frequency domain at all propagation constants before the strain of the sensing optical fiber, a reference Rayleigh scattering signal intensity array is constituted.

[0060] Step 13: After the sensing optical fiber undergoes strain, the OFDR system injects a laser with linearly tuned frequency into the sensing optical fiber, and simultaneously acquires the backward Rayleigh scattering signals at the propagation constants corresponding to different laser frequencies, obtaining the backward Rayleigh scattering signals in the frequency domain at multiple propagation constants after the strain of the sensing optical fiber.

[0061] Step 14: Based on the intensities of the backward Rayleigh scattering signals in the frequency domain at all propagation constants after the strain of the sensing optical fiber, a sensing Rayleigh scattering signal intensity array is constituted.

[0062] Specifically, before and after the sensing optical fiber undergoes strain, the computer controls the OFDR system to inject a laser with linearly tuned frequency into the sensing optical fiber, and simultaneously triggers the detector to acquire, detecting the backward Rayleigh scattering signals at the propagation constants corresponding to different laser frequencies, that is, the backward Rayleigh scattering signals in the frequency domain at multiple propagation constants of the sensing optical fiber:

[0063]

[0064] Among them, E(β) is the backward Rayleigh scattering signal in the frequency domain at the propagation constant β; E0 is the electric field strength of the incident laser. Since the scattered light is very weak and its influence on the electric field of the incident laser can be ignored, it is assumed that the electric field of the incident laser remains unchanged; β is the propagation constant of the laser, which is proportional to the laser frequency ω; z is the spatial distance; j is the imaginary unit; L1 is the length of the sensing optical fiber; κ(z) is the scattering cross-section of the sensing optical fiber at the spatial distance z along the length direction.

[0065] When actually measuring with the OFDR system, the light intensity is measured by coherent detection, and the amplitude and phase information of the backward Rayleigh scattering signal E(β) are retained. Since the acquisition and processing of data are discrete, next, for the backward Rayleigh scattering signals in the frequency domain at all propagation constants before the strain of the sensing optical fiber, the reference Rayleigh scattering signal intensity array I = {I n = I(nΔβ), 1 ≤ n < N} is used to represent, I n is the intensity of the backward Rayleigh scattering signal in the frequency domain at the nth propagation constant in the reference Rayleigh scattering signal intensity array, N is the total number of propagation constants, and Δβ is the difference between adjacent propagation constants. For the backward Rayleigh scattering signals in the frequency domain at all propagation constants after the strain of the sensing optical fiber, the sensing Rayleigh scattering signal intensity array I' = {I' n = I(nΔβ), 1 ≤ n < N} is used to represent, I' nis the intensity of the backward Rayleigh scattered radio frequency signal under the nth propagation constant in the sensing Rayleigh scattering signal intensity array.

[0066] Step 2: According to the preset demodulation parameters, the reference Rayleigh scattering signal intensity array and the sensing Rayleigh scattering signal intensity array are cross-correlatedly demodulated to obtain the under-sampling demodulated two-dimensional strain distribution of the sensing optical fiber.

[0067] In this application, a sliding large spatiotemporal window is used for sampling and demodulation.

[0068] As an optional implementation, the preset demodulation parameters include: a preset frequency domain data point step, a preset frequency domain data point window length, a preset time step, a preset time window length, a preset distance domain data point step, a preset distance domain data point window length, a preset distance step and a preset distance window length; the time step corresponding to the preset frequency domain data point step is the preset time step, the time step corresponding to the preset frequency domain data point window length is the preset time window length, the distance step corresponding to the preset distance domain data point step is the preset distance step, and the distance step corresponding to the preset distance domain data point window length is the preset distance window length.

[0069] As an optional implementation, Figure 3 As shown, step 2 includes:

[0070] Step 21: Divide the reference Rayleigh scattering signal intensity array according to a preset frequency domain data point step size and a preset frequency domain data point window length to obtain a plurality of reference sub-segment spectrum signals; each reference sub-segment spectrum signal is arranged according to a preset time step size and a preset time window length.

[0071] Specifically, the reference Rayleigh scattering signal intensity array I is divided according to the preset frequency domain data point step size Δm and the preset frequency domain data point window length M, and the obtained Reference sub-segment spectrum signal [I1,I M ],[I 1+Δm ,I M+Δm ],···,[I N-M+1 ,I N ]. Wherein, M is an integer multiple of Δm, the time step corresponding to Δm frequency domain data points is the preset time step Δt, and the time step corresponding to M frequency domain data points is the preset time window length T.

[0072] Step 22: Perform discrete Fourier transform on each reference sub-segment spectrum signal to obtain multiple reference Rayleigh scattering range domain signals.

[0073] Specifically, for any reference sub-segment spectrum signal I m Performing discrete Fourier transform, the corresponding expressions of multiple reference Rayleigh scattering distance domain signals are obtained as follows:

[0074]

[0075] Among them, is the k-th reference Rayleigh scattering distance domain signal, where 1 ≤ k < K and k is an integer.

[0076] The spatial distance between any two adjacent reference Rayleigh scattering distance domain signals is:

[0077]

[0078] where λ is the central wavelength of the laser; n g is the group refractive index of the sensing optical fiber; Δλ is the wavelength range of laser scanning. It can be seen that the spatial resolution of the data and the wavelength range of laser scanning (i.e., the time resolution ability) limit each other.

[0079] Step 23: Sample each reference Rayleigh scattering distance domain signal according to the preset distance domain data point step size and the preset distance domain data point window length, to obtain multiple reference Rayleigh scattering signals at random times and random positions, thereby constituting an undersampled reference Rayleigh scattering signal; each reference Rayleigh scattering signal is arranged according to the preset distance step size and the preset distance window length.

[0080] Specifically, for the reference Rayleigh scattering distance domain signal take a part of the data subset at a random position with Δl as the preset distance step size and L as the preset distance window length, to obtain an undersampled reference Rayleigh scattering signal

[0081] For example, randomly take 20% of the position reference Rayleigh scattering distance domain signals: [[ID=J36]]where l is an integer within the range 1 ≤ l < L, L is an integer multiple of Δl, the distance step size corresponding to Δl distance domain data points is the preset distance step size Δz, and the distance step size corresponding to L distance domain data points is the preset distance window length Z.

[0082] Step 24: Divide the sensing Rayleigh scattering signal intensity array according to the preset frequency domain data point step size and the preset frequency domain data point window length, to obtain multiple sensing sub-segment spectrum signals; each sensing sub-segment spectrum signal is arranged according to the preset time step size and the preset time window length.

[0083] Specifically, divide the sensing Rayleigh scattering signal intensity array I according to the preset frequency domain data point step size Δm and the preset frequency domain data point window length M, to obtain sensing sub-segment spectrum signals [I'1, I' M ,[I'1+Δm ,I' M+Δm ],···,[I' N-M+1 ,I' N ].

[0084] Step 25: Perform discrete Fourier transform on the spectrum signal of each sensor segment to obtain multiple sensor CorelDRAW range domain signals.

[0085] Specifically, for any sensor segment spectrum signal I' m Performing discrete Fourier transform, the corresponding expressions of multiple sensor Rayleigh scattering distance domain signals are obtained as follows:

[0086]

[0087] in, is the kth sensor Rayleigh scattering range domain signal.

[0088] The spatial distance between any two adjacent sensor Rayleigh scattering range domain signals is:

[0089]

[0090] It can be seen that the spatial resolution of the data and the scanning wavelength range of the laser (i.e., the temporal resolution) are mutually restricted.

[0091] Step 26: Sample each sensor Rayleigh scattering distance domain signal according to a preset distance domain data point step size and a preset distance domain data point window length, to obtain a plurality of sensor Rayleigh scattering signals at random times and random positions, thereby forming an undersampled sensor Rayleigh scattering signal; and arrange each sensor Rayleigh scattering signal according to the preset distance step size and the preset distance window length.

[0092] Specifically, for sensing Rayleigh scattering range domain signal Take a subset of data at random positions with Δl as the preset distance step and L as the preset distance window length to obtain the undersampled sensing Rayleigh scattering signal

[0093] Step 27: Using a cross-correlation algorithm, perform strain demodulation on the undersampled reference Rayleigh scattering signal and the undersampled sensing Rayleigh scattering signal to obtain an undersampled demodulated two-dimensional strain distribution of the sensing optical fiber; the undersampled demodulated two-dimensional strain distribution is a demodulated two-dimensional strain distribution of the sensing optical fiber in a random distribution with a preset time step and a preset time window length and a demodulation window with a preset distance step and a preset distance window length.

[0094] Specifically, the cross-correlation algorithm is a traditional cross-correlation algorithm. By performing strain demodulation according to the traditional cross-correlation algorithm, the under-sampled demodulated two-dimensional strain distribution y can be calculated.

[0095] Step 3: Input the undersampled demodulated two-dimensional strain distribution into the two-dimensional strain distribution restoration model to obtain the restored two-dimensional strain distribution of the sensing optical fiber.

[0096] Among them, the two-dimensional strain distribution restoration model is obtained by training the compressed sensing deep neural network.

[0097] As an optional implementation, the process of determining the two-dimensional strain distribution restoration model includes:

[0098] Step 31: Initialize the compressed sensing deep neural network; the compressed sensing deep neural network includes multiple reconstruction layers, multiple auxiliary variable update blocks and multiple multiplier update layers; wherein the current reconstruction layer is connected to the current auxiliary variable update block and the current multiplier update layer respectively, the current auxiliary variable update block is connected to the current multiplier update layer, and the current auxiliary variable update block and the current multiplier update layer are both connected to the next reconstruction layer.

[0099] Specifically, the structure of the compressed sensing deep neural network is as follows Figure 4 As shown. Among them, X (1) is the first reconstruction layer, X (p-1) is the p-1th reconstruction layer, Z (p-1) Update block for the p-1th auxiliary variable, M (p-1) For the p-1th multiplier update layer, X (p) is the p-th reconstruction layer, Z (p) Update block for the pth auxiliary variable, M (p) For the p-th multiplier update layer, X (p+1) is the p+1th reconstruction layer, Z (p+1) Update block for the p+1th auxiliary variable, M (p+1) For the p+1th multiplier update layer, X (P) For X (1) To X (p+1) The comprehensive matrix, A (p,1) Update the first added layer within the block for the pth auxiliary variable, A (p,q-1) Update the q-1th added layer within the pth auxiliary variable block, Update the qth first convolutional layer in the block for the pth auxiliary variable, H (p,q) Update the qth nonlinear activation layer within the block for the pth auxiliary variable, Update the qth second convolutional layer within the block for the pth auxiliary variable, A (p,q) Update the qth added layer within the block for the pth auxiliary variable, A (p,Q) A (p,1) -A (p,q) The comprehensive matrix.

[0100] The compressed sensing deep neural network is represented as the following matrix:

[0101] y=φx+w。

[0102] Where φ is the compressed sensing sampling matrix; x is the undersampled demodulated two-dimensional strain distribution; and w is the noise.

[0103] Step 32: Obtain a training set; the training set includes: a plurality of samples of two-dimensional strain distributions to be restored and the corresponding two-dimensional strain distributions after restoration.

[0104] Step 33: Use the training set to train the compressed sensing deep neural network to obtain a two-dimensional strain distribution restoration model.

[0105] Step 4: Using the sliding average matrix corresponding to the preset demodulation parameters, perform a two-dimensional deconvolution calculation on the restored two-dimensional strain distribution to obtain the demodulated two-dimensional strain distribution of the sensing optical fiber.

[0106] As an optional implementation, step 4 includes:

[0107] Step 41: Determine a sliding average matrix based on the time ratio and the distance ratio; the time ratio is the ratio of the preset time window length to the preset time step length, and the distance ratio is the ratio of the preset distance window length to the preset distance step length.

[0108] Step 42: Using the sliding average matrix, perform a two-dimensional deconvolution calculation on the restored two-dimensional strain distribution to obtain the demodulated two-dimensional strain distribution of the sensing optical fiber.

[0109] Specifically, during the OFDR demodulation process, due to the convolution-like effect of the cross-correlation algorithm, within the interval of uniform and linear strain changes, the strain value demodulated in a certain time or space window is the mean of the strain distribution in this window. Approximately, the mean effect still exists in the area of small strain changes. Therefore, in this application, strain demodulation is performed with a small sliding step Δt, Δz in a time and space window of length T, Z. The demodulated undersampled demodulated two-dimensional strain distribution x is equal to the sliding average effect of the strain value x' (i.e., the demodulated two-dimensional strain distribution of the sensing optical fiber) under the small sliding window Δt, Δz. The size of the sliding average matrix depends on the ratio of T, Z to Δt, Δz. For example, Figure 5 and Figure 6 In the data shown, T = 3Δt, Z = 5Δz, that is, the time ratio is 3 and the distance ratio is 5. Therefore, the sliding average matrix h is:

[0110]

[0111] Since the sliding average matrix is flip symmetric, the sliding average filter is equivalent to the convolution operation:

[0112] x=h*x'.

[0113] Wherein, x' is the demodulated two-dimensional strain distribution of the sensing fiber.

[0114] Therefore, through two-dimensional deconvolution calculation, the strain distribution x' at the sliding step size Δt, Δz resolution is restored using x, breaking through the spatiotemporal resolution limitation of OFDR demodulation.

[0115] In this embodiment, the two-dimensional deconvolution is the Lucky-Richardson deconvolution algorithm, and the obtained x' is as follows Figure 6 shown.

[0116] This application introduces compressed sensing deep neural network and two-dimensional deconvolution into the time-resolved OFDR scheme, which takes advantage of the compressed sensing deep neural network's ability to efficiently and quickly restore original data with under-sampled data to improve the demodulation speed of the time-resolved OFDR scheme, and uses two-dimensional deconvolution to deconvolve the strain data under a large sliding window to obtain the accurate strain distribution under a small sliding step size, thereby improving the demodulation spatiotemporal resolution of the time-resolved scheme in principle and breaking through the mutual limitations of spatial and temporal resolution and measurement accuracy of traditional time-resolved OFDR.

[0117] In an exemplary embodiment, a dynamic strain sensing signal demodulation system of an OFDR system is provided to implement a dynamic strain sensing signal demodulation method of an OFDR system. The dynamic strain sensing signal demodulation system of the OFDR system includes:

[0118] The array acquisition module is used to acquire a reference Rayleigh scattering signal intensity array and a sensing Rayleigh scattering signal intensity array using an OFDR system.

[0119] The cross-correlation demodulation module is used to perform cross-correlation demodulation on the reference Rayleigh scattering signal intensity array and the sensing Rayleigh scattering signal intensity array according to preset demodulation parameters to obtain the under-sampling demodulated two-dimensional strain distribution of the sensing optical fiber.

[0120] The restoration module is used to input the undersampled demodulated two-dimensional strain distribution into the two-dimensional strain distribution restoration model to obtain the restored two-dimensional strain distribution of the sensing optical fiber; the two-dimensional strain distribution restoration model is obtained by training the compressed sensing deep neural network.

[0121] The deconvolution module is used to perform a two-dimensional deconvolution calculation on the restored two-dimensional strain distribution using a sliding average matrix corresponding to a preset demodulation parameter to obtain the demodulated two-dimensional strain distribution of the sensing optical fiber.

[0122] In an exemplary embodiment, a computer device is provided, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a dynamic strain sensing signal demodulation method for an OFDR system.

[0123] In an exemplary embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, a dynamic strain sensing signal demodulation method of an OFDR system is implemented.

[0124] In an exemplary embodiment, a computer program product is provided, including a computer program, which, when executed by a processor, implements a dynamic strain sensing signal demodulation method of an OFDR system.

[0125] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 7 As shown. The computer device includes a processor, a memory, an input / output interface (I / O) and a communication interface. The processor, the memory and the input / output interface are connected via a system bus, and the communication interface is connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a dynamic strain sensing signal demodulation method of an OFDR system is implemented.

[0126] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0127] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0128] The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may include, but are not limited to, general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic units, data processing logic units based on quantum computing, and the like.

[0129] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0130] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0131] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, for those skilled in the art, based on the concept of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A dynamic strain sensing signal demodulation method for an OFDR system, characterized in that: The dynamic strain sensing signal demodulation method of the OFDR system includes: Obtaining a reference Rayleigh scattering signal intensity array and a sensing Rayleigh scattering signal intensity array using an OFDR system; According to the preset demodulation parameters, the reference Rayleigh scattering signal intensity array and the sensing Rayleigh scattering signal intensity array are cross-correlatedly demodulated to obtain the under-sampling demodulated two-dimensional strain distribution of the sensing optical fiber; Inputting the undersampled demodulated two-dimensional strain distribution into a two-dimensional strain distribution restoration model to obtain a restored two-dimensional strain distribution of the sensing optical fiber; the two-dimensional strain distribution restoration model is obtained by training a compressed sensing deep neural network; A two-dimensional deconvolution calculation is performed on the restored two-dimensional strain distribution using the sliding average matrix corresponding to the preset demodulation parameters to obtain the demodulated two-dimensional strain distribution of the sensing optical fiber.

2. The dynamic strain sensing signal demodulation method of the OFDR system according to claim 1, characterized in that: The OFDR system is used to obtain a reference Rayleigh scattering signal intensity array and a sensing Rayleigh scattering signal intensity array, including: Before the sensing fiber is strained, the OFDR system injects a frequency-linearly tuned laser into the sensing fiber and simultaneously collects Rayleigh backscattered signals at propagation constants corresponding to different laser frequencies, obtaining Rayleigh backscattered frequency domain signals at multiple propagation constants before the sensing fiber is strained. Based on the intensities of the backward Rayleigh scattering frequency domain signals under all propagation constants before the strain of the sensing optical fiber, a reference Rayleigh scattering signal intensity array is constructed; After the sensing fiber is strained, the OFDR system injects a frequency-linearly tuned laser into the sensing fiber and simultaneously collects backscattered Rayleigh signals at propagation constants corresponding to different laser frequencies, obtaining backscattered Rayleigh frequency domain signals at multiple propagation constants after the strain of the sensing fiber. Based on the intensity of the back Rayleigh scattering frequency domain signal under all propagation constants after the strain of the sensing optical fiber, a sensing Rayleigh scattering signal intensity array is constructed.

3. The dynamic strain sensing signal demodulation method of the OFDR system according to claim 1, characterized in that: The preset demodulation parameters include: preset demodulation parameters, including: preset frequency domain data point step, preset frequency domain data point window length, preset time step, preset time window length, preset distance domain data point step, preset distance domain data point window length, preset distance step and preset distance window length; the time step corresponding to the preset frequency domain data point step is the preset time step, the time step corresponding to the preset frequency domain data point window length is the preset time window length, the distance step corresponding to the preset distance domain data point step is the preset distance step, and the distance step corresponding to the preset distance domain data point window length is the preset distance window length.

4. The dynamic strain sensing signal demodulation method of the OFDR system according to claim 3, characterized in that: According to preset demodulation parameters, the reference Rayleigh scattering signal intensity array and the sensing Rayleigh scattering signal intensity array are cross-correlatedly demodulated to obtain the under-sampling demodulated two-dimensional strain distribution of the sensing optical fiber, including: Dividing the reference Rayleigh scattering signal intensity array according to the preset frequency domain data point step size and the preset frequency domain data point window length to obtain a plurality of reference sub-segment spectrum signals; the reference sub-segment spectrum signals are arranged according to the preset time step size and the preset time window length; Performing discrete Fourier transform on each reference sub-segment spectrum signal to obtain multiple reference Rayleigh scattering range domain signals; sampling each reference Rayleigh scattering distance domain signal according to the preset distance domain data point step size and the preset distance domain data point window length, to obtain a plurality of reference Rayleigh scattering signals at random times and random positions, thereby forming an undersampled reference Rayleigh scattering signal; and arranging the reference Rayleigh scattering signals according to the preset distance step size and the preset distance window length; The sensor Rayleigh scattering signal intensity array is divided according to the preset frequency domain data point step size and the preset frequency domain data point window length to obtain a plurality of sensor sub-segment spectrum signals; the spectrum signals of each sensor sub-segment are arranged according to the preset time step size and the preset time window length; Performing discrete Fourier transform on the spectrum signal of each sensor segment respectively to obtain multiple sensor CorelDRAW range domain signals; sampling each sensor Rayleigh scattering distance domain signal according to the preset distance domain data point step size and the preset distance domain data point window length, to obtain a plurality of sensor Rayleigh scattering signals at random times and random positions, thereby forming an undersampled sensor Rayleigh scattering signal; and arranging each sensor Rayleigh scattering signal according to the preset distance step size and the preset distance window length; The undersampled reference Rayleigh scattering signal and the undersampled sensing Rayleigh scattering signal are strain demodulated by using a cross-correlation algorithm to obtain an undersampled demodulated two-dimensional strain distribution of the sensing optical fiber; the undersampled demodulated two-dimensional strain distribution is a demodulated two-dimensional strain distribution of the sensing optical fiber in a random distribution with a preset time step and a preset time window length and a demodulation window with a preset distance step and a preset distance window length.

5. The dynamic strain sensing signal demodulation method of the OFDR system according to claim 1, characterized in that: The process of determining the two-dimensional strain distribution restoration model includes: Initializing the compressed sensing deep neural network; the compressed sensing deep neural network includes multiple reconstruction layers, multiple auxiliary variable update blocks, and multiple multiplier update layers; wherein the current reconstruction layer is connected to the current auxiliary variable update block and the current multiplier update layer, respectively, the current auxiliary variable update block is connected to the current multiplier update layer, and the current auxiliary variable update block and the current multiplier update layer are both connected to the next reconstruction layer; Acquire a training set; the training set includes: a plurality of samples of two-dimensional strain distributions to be restored and corresponding two-dimensional strain distributions after restoration; The training set is used to train a compressed sensing deep neural network to obtain the two-dimensional strain distribution restoration model.

6. The dynamic strain sensing signal demodulation method of the OFDR system according to claim 3, characterized in that: Using the sliding average matrix corresponding to the preset demodulation parameters, a two-dimensional deconvolution calculation is performed on the restored two-dimensional strain distribution to obtain the demodulated two-dimensional strain distribution of the sensing optical fiber, including: Determining the sliding average matrix based on a time ratio and a distance ratio; the time ratio is a ratio of the preset time window length to the preset time step length, and the distance ratio is a ratio of the preset distance window length to the preset distance step length; The sliding average matrix is used to perform a two-dimensional deconvolution calculation on the restored two-dimensional strain distribution to obtain a demodulated two-dimensional strain distribution of the sensing optical fiber.

7. A dynamic strain sensing signal demodulation system of an OFDR system, for implementing the dynamic strain sensing signal demodulation method of an OFDR system according to any one of claims 1 to 6, characterized in that: The dynamic strain sensing signal demodulation system of the OFDR system includes: an array acquisition module, for acquiring a reference Rayleigh scattering signal intensity array and a sensing Rayleigh scattering signal intensity array using an OFDR system; A cross-correlation demodulation module is used to perform cross-correlation demodulation on the reference Rayleigh scattering signal intensity array and the sensing Rayleigh scattering signal intensity array according to preset demodulation parameters to obtain an undersampled demodulated two-dimensional strain distribution of the sensing optical fiber; a restoration module, configured to input the undersampled demodulated two-dimensional strain distribution into a two-dimensional strain distribution restoration model to obtain a restored two-dimensional strain distribution of the sensing optical fiber; the two-dimensional strain distribution restoration model is obtained by training a compressed sensing deep neural network; The deconvolution module is used to perform a two-dimensional deconvolution calculation on the restored two-dimensional strain distribution using the sliding average matrix corresponding to the preset demodulation parameters to obtain the demodulated two-dimensional strain distribution of the sensing optical fiber.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the dynamic strain sensing signal demodulation method of the OFDR system according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the dynamic strain sensing signal demodulation method of the OFDR system according to any one of claims 1 to 6 is implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the dynamic strain sensing signal demodulation method of the OFDR system according to any one of claims 1 to 6 is implemented.

Images