A frequency domain reflected strain measurement method, device, electronic equipment and medium

Abstract

The application discloses a frequency domain reflection strain measurement method and device, electronic equipment and medium, and relates to the technical field of distributed optical fiber strain sensing. The application obtains a cross-correlation numerical vector, a cross-correlation index vector and an initial strain distribution by carrying out a series of processing on a reference state beat signal and a measurement state beat signal, completes accurate identification in combination with a preset threshold value and a median and a standard deviation calculated on the basis of a local data window, extracts an extreme value peak index corresponding to a bad point, screens out a real spectrum offset matching the local median, and corrects strain data. The method only reuses the generated cross-correlation numerical vector and cross-correlation index vector to complete secondary operation, relies on software logic throughout the process, does not need to change the hardware structure of the optical frequency domain reflection distributed strain sensing system, effectively solves the demodulation deviation problem caused by the cross-correlation false peak, makes the final output strain distribution continuous and reliable, and controls the overall operation cost while ensuring the measurement accuracy.

Description

Technical Field

[0001] This application relates to the field of distributed optical fiber strain sensing technology, and in particular to a frequency domain reflection strain measurement method, device, electronic device and medium. Background Technology

[0002] Optical Frequency Domain Reflectometry (OFDR) is a high spatial resolution distributed fiber optic sensing technology widely used in fiber optic strain sensing, structural health monitoring, and medical catheter tracking. OFDR uses a tunable laser to emit linearly swept-frequency light. It generates beat signals by interfering with the backscattered Rayleigh light within the fiber and the local oscillator light. By performing spectral analysis or cross-correlation calculations on the beat signals, physical parameters such as strain and temperature at various locations along the fiber can be demodulated. Compared to optical time domain reflectometry, OFDR offers millimeter-level or even sub-millimeter-level spatial resolution, a high signal-to-noise ratio, and fast measurement speed, making it significantly advantageous in short-distance, high-precision measurement scenarios.

[0003] In OFDR-based distributed strain sensing, a common demodulation method is to perform cross-correlation on the Rayleigh scattering spectra of the reference and measurement states. The spectral shift is obtained by finding the position of the maximum peak in the cross-correlation result, and then the strain value is calculated. However, in practical applications, the spatial resolution of strain demodulation is limited by several factors. First, the frequency modulation nonlinearity of the laser causes broadening of the beat signal spectrum, reducing the sharpness of the cross-correlation peak. Second, to achieve high spatial resolution, a small sliding window number is required, but too few window points will result in insufficient effective data points for the cross-correlation operation, leading to spurious peaks in the cross-correlation result. Furthermore, when the strain is large, the true peak intensity decreases significantly, and the spurious peak value may exceed the true peak, causing demodulation errors. In addition, traditional methods to improve spatial resolution, such as external clock resampling, require long-delay optical fibers and have large system size; while image processing-based methods such as median filtering, two-dimensional wavelet denoising, and nonlocal mean filtering can suppress noise, they are computationally intensive and difficult to meet real-time requirements; deep learning-based methods require a large amount of training data and have limited generalization ability. Summary of the Invention

[0004] This application provides a frequency domain reflection strain measurement method, device, electronic device, and medium to at least address the problem of how to effectively improve the spatial resolution of OFDR distributed strain sensing while ensuring strain measurement accuracy and low computational complexity.

[0005] This application provides a frequency-domain reflection strain measurement method for an optical frequency-domain reflection distributed strain sensing system. The optical frequency-domain reflection distributed strain sensing system includes a tunable laser, a coupler, an auxiliary interferometer, a measuring interferometer, a circulator, a sensing fiber, and a balanced detector. The tunable laser outputs a linearly swept frequency optical signal. After being split by the coupler, one optical signal is sent to the auxiliary interferometer to complete laser frequency modulation nonlinearity correction, and the other optical signal is sent to the measuring interferometer. The optical signal sent to the measuring interferometer is split again, one path serving as the local oscillator signal, and the other path passing through the circulator and incident on the sensing fiber. The backscattered Rayleigh light signal returned by the sensing fiber interferes with the local oscillator signal and is converted into an electrical signal by the balanced detector. The method includes:

[0006] The reference beat signal and the measurement beat signal corresponding to the sensing fiber are acquired. Frequency domain transformation, segmentation, spectral reconstruction, spectral cross-correlation and initial strain calculation are performed on the reference beat signal and the measurement beat signal respectively to obtain the cross-correlation numerical vector, cross-correlation index vector and initial strain distribution of each sensing point.

[0007] Traverse each sensing point in the initial strain distribution, construct a local data window and calculate the median and standard deviation of the strain data within the window. Based on the median, the standard deviation and a preset threshold, identify sensing points that deviate from the normal range as bad points and record the locations of the bad points.

[0008] For each bad point, extract the extreme peak index values ​​corresponding to all extreme peaks from the cross-correlation numerical vector and cross-correlation index vector corresponding to the bad point, select the extreme peak index value with the smallest difference between the median of the strain data in the local data window corresponding to the bad point as the true spectral offset, recalculate the strain value of the bad point and replace the original bad point value to update the initial strain distribution;

[0009] The sensor fiber is traversed to correct all defects, and the corrected strain distribution is obtained and output.

[0010] Furthermore, the acquisition of the reference beat signal and the measurement beat signal corresponding to the sensing fiber involves performing frequency domain transformation, segmentation, spectral reconstruction, spectral cross-correlation, and initial strain calculation on the reference beat signal and the measurement beat signal, respectively, to obtain the cross-correlation numerical vector, cross-correlation index vector, and initial strain distribution for each sensing point, including:

[0011] The reference beat signal under the unstrained state of the sensing fiber and the measurement beat signal under the strained state of the sensing fiber were collected respectively.

[0012] The reference state beat signal and the measurement state beat signal are respectively subjected to fast Fourier transform to obtain the reference state range domain signal and the measurement state range domain signal, and the effective length range of the sensing fiber is determined.

[0013] The number of points and window overlap ratio of the sliding rectangular window are set. Based on the sliding rectangular window, signal segments are obtained from the reference state range domain signal and the measurement state range domain signal segment by segment. For each signal segment, interpolation and zero-padding and inverse fast Fourier transform are performed sequentially to reconstruct the reference state Rayleigh scattering spectrum and the measurement state Rayleigh scattering spectrum corresponding to each sensing point on the sensing fiber. The sensing point is a discrete measurement position divided along the axis of the sensing fiber.

[0014] Cross-correlation calculations are performed on the reference state Rayleigh scattering spectrum and the measurement state Rayleigh scattering spectrum of each sensing point to obtain the cross-correlation numerical vector and cross-correlation index vector of each sensing point.

[0015] The global maximum peak value is selected from the cross-correlation numerical vector of each sensing point, and the index value corresponding to the global maximum peak value in the cross-correlation index vector is retrieved as the spectral offset. Combined with the preset scaling factor, the strain of each sensing point is calculated in sequence to obtain the initial strain distribution.

[0016] Furthermore, the cross-correlation operation is performed on the reference-state Rayleigh scattering spectrum and the measurement-state Rayleigh scattering spectrum of each sensing point to obtain the cross-correlation numerical vector and cross-correlation index vector for each sensing point, including:

[0017] ;

[0018] In the formula Represents a cross-correlation numerical vector. Represents a vector of cross-correlation index values. and These are represented as cross-correlation numerical values ​​and cross-correlation index values, respectively.

[0019] The step of selecting the global maximum peak value from the cross-correlation numerical vector at each sensing point and retrieving the index value corresponding to the global maximum peak value in the cross-correlation index vector as the spectral shift includes:

[0020] In the formula This is the spectral shift. This is the scaling factor between the cross-correlation index value and the spectral offset;

[0021] The process of sequentially calculating the strain at each sensing point using a preset proportional coefficient to obtain the initial strain distribution includes:

[0022] In the formula The value represents the strain at each sensing point, s is the numerical value corresponding to a single spatial resolution, and K represents the scaling factor between the strain and the spectral offset.

[0023] Furthermore, the process of traversing each sensing point in the initial strain distribution, constructing a local data window, and calculating the median and standard deviation of the strain data within the window, and identifying sensing points deviating from the normal range as bad points based on the median, the standard deviation, and a preset threshold, and recording the locations of the bad points, includes:

[0024] Traverse each sensing point in the initial strain distribution, and take the current sensing point as the center to select several sensing points before and after the current sensing point to form a local data window;

[0025] Calculate the local median and local standard deviation of all strain data within each local data window;

[0026] Based on the laser frequency modulation nonlinearity and signal-to-noise ratio parameters, a preset discrimination threshold is set, and the absolute value of the difference between the strain value of the current sensing point and the corresponding local median is calculated.

[0027] If the absolute value of the difference is greater than the product of the local standard deviation and the preset discrimination threshold, then the current sensing point is determined to be a bad point;

[0028] By traversing all sensing points on the sensing fiber, all locations identified as bad points are summarized to obtain a set of bad point locations.

[0029] Furthermore, the calculation of the local median and local standard deviation of all strain data within each local data window; setting a preset discrimination threshold based on laser frequency modulation nonlinearity and signal-to-noise ratio parameters; calculating the absolute value of the difference between the strain value of the current sensing point and the corresponding local median; if the absolute value of the difference is greater than the product of the local standard deviation and the preset discrimination threshold, then the current sensing point is determined to be a bad point, including:

[0030] For each sensing point, take n points before and after it to form a local data window, and calculate the median and standard deviation of the strain data within this local data window:

[0031] ;

[0032] In the formula, median represents the median. This represents the local median of all strain data within the local data window. This represents the local standard deviation of all strain data within a local data window. Represented as each dependent variable within a local data window. represents the original dependent variable of the i-th sensing point, n represents the local data window formed by taking n points before and after each sensing point, and N is the preset discrimination threshold.

[0033] Furthermore, for each bad point, the extreme peak index values ​​corresponding to all extreme peaks are extracted from the cross-correlation numerical vector and cross-correlation index vector corresponding to the bad point. The extreme peak index value with the smallest median difference of strain data within the local data window corresponding to the bad point is selected as the true spectral offset. The strain value of the bad point is recalculated and replaced with the original strain value to update the initial strain distribution, including:

[0034] Based on the location of each bad point, retrieve the cross-correlation numerical vector and cross-correlation index vector of the corresponding sensor point, and extract the extreme peak index corresponding to all extreme peaks.

[0035] Using the local median of all strain data within the local data window corresponding to the bad point as the reference true value, and combining it with the strain transformation scaling factor, calculate the strain corresponding to each extreme peak index, and solve for the absolute difference between each strain and the local median.

[0036] The index of the extreme peak with the smallest absolute difference is selected as the true spectral offset. The strain value of the bad point is recalculated in combination with the scaling factor, and the original strain value corresponding to the initial strain distribution is replaced.

[0037] The original strain values ​​corresponding to all bad spots in the set of bad spot locations are corrected sequentially.

[0038] Furthermore, the step of selecting the index of the extreme peak with the smallest absolute difference as the true spectral shift includes:

[0039] The true spectral shift is ;

[0040] In the formula, min represents the minimum value, and f n For the cross-correlation index corresponding to each extreme peak, K represents the scaling factor between the cross-correlation index value and the spectral offset.

[0041] This application also provides a frequency domain reflection strain measurement device, the device comprising:

[0042] The signal acquisition module is used to acquire the reference beat signal and the measurement beat signal corresponding to the sensing fiber.

[0043] The signal processing and initial calculation module is connected to the signal acquisition module and is used to perform frequency domain transformation, segmentation, spectral reconstruction, spectral cross-correlation and initial strain calculation on the reference beat signal and the measured beat signal, respectively, to obtain the cross-correlation numerical vector, cross-correlation index vector and initial strain distribution of each sensing point.

[0044] The defective pixel identification module, connected to the signal processing and initial solution module, is used to traverse each sensing point in the initial strain distribution, construct a local data window and calculate the median and standard deviation of the strain data within the window, and identify sensing points that deviate from the normal range as defective pixels based on the median, the standard deviation and a preset threshold, and record the location of the defective pixels.

[0045] The peak discrimination and bad spot correction module is connected to the signal processing and initial solution module and the bad spot identification module, respectively. For each bad spot, it extracts the extreme peak index values ​​corresponding to all extreme peaks from the cross-correlation numerical vector and cross-correlation index vector corresponding to the bad spot, selects the extreme peak index value with the smallest median difference of strain data in the local data window corresponding to the bad spot as the true spectral offset, recalculates the strain value of the bad spot and replaces the original bad spot value to update the initial strain distribution.

[0046] The result output module, connected to the peak discrimination and bad pixel correction module, is used to traverse the sensing fiber, complete the correction of all bad pixels, obtain the corrected strain distribution, and output it.

[0047] This application also provides an electronic device, comprising: a memory for storing a computer program; and a processor for executing the computer program to implement the steps of any of the above-described frequency domain reflection strain measurement methods.

[0048] This application also provides a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, it implements the steps of any of the above-described frequency domain reflection strain measurement methods.

[0049] This application utilizes a series of processing steps on the reference and measured beat signals to obtain cross-correlation numerical vectors, cross-correlation index vectors, and an initial strain distribution. To address the issue of abnormal outliers in the initial strain distribution, accurate identification is achieved by combining the median and standard deviation calculated using a local data window with a preset threshold. Then, the extreme peak indices corresponding to the outliers are extracted, and the true spectral shifts matching the local median are selected to correct the strain data. This method reuses only the generated cross-correlation numerical vectors and cross-correlation index vectors to perform secondary calculations, relying entirely on software logic. It requires no modification to the hardware structure of the optical frequency domain reflective distributed strain sensing system, effectively solving the demodulation deviation problem caused by cross-correlation spurious peaks. This ensures a continuous and reliable strain distribution output, controls overall computational overhead while maintaining measurement accuracy, and improves the overall stability of distributed strain measurement. Attached Figure Description

[0050] To more clearly illustrate the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 This is a schematic diagram of the structure of a distributed strain sensing system with optical frequency domain reflection in one embodiment of this application;

[0052] Figure 2 This is a logic diagram of a frequency domain reflection strain measurement method in one embodiment of this application;

[0053] Figure 3 This is a schematic diagram of the initial strain distribution with defects in one embodiment of this application;

[0054] Figure 4 This is a schematic diagram of the initial strain distribution after defect identification in one embodiment of this application;

[0055] Figure 5 This is a schematic diagram of the simulation results of the modified strain distribution in one embodiment of this application;

[0056] Figure 6 This is a structural block diagram of a frequency domain reflection strain measurement device in one embodiment of this application;

[0057] Figure 7 This is an internal structural diagram of a computer device in one embodiment of this application. Detailed Implementation

[0058] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the protection scope of this application.

[0059] It should be noted that, in the description of this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. The terms "first," "second," etc., in this application are used to distinguish similar objects and are not used to describe a specific order or sequence.

[0060] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0061] This application provides a frequency-domain reflection strain measurement method for optical frequency-domain reflection distributed strain sensing systems, such as... Figure 1 As shown, the optical frequency domain reflection distributed strain sensing system includes a tunable laser, a coupler, an auxiliary interferometer, a measuring interferometer, a circulator, a sensing fiber, and a balanced detector. The tunable laser outputs a linearly swept frequency optical signal. After being split by the coupler, one optical signal is sent to the auxiliary interferometer to complete the laser frequency modulation nonlinearity correction, and the other optical signal is sent to the measuring interferometer. The optical signal sent to the measuring interferometer is split again, one of which is used as the local oscillator signal, and the other is then incident on the sensing fiber through the circulator. The backscattered Rayleigh light signal returned by the sensing fiber interferes with the local oscillator signal and is converted into an electrical signal by the balanced detector.

[0062] like Figure 2 As shown, the frequency domain reflection strain measurement method includes:

[0063] The reference beat signal and the measurement beat signal corresponding to the sensing fiber are acquired. Frequency domain transformation, segmentation, spectral reconstruction, spectral cross-correlation and initial strain calculation are performed on the reference beat signal and the measurement beat signal respectively to obtain the cross-correlation numerical vector, cross-correlation index vector and initial strain distribution of each sensing point.

[0064] Traverse each sensing point in the initial strain distribution, construct a local data window and calculate the median and standard deviation of the strain data within the window. Based on the median, the standard deviation and a preset threshold, identify sensing points that deviate from the normal range as bad points and record the locations of the bad points.

[0065] For each bad point, extract the extreme peak index values ​​corresponding to all extreme peaks from the cross-correlation numerical vector and cross-correlation index vector corresponding to the bad point, select the extreme peak index value with the smallest difference between the median of the strain data in the local data window corresponding to the bad point as the true spectral offset, recalculate the strain value of the bad point and replace the original bad point value to update the initial strain distribution;

[0066] The sensing fiber is traversed to correct all defects, and the corrected strain distribution is obtained and output.

[0067] Furthermore, the acquisition of the reference beat signal and the measurement beat signal corresponding to the sensing fiber involves performing frequency domain transformation, segmentation, spectral reconstruction, spectral cross-correlation, and initial strain calculation on the reference beat signal and the measurement beat signal, respectively, to obtain the cross-correlation numerical vector, cross-correlation index vector, and initial strain distribution for each sensing point, including:

[0068] The reference beat signal under the unstrained state of the sensing fiber and the measurement beat signal under the strained state of the sensing fiber were collected respectively.

[0069] The reference state beat signal and the measurement state beat signal are respectively subjected to fast Fourier transform to obtain the reference state range domain signal and the measurement state range domain signal, and the effective length range of the sensing fiber is determined.

[0070] The number of points and window overlap ratio of the sliding rectangular window are set. Based on the sliding rectangular window, signal segments are obtained from the reference state range domain signal and the measurement state range domain signal segment by segment. For each signal segment, interpolation and zero-padding and inverse fast Fourier transform are performed sequentially to reconstruct the reference state Rayleigh scattering spectrum and the measurement state Rayleigh scattering spectrum corresponding to each sensing point on the sensing fiber. The sensing point is a discrete measurement position divided along the axis of the sensing fiber.

[0071] Cross-correlation calculations are performed on the reference state Rayleigh scattering spectrum and the measurement state Rayleigh scattering spectrum of each sensing point to obtain the cross-correlation numerical vector and cross-correlation index vector of each sensing point.

[0072] The global maximum peak value is selected from the cross-correlation numerical vector of each sensing point, and the index value corresponding to the global maximum peak value in the cross-correlation index vector is retrieved as the spectral offset. Combined with the preset scaling factor, the strain of each sensing point is calculated in sequence to obtain the initial strain distribution.

[0073] Furthermore, the cross-correlation operation is performed on the reference-state Rayleigh scattering spectrum and the measurement-state Rayleigh scattering spectrum of each sensing point to obtain the cross-correlation numerical vector and cross-correlation index vector for each sensing point, including:

[0074] ;

[0075] In the formula Represents a cross-correlation numerical vector. Represents a vector of cross-correlation index values. and These are represented as cross-correlation numerical values ​​and cross-correlation index values, respectively.

[0076] The step of selecting the global maximum peak value from the cross-correlation numerical vector at each sensing point and retrieving the index value corresponding to the global maximum peak value in the cross-correlation index vector as the spectral shift includes:

[0077] In the formula This is the spectral shift. This is the scaling factor between the cross-correlation index value and the spectral offset;

[0078] The process of sequentially calculating the strain at each sensing point using a preset proportional coefficient to obtain the initial strain distribution includes:

[0079] In the formula The value represents the strain at each sensing point, s is the numerical value corresponding to a single spatial resolution, and K represents the scaling factor between the strain and the spectral offset.

[0080] The complete cross-correlation results (numerical vector and index vector) for each sensing point are saved. The spatial resolution is then determined by the number of points in the sliding window; a smaller window size results in higher spatial resolution, but may also introduce numerous bad points in the initial strain distribution. Figure 3 As shown.

[0081] Furthermore, the process of traversing each sensing point in the initial strain distribution, constructing a local data window, and calculating the median and standard deviation of the strain data within the window, and identifying sensing points deviating from the normal range as bad points based on the median, the standard deviation, and a preset threshold, and recording the locations of the bad points, includes:

[0082] Traverse each sensing point in the initial strain distribution, and take the current sensing point as the center to select several sensing points before and after the current sensing point to form a local data window;

[0083] Calculate the local median and local standard deviation of all strain data within each local data window;

[0084] Based on the laser frequency modulation nonlinearity and signal-to-noise ratio parameters, a preset discrimination threshold is set, and the absolute value of the difference between the strain value of the current sensing point and the corresponding local median is calculated.

[0085] If the absolute value of the difference is greater than the product of the local standard deviation and the preset discrimination threshold, then the current sensing point is determined to be a bad point;

[0086] By traversing all sensing points on the sensing fiber, all locations identified as bad points are summarized to obtain a set of bad point locations.

[0087] Furthermore, the calculation of the local median and local standard deviation of all strain data within each local data window; setting a preset discrimination threshold based on laser frequency modulation nonlinearity and signal-to-noise ratio parameters; calculating the absolute value of the difference between the strain value of the current sensing point and the corresponding local median; if the absolute value of the difference is greater than the product of the local standard deviation and the preset discrimination threshold, then the current sensing point is determined to be a bad point, including:

[0088] For each sensing point, take n points before and after it to form a local data window, and calculate the median and standard deviation of the strain data within this local data window:

[0089] ;

[0090] In the formula, median represents the median. This represents the local median of all strain data within the local data window. This represents the local standard deviation of all strain data within a local data window. Represented as each dependent variable within a local data window. Let represent the original strain at the i-th sensing point, n represent the local data window formed by taking n points before and after each sensing point, and N be the preset discrimination threshold. The preset discrimination threshold N can be preset according to system parameters (such as laser sweep frequency nonlinearity, signal-to-noise ratio, etc.). Figure 4 The strain distribution after defect identification.

[0091] Furthermore, for each bad point, the extreme peak index values ​​corresponding to all extreme peaks are extracted from the cross-correlation numerical vector and cross-correlation index vector corresponding to the bad point. The extreme peak index value with the smallest median difference of strain data within the local data window corresponding to the bad point is selected as the true spectral offset. The strain value of the bad point is recalculated and replaced with the original strain value to update the initial strain distribution, including:

[0092] Based on the location of each bad point, retrieve the cross-correlation numerical vector and cross-correlation index vector of the corresponding sensor point, and extract the extreme peak index corresponding to all extreme peaks.

[0093] Using the local median of all strain data within the local data window corresponding to the bad point as the reference true value, and combining it with the strain transformation scaling factor, calculate the strain corresponding to each extreme peak index, and solve for the absolute difference between each strain and the local median.

[0094] The index of the extreme peak with the smallest absolute difference is selected as the true spectral offset. The strain value of the bad point is recalculated in combination with the scaling factor, and the original strain value corresponding to the initial strain distribution is replaced.

[0095] The original strain values ​​corresponding to all bad spots in the set of bad spot locations are corrected sequentially.

[0096] The corrected strain distribution eliminates the influence of cross-correlation spurious peaks, and while maintaining the high spatial resolution brought about by the small sliding window, the strain values ​​are continuous and free of abrupt bad spots. Figure 5 Simulation results of the corrected strain distribution are presented.

[0097] Furthermore, the step of selecting the index of the extreme peak with the smallest absolute difference as the true spectral shift includes:

[0098] The true spectral shift is ;

[0099] In the formula, min represents the minimum value, and f n For the cross-correlation index corresponding to each extreme peak, K represents the scaling factor between the cross-correlation index value and the spectral offset.

[0100] This embodiment effectively eliminates the interference of cross-correlation spurious peaks on demodulation results under small sliding windows by utilizing only the original cross-correlation results, and significantly improves the spatial resolution of OFDR distributed strain sensing without increasing any hardware costs.

[0101] This embodiment demonstrates excellent adaptability to continuous variations in fiber strain, applicable not only to single constant strain but also to strain distributions varying along the sensing length (such as bending deformation), exhibiting universality. Furthermore, this method can be combined with sliding window overlapping movement technology to further improve effective spatial resolution without adding extra signal acquisition overhead.

[0102] This embodiment is implemented entirely based on software algorithms, without requiring any changes to the optical path structure of the existing OFDR system or the addition of auxiliary hardware such as long-delay optical fibers or optical switches. The system is small in size and low in cost, and is particularly suitable for application scenarios with high spatial resolution requirements, such as medical catheter tracking, flexible robot shape perception, and structural health monitoring.

[0103] In the aforementioned frequency domain reflection strain measurement method, a series of processing steps are performed on the reference beat signal and the measured beat signal to obtain the cross-correlation numerical vector, cross-correlation index vector, and initial strain distribution. To address the issue of abnormal bad spots in the initial strain distribution, accurate identification is achieved by combining the median and standard deviation calculated using a local data window with a preset threshold. Then, the extreme peak index corresponding to the bad spots is extracted, and the true spectral shift matching the local median is selected to correct the strain data. This method only reuses the already generated cross-correlation numerical vector and cross-correlation index vector to complete the secondary calculations, relying entirely on software logic. It requires no modification to the hardware structure of the optical frequency domain reflection distributed strain sensing system, effectively solving the demodulation deviation problem caused by cross-correlation spurious peaks, ensuring the continuous and reliable output strain distribution. While guaranteeing measurement accuracy, it controls the overall computational overhead, improving the overall stability of distributed strain measurement.

[0104] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method.

[0105] In one embodiment, such as Figure 6 As shown, a frequency domain reflection strain measurement device is provided, the device comprising:

[0106] The signal acquisition module is used to acquire the reference beat signal and the measurement beat signal corresponding to the sensing fiber.

[0107] The signal processing and initial calculation module is connected to the signal acquisition module and is used to perform frequency domain transformation, segmentation, spectral reconstruction, spectral cross-correlation and initial strain calculation on the reference beat signal and the measured beat signal, respectively, to obtain the cross-correlation numerical vector, cross-correlation index vector and initial strain distribution of each sensing point.

[0108] The defective pixel identification module, connected to the signal processing and initial solution module, is used to traverse each sensing point in the initial strain distribution, construct a local data window and calculate the median and standard deviation of the strain data within the window, and identify sensing points that deviate from the normal range as defective pixels based on the median, the standard deviation and a preset threshold, and record the location of the defective pixels.

[0109] The peak discrimination and bad spot correction module is connected to the signal processing and initial solution module and the bad spot identification module, respectively. For each bad spot, it extracts the extreme peak index values ​​corresponding to all extreme peaks from the cross-correlation numerical vector and cross-correlation index vector corresponding to the bad spot, selects the extreme peak index value with the smallest median difference of strain data in the local data window corresponding to the bad spot as the true spectral offset, recalculates the strain value of the bad spot and replaces the original bad spot value to update the initial strain distribution.

[0110] The result output module, connected to the peak discrimination and bad pixel correction module, is used to traverse the sensing fiber, complete the correction of all bad pixels, obtain the corrected strain distribution, and output it.

[0111] In the aforementioned frequency-domain reflectance strain measurement device, a series of processes are performed on the reference beat signal and the measured beat signal to obtain a cross-correlation numerical vector, a cross-correlation index vector, and an initial strain distribution. To address the issue of abnormal dead spots in the initial strain distribution, accurate identification is achieved by combining the median and standard deviation calculated using a local data window with a preset threshold. Then, the extreme peak index corresponding to the dead spots is extracted, and the true spectral shift matching the local median is selected to correct the strain data. This method only reuses the already generated cross-correlation numerical vector and cross-correlation index vector to complete the secondary calculations, relying entirely on software logic. It requires no modification to the hardware structure of the optical frequency-domain reflectance distributed strain sensing system, effectively solving the demodulation deviation problem caused by cross-correlation spurious peaks. This ensures that the final output strain distribution is continuous and reliable, controlling overall computational overhead while guaranteeing measurement accuracy and improving the overall stability of distributed strain measurement.

[0112] For a description of the features in the embodiment corresponding to the frequency domain reflection strain measurement device, please refer to the relevant description of the embodiment corresponding to the frequency domain reflection strain measurement method, which will not be repeated here.

[0113] Embodiments of this application also provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to perform the steps in any of the above-described frequency domain reflection strain measurement method embodiments.

[0114] In one embodiment, the electronic device may be a server, and its internal structure diagram may be as follows: Figure 7As shown, the electronic device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computational and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores frequency-domain reflection strain measurement data. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements a frequency-domain reflection strain measurement method.

[0115] Embodiments of this application also provide a computer-readable storage medium storing a computer program configured to execute the steps in any of the above-described frequency domain reflection strain measurement method embodiments.

[0116] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard disk, magnetic disk, or optical disk.

[0117] The embodiments of this application also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in any of the above-described frequency domain reflection strain measurement method embodiments.

[0118] Embodiments of this application also provide another computer program product, including a non-volatile computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in any of the above-described frequency domain reflection strain measurement method embodiments.

[0119] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0120] The foregoing has provided a detailed description of a frequency domain reflection strain measurement method, apparatus, electronic device, and medium provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are only intended to aid in understanding the method and core ideas of this application. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of the claims of this application.

Claims

1. A frequency domain reflection strain measurement method for an optical frequency domain reflection distributed strain sensing system, wherein the optical frequency domain reflection distributed strain sensing system comprises a tunable laser, a coupler, an auxiliary interferometer, a measurement interferometer, a circulator, a sensing fiber, and a balanced detector; The tunable laser outputs a linearly swept frequency optical signal. After being split by a coupler, one optical signal is sent to an auxiliary interferometer to complete the laser's frequency modulation nonlinearity correction, and the other optical signal is sent to a measurement interferometer. The optical signal sent to the measurement interferometer is split again, one path serving as the local oscillator signal, and the other path passing through a circulator and incident on a sensing fiber. The backscattered Rayleigh light signal returned from the sensing fiber interferes with the local oscillator signal, and is converted into an electrical signal by a balanced detector; characterized in that... The method includes: The reference beat signal and the measurement beat signal corresponding to the sensing fiber are acquired. Frequency domain transformation, segmentation, spectral reconstruction, spectral cross-correlation and initial strain calculation are performed on the reference beat signal and the measurement beat signal respectively to obtain the cross-correlation numerical vector, cross-correlation index vector and initial strain distribution of each sensing point. Traverse each sensing point in the initial strain distribution, construct a local data window and calculate the median and standard deviation of the strain data within the window. Based on the median, the standard deviation and a preset threshold, identify sensing points that deviate from the normal range as bad points and record the locations of the bad points. For each bad point, extract the extreme peak index values ​​corresponding to all extreme peaks from the cross-correlation numerical vector and cross-correlation index vector corresponding to the bad point, select the extreme peak index value with the smallest difference between the median of the strain data in the local data window corresponding to the bad point as the true spectral offset, recalculate the strain value of the bad point and replace the original bad point value to update the initial strain distribution; The sensing fiber is traversed to correct all defects, and the corrected strain distribution is obtained and output.

2. The frequency domain reflection strain measurement method according to claim 1, characterized in that, The process involves acquiring the reference beat signal and the measurement beat signal corresponding to the sensing fiber, and performing frequency domain transformation, segmentation, spectral reconstruction, spectral cross-correlation, and initial strain calculation on the reference beat signal and the measurement beat signal, respectively, to obtain the cross-correlation numerical vector, cross-correlation index vector, and initial strain distribution for each sensing point, including: The reference beat signal under the unstrained state of the sensing fiber and the measurement beat signal under the strained state of the sensing fiber were collected respectively. The reference state beat signal and the measurement state beat signal are respectively subjected to fast Fourier transform to obtain the reference state range domain signal and the measurement state range domain signal, and the effective length range of the sensing fiber is determined. The number of points and window overlap ratio of the sliding rectangular window are set. Based on the sliding rectangular window, signal segments are obtained from the reference state range domain signal and the measurement state range domain signal segment by segment. For each signal segment, interpolation and zero-padding and inverse fast Fourier transform are performed sequentially to reconstruct the reference state Rayleigh scattering spectrum and the measurement state Rayleigh scattering spectrum corresponding to each sensing point on the sensing fiber. The sensing point is a discrete measurement position divided along the axis of the sensing fiber. Cross-correlation calculations are performed on the reference state Rayleigh scattering spectrum and the measurement state Rayleigh scattering spectrum of each sensing point to obtain the cross-correlation numerical vector and cross-correlation index vector of each sensing point. The global maximum peak value is selected from the cross-correlation numerical vector of each sensing point, and the index value corresponding to the global maximum peak value in the cross-correlation index vector is retrieved as the spectral offset. Combined with the preset scaling factor, the strain of each sensing point is calculated in sequence to obtain the initial strain distribution.

3. The frequency domain reflection strain measurement method according to claim 2, characterized in that, The process of performing cross-correlation calculations on the reference-state Rayleigh scattering spectrum and the measurement-state Rayleigh scattering spectrum at each sensing point yields a cross-correlation numerical vector and a cross-correlation index vector for each sensing point, including: ； In the formula Represents a cross-correlation numerical vector. Represents a vector of cross-correlation index values. and These are represented as cross-correlation numerical values ​​and cross-correlation index values, respectively. The step of selecting the global maximum peak value from the cross-correlation numerical vector at each sensing point and retrieving the index value corresponding to the global maximum peak value in the cross-correlation index vector as the spectral shift includes: In the formula This is the spectral shift. This is the scaling factor between the cross-correlation index value and the spectral offset; The process of sequentially calculating the strain at each sensing point using a preset proportional coefficient to obtain the initial strain distribution includes: In the formula The value represents the strain at each sensing point, s is the numerical value corresponding to a single spatial resolution, and K represents the scaling factor between the strain and the spectral offset.

4. The frequency domain reflection strain measurement method according to claim 1, characterized in that, The process involves traversing each sensing point in the initial strain distribution, constructing a local data window, and calculating the median and standard deviation of the strain data within the window. Sensing points deviating from the normal range are identified as bad points based on the median, the standard deviation, and a preset threshold. The locations of these bad points are recorded, including: Traverse each sensing point in the initial strain distribution, and take the current sensing point as the center to select several sensing points before and after the current sensing point to form a local data window; Calculate the local median and local standard deviation of all strain data within each local data window; Based on the laser frequency modulation nonlinearity and signal-to-noise ratio parameters, a preset discrimination threshold is set, and the absolute value of the difference between the strain value of the current sensing point and the corresponding local median is calculated. If the absolute value of the difference is greater than the product of the local standard deviation and the preset discrimination threshold, then the current sensing point is determined to be a bad point; By traversing all sensing points on the sensing fiber, all locations identified as bad points are summarized to obtain a set of bad point locations.

5. The frequency domain reflection strain measurement method according to claim 4, characterized in that, The calculation involves finding the local median and local standard deviation of all strain data within each local data window; and setting a preset discrimination threshold based on the laser frequency modulation nonlinearity and signal-to-noise ratio parameters to calculate the absolute value of the difference between the strain value of the current sensing point and the corresponding local median. If the absolute value of the difference is greater than the product of the local standard deviation and the preset discrimination threshold, then the current sensing point is determined to be a bad point, including: For each sensing point, take n points before and after it to form a local data window, and calculate the median and standard deviation of the strain data within this local data window: ； In the formula, median represents the median. This represents the local median of all strain data within the local data window. This represents the local standard deviation of all strain data within a local data window. Represented as each dependent variable within a local data window. represents the original dependent variable of the i-th sensing point, n represents the local data window formed by taking n points before and after each sensing point, and N is the preset discrimination threshold.

6. The frequency domain reflection strain measurement method according to claim 5, characterized in that, For each bad point, the extreme peak index values ​​corresponding to all extreme peaks are extracted from the cross-correlation numerical vector and cross-correlation index vector corresponding to the bad point. The extreme peak index value with the smallest median difference of strain data within the local data window corresponding to the bad point is selected as the true spectral offset. The strain value of the bad point is recalculated and replaced with the original strain value to update the initial strain distribution, including: Based on the location of each bad point, retrieve the cross-correlation numerical vector and cross-correlation index vector of the corresponding sensor point, and extract the extreme peak index corresponding to all extreme peaks. Using the local median of all strain data within the local data window corresponding to the bad point as the reference true value, and combining it with the strain transformation scaling factor, calculate the strain corresponding to each extreme peak index, and solve for the absolute difference between each strain and the local median. The index of the extreme peak with the smallest absolute difference is selected as the true spectral offset. The strain value of the bad point is recalculated in combination with the scaling factor, and the original strain value corresponding to the initial strain distribution is replaced. The original strain values ​​corresponding to all bad spots in the set of bad spot locations are corrected sequentially.

7. The frequency domain reflection strain measurement method according to claim 6, characterized in that, The step of selecting the extreme peak index with the smallest absolute difference as the true spectral offset includes: The true spectral shift is ; In the formula, min represents the minimum value, and f n For the cross-correlation index corresponding to each extreme peak, K represents the scaling factor between the cross-correlation index value and the spectral offset.

8. A frequency domain reflection strain measuring device, characterized in that, The device includes: The signal acquisition module is used to acquire the reference beat signal and the measurement beat signal corresponding to the sensing fiber. The signal processing and initial calculation module is connected to the signal acquisition module and is used to perform frequency domain transformation, segmentation, spectral reconstruction, spectral cross-correlation and initial strain calculation on the reference beat signal and the measured beat signal, respectively, to obtain the cross-correlation numerical vector, cross-correlation index vector and initial strain distribution of each sensing point. The defective pixel identification module, connected to the signal processing and initial solution module, is used to traverse each sensing point in the initial strain distribution, construct a local data window and calculate the median and standard deviation of the strain data within the window, and identify sensing points that deviate from the normal range as defective pixels based on the median, the standard deviation and a preset threshold, and record the location of the defective pixels. The peak discrimination and bad spot correction module is connected to the signal processing and initial solution module and the bad spot identification module, respectively. For each bad spot, it extracts the extreme peak index values ​​corresponding to all extreme peaks from the cross-correlation numerical vector and cross-correlation index vector corresponding to the bad spot, selects the extreme peak index value with the smallest median difference of strain data in the local data window corresponding to the bad spot as the true spectral offset, recalculates the strain value of the bad spot and replaces the original bad spot value to update the initial strain distribution. The result output module, connected to the peak discrimination and bad pixel correction module, is used to traverse the sensing fiber, complete the correction of all bad pixels, obtain the corrected strain distribution, and output it.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the steps of the frequency domain reflection strain measurement method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, it implements the steps of the frequency domain reflection strain measurement method as described in any one of claims 1 to 7.

Images