Distributed optical fiber sensing positioning system and positioning method based on asymmetric double Mach-Zehnder

Abstract

The invention discloses a distributed optical fiber sensing positioning system and positioning method based on asymmetric double Mach-Zehnder. The distributed optical fiber sensing positioning system comprises an asymmetric double Mach-Zehnder distributed optical fiber sensing module and a disturbance positioning module. The optical fiber sensing module comprises a first light source, a second light source, an asymmetric double Mach-Zehnder interferometer, a data acquisition card and an industrial personal computer; two beams of light emitted by the first light source and the second light source are respectively propagated clockwise and anticlockwise in a sensing loop, and interfere in couplers at opposite ends in the interferometer to form two paths of interference signals, and the data acquisition card acquires the two paths of interference signals, converts the two paths of interference signals into electric signals, and sends the electric signals to the industrial personal computer for signal processing; the disturbance positioning module is used for obtaining a positioning result, a double-endpoint detection method based on median filtering is adopted in the aspect of endpoint detection, iterative variational mode decomposition is used in the aspect of time-frequency change feature extraction, and envelope extraction and generalized secondary cross-correlation time delay estimation algorithms are used in the aspect of time delay estimation positioning.

Description

Technical Field

[0001] The present invention belongs to the technical field of sensing and detection, and particularly relates to a distributed optical fiber sensing and positioning system and a positioning method based on double-end detection and iterative variational mode decomposition. Background Art

[0002] With the continuous progress of society and the continuous development of technology, the field of security monitoring has received increasing attention, and perimeter security monitoring has been widely applied in people's livelihood fields such as oil and gas pipelines, high-speed railway lines, and power transmission networks, as well as in fields with high security requirements such as airports. Among them, distributed optical fiber sensing technology is a very important one.

[0003] The distributed optical fiber sensing technology based on the principle of asymmetric dual Mach-Zehnder interference uses optical wave interference technology to achieve intrusion disturbance detection and positioning. Compared with the more traditional dual Mach-Zehnder sensing system, it further extends the sensing distance in its inherent advantages. However, the introduction of dual wavelengths reduces the Rayleigh reflection noise on the one hand and makes the system asymmetric, and the similarity between the two signals. Currently, the primary problem of the positioning algorithm based on the principle of asymmetric dual Mach-Zehnder interference is to solve this problem. Researchers have well solved this problem from two aspects: time-frequency analysis and phase demodulation. In time-frequency analysis, wavelet transform, short-time Fourier transform, S transform, etc. are mostly used, but the selection of its relevant parameters affects its analysis ability for signals in different frequency bands. This makes it only effective for specific frequency bands, lacking self-adaptability and wide-frequency positioning ability. At the same time, there is less research on weak disturbance positioning at the present stage.

[0004] Therefore, a method based on the principle of asymmetric dual Mach-Zehnder interference is needed to achieve the wide-frequency positioning function on the basis of overcoming asymmetry. At the same time, the positioning function of weak disturbances is improved. Summary of the Invention

[0005] The purpose of the present invention is to solve the problems of difficult wide-frequency domain positioning and low weak disturbance positioning accuracy of the current distributed optical fiber sensing technology positioning algorithm based on the principle of asymmetric dual Mach-Zehnder interference, and to propose a distributed optical fiber sensing and positioning system and a distributed optical fiber sensing wide-frequency positioning method based on double-end detection and iterative variational mode decomposition with fast positioning ability. This method uses a double-end detection method based on median filtering in endpoint detection, iterative variational mode decomposition in time-frequency change feature extraction, and envelope extraction and generalized cross-correlation time delay estimation in time delay estimation and positioning. The combined use of the three solves the problem of wide-frequency positioning and improves the weak disturbance positioning accuracy.

[0006] A distributed optical fiber sensing and positioning system based on an asymmetric dual Mach-Zehnder includes

[0007] an asymmetric dual Mach-Zehnder distributed optical fiber sensing module and a disturbance positioning module; the optical fiber sensing module includes a first light source, a second light source, an asymmetric dual Mach-Zehnder interferometer, a data acquisition card, and an industrial control computer; two beams of light emitted by the first light source and the second light source propagate along the clockwise and counterclockwise directions respectively in the sensing loop, and interfere at the coupler at the opposite ends in the interferometer to form two interference signals, and the data acquisition card collects the two interference signals, converts them into electrical signals, and then sends them to the industrial control computer for signal processing;

[0008] the disturbance positioning module is used to obtain the positioning result, in which a dual-endpoint detection method based on median filtering is adopted for endpoint detection, iterative variational mode decomposition is used for time-frequency variation feature extraction, and envelope extraction and generalized cross-correlation time delay estimation algorithm are used for time delay estimation and positioning. Further, the optical path of the light emitted by the first light source is the first optical circulator - the first polarization controller - the asymmetric dual Mach-Zehnder interferometer - the second polarization controller - the second optical circulator - the second dense wavelength division multiplexer - the second photodetector;

[0009] the optical path of the light emitted by the second light source is the second optical circulator - the second polarization controller - the asymmetric dual Mach-Zehnder interferometer - the first polarization controller - the first optical circulator - the first dense wavelength division multiplexer - the first photodetector.

[0010] A distributed optical fiber sensing broadband positioning method based on dual-endpoint detection and iterative variational mode decomposition for the distributed optical fiber sensing positioning system includes:

[0011] Step 1: Based on the optical fiber sensing and positioning system, the industrial control computer receives two voltage signals collected by the data acquisition card, and obtains two numerical signals through analog-to-digital conversion, namely the first numerical signal s_1 and the second numerical signal s_2;

[0012] Step 2: Perform normalization and median filtering on the first numerical signal s_1 and the second numerical signal s_2 obtained in Step 1 in sequence, and then perform dual-endpoint detection to obtain the endpoint offset Dev, and the first effective signal g_1 and the second effective signal g_2 corresponding to the two signals based on the theoretical maximum time delay TDMax;

[0013] Step 3: Perform iterative variational mode decomposition on the first effective signal g_1 and the second effective signal g_2 obtained in Step 2;

[0014] Step 4: Perform envelope extraction and generalized cross-correlation time-delay estimation on the decomposed first-layer sub-signals imf_1 and imf_2 obtained in Step 3 to obtain the time-delay estimation value;

[0015] Step 5: Add the time-delay estimation value obtained in Step 4 to the endpoint offset Dev obtained in Step 2 to obtain the final time-delay estimation value.

[0016] Further, the specific steps of Step 2 include:

[0017] S21: Synchronously perform normalization and median filtering on the two original signals in sequence to obtain numerical signals ss_1 and ss_2, and obtain the high-pass filtered results Sss_1 and Sss_2 by taking the difference between the normalized numerical signal and the median filtered signal;

[0018] S22: Preset a threshold Thr, and respectively find the positions Pos_1 and Pos_2 of the first value greater than the threshold Thr in Sss_1 and Sss_2. Subtract the two positions Pos_1 and Pos_2 and denote it as the endpoint offset Dev;

[0019] S23: Judge whether the absolute value of the endpoint offset Dev is greater than 1.2 - 2 times the theoretical maximum time delay TDMax. If it is greater, execute Step S24; otherwise, execute Step S25;

[0020] S24: In the first numerical signal s_1 and the second numerical signal s_2, extract the first effective signal g_1 and the second effective signal g_2 with a preset time length from 1.2 times the theoretical maximum time delay TDMax before Pos_1, and perform normalization again; let the endpoint offset Dev = 0;

[0021] S25: In the first numerical signal s_1, extract the first effective signal g_1 with a preset time length from 1.2 times the theoretical maximum time delay TDMax before Pos_1, and perform normalization again; in the second numerical signal s_2, extract the second effective signal g_2 with a preset time length from 1.2 times the theoretical maximum time delay TDMax before Pos_2, and perform normalization again, and output the endpoint offset Dev.

[0022] Further, the theoretical maximum time delay TDMax is obtained by the following formula:

[0023]

[0024] where n is the refractive index of the sensing optical fiber, c is the speed of light in vacuum, and L is the total length of the sensing optical fiber;

[0025] Further, the specific steps of Step 3 include:

[0026] S31: Calculate the 99% power occupancy bandwidths B_1 and B_2 of the two effective signals g_1 and g_2;

[0027] S32: Preset the decomposition layer number K = 2, initialize the quadratic penalty factor α, and set the iterative update step size UL;

[0028] S33: Perform variational mode decomposition and determine whether to complete the iteration according to Condition 1 and Condition 2;

[0029] Among them, Condition 1: The center frequency of the first layer component of the variational mode decomposition is close to the bandwidth, and the center frequencies of the remaining modes are within the bandwidth;

[0030] Condition 2: The center frequency of the first layer mode of the variational mode decomposition is 1.2 times that of the second layer mode;

[0031] Judge whether both Condition 1 and Condition 2 are satisfied. If so, perform iterative variational mode decomposition and output the first layer sub-signals imf_1 and imf_2 of the decomposition of the two effective signals;

[0032] Otherwise, update the quadratic penalty factor α according to the following formula

[0033]

[0034] When one of the center frequencies of the first layer components of the variational mode decomposition of the two effective signals g_1 and g_2 is much smaller than B_1 and B_2, execute Equation (2); when one of them is much larger than B_1 and B_2, execute Equation (1); when Condition 2 is not satisfied, execute Equation (1);

[0035] Update the quadratic penalty factor α and perform variational mode decomposition on the two effective signals g_1 and g_2 again;

[0036] Until both Condition 1 and Condition 2 are satisfied, complete the iterative variational mode decomposition and output the first layer sub-signals imf_1 and imf_2 of the decomposition of the two effective signals.

[0037] Furthermore, Step Four further includes:

[0038] S41: Perform Hilbert transform on the first layer sub-signals imf_1 and imf_2 obtained after decomposition in Step Four to extract the envelope signals, denoted as the first envelope signal z1 and the second envelope signal z2;

[0039] S42: Perform autocorrelation on the first envelope signal z1 to obtain R1, perform cross-correlation on the first envelope signal z1 and the second envelope signal z2 to obtain R2, and perform quadratic cross-correlation on R1 and R2 to obtain R3.

[0040] S43: Multiply R3 by the generalized weighting function G to obtain the generalized bispectral density, and then perform an inverse Fourier transform on the obtained generalized bispectral density to obtain the generalized cross-correlation function;

[0041] S44: Then perform peak detection to finally obtain the time delay estimation value TD_2.

[0042] Furthermore, step five also includes obtaining the actual disturbance point position according to the obtained final time delay estimation value by using the following formula:

[0043]

[0044] where x is the position of the disturbance point from the first optical coupler or the second optical coupler, n is the refractive index of the sensing optical fiber, L is the total length of the sensing optical fiber, τ is the final time delay estimation value between the two signals obtained in step five, and c is the speed of light in vacuum.

[0045] Compared with the prior art, the beneficial effects of the present invention are:

[0046] The present invention proposes a distributed optical fiber sensing and positioning method based on double-endpoint detection and iterative variational mode decomposition. Most of the current positioning methods adopt single-endpoint detection, time-frequency analysis methods with fixed parameters, and direct cross-correlation time delay estimation methods.

[0047] The method of single-endpoint detection makes the mismatch degree of the two intercepted signals reach the maximum when the disturbance behavior occurs at the head end and the tail end. The method of double-endpoint detection essentially borrows the simultaneous endpoint detection of the two signals to perform a time delay estimation with low accuracy, which greatly overcomes this mismatch degree.

[0048] The time-frequency analysis method with fixed parameters has relatively strict requirements for the changes brought by the disturbance behavior, generally resulting in the failure of locating low-frequency disturbance signals caused by weak disturbances. The iterative variational mode decomposition effectively captures the influence brought by the disturbance by iteratively updating its important parameter, the quadratic penalty factor α, which can effectively capture the disturbance information and provides a basis for realizing wide-band positioning.

[0049] Direct cross-correlation has the advantage of fast calculation, but its ability to resist noise is weak. Generalized cross-correlation improves the anti-noise performance through quadratic correlation and the generalized weighting function. In the case where the low-frequency disturbance signals caused by weak disturbances are aliased with environmental noise in both the time domain and the frequency domain and the signal-to-noise ratio is low, generalized cross-correlation is a better solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 is a schematic diagram of the principle of a distributed optical fiber sensing and positioning system based on the principle of asymmetric dual Mach-Zehnder interferometry;

[0051] Figure 2 Schematic diagram of a distributed optical fiber sensing broadband positioning method based on double-endpoint detection and iterative variational mode decomposition;

[0052] Figure 3 Flow chart of double-endpoint detection in step 2;

[0053] Figure 4 Flow chart of iterative variational mode decomposition in step 3;

[0054] Figure 5 Flow chart of envelope extraction and generalized cross-correlation time delay estimation algorithm in step 4;

[0055] Figure 6 Shows two effective signals intercepted after double-endpoint detection;

[0056] Figure 7 Shows two envelope signals extracted by this method for high-frequency disturbance signals;

[0057] Figure 8 Shows two envelope signals extracted by this method for low-frequency disturbance signals;

[0058] Figure 9 Shows the final time delay estimation peak of this method.

[0059] Figure 1 In:

[0060] 1: First light source 2: Second light source 3: First optical circulator

[0061] 4: Second optical circulator 5: First polarization controller 6 Second polarization controller

[0062] 7: First optical coupler 8: Second optical coupler 9: First dense wavelength division multiplexer

[0063] 10: Second dense wavelength division multiplexer 11: First photodetector

[0064] 12: Second photodetector 13: Data acquisition card 14: Industrial control computer

[0065] 15: Sensing optical fiber Specific implementation method

[0066] To make the objectives, technical solutions, beneficial effects, and remarkable improvements of the embodiments of the present invention clearer, the following describes the technical solutions in the embodiments of the present invention clearly and completely in conjunction with the accompanying drawings provided in the embodiments of the present invention. Obviously, all these described embodiments are only partial embodiments of the present invention, rather than all embodiments; based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0067] A distributed optical fiber sensing and positioning system based on an asymmetric dual Mach-Zehnder, the schematic diagram of its principle is as Figure 1 shown. Two beams of light emitted by the first light source 1 and the second light source 2 propagate in the clockwise and counterclockwise directions respectively in the sensing loop, and interfere at the coupler at the opposite end. It includes:

[0068] The first light source 1 and the second light source 2: The wavelengths of the two light sources are 1550.12 nm and 1550.92 nm respectively, and the power of both is 10 mw;

[0069] The first optical circulator 3 and the second optical circulator 4: used to transmit the light emitted by the light source to the object to be detected and collect the reflected signal light;

[0070] The first dense wavelength division multiplexer 9 and the second dense wavelength division multiplexer 10: The passband bandwidths of the first dense wavelength division multiplexer 9 and the second dense wavelength division multiplexer 10 are 1550.92 ± 0.2 nm and 1550.12 ± 0.2 nm respectively;

[0071] The first photodetector 11 and the second photodetector 12: indium gallium arsenide photodetector PD, receive the optical signal, perform photoelectric conversion and amplification;

[0072] The first fiber optic coupler 7 and the second optical coupler 8: 3 dB fiber optic coupler, used for 1:2 optical splitting;

[0073] The first polarization controller 5 and the second polarization controller 6: adjust the polarization states of the two interference signals to ensure the signal visibility;

[0074] The sensing optical fiber 15: The total length L is 90 km, of which 1 km is armored optical cable and the remaining 89 km is bare fiber, used to sense external disturbances and conduct optical signals;

[0075] The data acquisition card 13: The sampling frequency is 5 M / s, the sampling time is 0.3 s, and the sampling time corresponding to the effective signal is 0.005 s, used to collect the voltage signals of the two photodetectors and send them to the processing unit for processing;

[0076] Industrial control computer: Analyze and demodulate the interference signals received by the data acquisition card 13 to demodulate the amplitude-frequency information of the vibration signals applied to the sensing optical fiber.

[0077] Among them, the light emitted by the first light source 1 propagates along the path of the first optical circulator 3 - the first polarization controller 5 - the first optical coupler 7 - the sensing optical fiber 15 - the second optical coupler 8 - the second polarization controller 6 - the second optical circulator 4 - the second dense wavelength division multiplexer 10 - the second photodetector 12. The light emitted by the second light source 2 propagates along the path of the second optical circulator 4 - the second polarization controller 6 - the second optical coupler 8 - the sensing optical fiber 15 - the first optical coupler 7 - the first polarization controller 5 - the first optical circulator 3 - the first dense wavelength division multiplexer 9 - the first photodetector 11. The voltage signals detected by the first photodetector 11 and the second photodetector 12 are sent to the data acquisition card 13 for acquisition, and then sent to the industrial control computer 14 for signal processing.

[0078] As Figure 2 shown, for the distributed optical fiber sensing wideband positioning method based on double-endpoint detection and iterative variational mode decomposition of the distributed optical fiber sensing positioning system, this method respectively performs analog-to-digital conversion on the two signals collected by the two photodetectors of the distributed perturbation sensor based on the asymmetric dual Mach-Zehnder interference principle to obtain two digital quantity signals, divides the signals into frames, and sets the frame length to 0.3 seconds. When a certain frame of signal is determined to be a perturbation frame, three links are sequentially performed on the two signals: double-endpoint detection, iterative variational mode decomposition, envelope extraction and generalized cross-correlation time delay estimation.

[0079] Specifically, it includes:

[0080] Step 1: Based on the above optical fiber sensing positioning system, the industrial control computer 14 receives the two voltage signals collected by the data acquisition card 13, obtains two numerical signals through analog-to-digital conversion, and names them the first numerical signal s_1 and the second numerical signal s_2.

[0081] Step 2: As Figure 3 shown, first perform double-endpoint detection on the first numerical signal s_1 and the second numerical signal s_2 obtained in Step 1.

[0082] S21: Normalize and median filter the first numerical signal s_1 and the second numerical signal s_2 respectively to obtain numerical signals ss_1 and ss_2; subtract the normalized numerical signal from the median-filtered signal to obtain the results of high-pass filtering Sss_1 and Sss_2. The specific formulas are as follows:

[0083] Sss_1 = s_1 - ss_1, Sss_2 = s_2 - ss_2;

[0084] S22: Find the positions Pos_1 and Pos_2 of the first values greater than the preset threshold Thr in Sss_1 and Sss_2 respectively, and calculate the endpoint offset Dev = Pos_1 - Pos_2.

[0085] S23: Determine whether the absolute value of the obtained endpoint offset Dev is greater than 1.2 - 2 times the theoretical maximum time delay TDMax.

[0086] From the principle of the asymmetric dual Mach-Zehnder fiber optic vibration sensing system, the positioning formula can be obtained:

[0087]

[0088] where τ is the time delay between the two signals, n is the refractive index of the sensing fiber, c is the speed of light in vacuum, L is the total length of the sensing fiber, x is the perturbation position, D is the fiber dispersion coefficient, λ1 is the wavelength of the first light source 1, and λ2 is the wavelength of the second light source 2. Generally, the latter term can be ignored. Then when x = 0 or x = L, the absolute value of the time delay between the two signals reaches the maximum. At this time, the theoretical maximum time delay TDMax can be obtained from the following formula:

[0089] , the theoretical maximum time delay TDMax in the double-endpoint detection can be calculated according to the following formula:

[0090]

[0091] Take the refractive index n of the fiber as 1.47, the speed of light c in vacuum as 3×10^8 m / s, TDMax as 0.441 ms, and preset a threshold Thr. According to the sampling rate of 5 M / s, TDMax is 2205 sampling points.

[0092] According to the sampling rate of 5 M / s, the sampling time corresponding to the valid signal is 0.005 s, and the sampling points of the corresponding first valid signal g_1 and second valid signal g_2 are 25000.

[0093] S24: If the absolute value of the endpoint offset Dev is greater than 1.2 times the theoretical maximum time delay TDMax, that is, when the absolute value of the endpoint offset Dev is greater than 1.2 times the theoretical maximum time delay TDMax, the positions of the output first valid signal g_1 and second valid signal g_2 are both Pos_1 - (1.2 - 2) times TDMax, that is, output the first valid signal g_1 with a time length of 5 ms at the position 1.2 times TDMax before the Pos_1 position in the first numerical signal s_1 and the second numerical signal S_2, the second valid signal g_2, set the endpoint offset Dev to 0, and output the endpoint offset Dev. The practical significance of this step 4 is that when the double-endpoint detection fails, this method becomes single-endpoint detection to ensure the rationality of subsequent signal processing.

[0094] S25: That is, if the absolute value of the obtained endpoint offset Dev (i.e., Pos_1 - Pos_2) is less than 1.2 times the maximum theoretical time delay TDMax, the positions of the two effective signal interceptions are Pos_1 - 1.2 times TDMax and Pos_2 - 1.2 times TDMax respectively. The position of the first effective signal g_1 with an output time length of 5 ms is: at the position 1.2 times TDMax before the Pos_1 position in the first numerical signal s_1, and the position of the effective signal g_2 with an output time length of 5 ms is: at the position 1.2 times TDMax before the Pos_2 position in the second numerical signal s_2.

[0095] Step three: As Figure 4 shown, iterative variational mode decomposition.

[0096] S31: First, calculate the 99% power occupancy bandwidths B_1 and B_2 of the first effective signal g_1 and the second effective signal g_2 obtained in step two.

[0097] S32: According to the first bandwidth B_1 and the second bandwidth B_2, initialize the quadratic penalty factor α. The initialization of the 99% power occupancy bandwidth and the initialization of the quadratic penalty factor α are given by the following table:

[0098]

[0099] For example, when the first bandwidth B_1 is 45 kHz, the quadratic penalty factor α is initialized to 3000, the preset decomposition layer number K is 2, and the iterative update step size UL is 0.5.

[0100] S33: Perform variational mode decomposition, setting two important conditions, where:

[0101] Condition 1: The center frequency of the first layer component of the variational mode decomposition is close to the bandwidth, and the center frequencies of the remaining modes are within the bandwidth;

[0102] Condition 2: The center frequency of the first layer mode of the variational mode decomposition is 1.2 times that of the second layer mode;

[0103] Judge whether conditions 1 and 2 are satisfied. If they are not satisfied simultaneously, update the quadratic penalty factor α according to the following formula

[0104]

[0105] When one of the center frequencies of the first layer components of the variational mode decomposition of the two effective signals g_1 and g_2 is much less than B_1 and B_2, execute formula (2), when one of them is much greater than B_1 and B_2, execute formula (1), and when condition 2 is not satisfied, execute formula (1).

[0106] Update the secondary penalty factor α, and perform variational mode decomposition on the two effective signals g_1 and g_2 again;

[0107] Until both Condition 1 and Condition 2 are satisfied simultaneously, complete the iterative variational mode decomposition, and output the first-layer sub-signals imf_1 and imf_2 of the decomposition of the two effective signals.

[0108] Step Four: As Figure 5 shown, perform envelope extraction and generalized cross-correlation time-delay estimation.

[0109] S41: Perform Hilbert transform on the first-layer sub-signals imf_1 and imf_2 obtained after decomposition in Step Four to extract the envelope signals, denoted as the first envelope signal z1 and the second envelope signal z2;

[0110] S42: Perform autocorrelation on the first envelope signal z1 to obtain R1, perform cross-correlation on the first envelope signal z1 and the second envelope signal z2 to obtain R2, and perform quadratic cross-correlation on R1 and R2 to obtain R3.

[0111] S43: Multiply R3 by the generalized weighting function G to obtain the generalized quadratic cross-power spectrum; where the generalized weighting function uses Roth weighting, which can be expressed by the following formula

[0112]

[0113] In this example, it is the reciprocal of R1, which is the autocorrelation of the first envelope signal z1.

[0114] S44: Then perform inverse Fourier transform on the obtained generalized quadratic cross-power spectrum to obtain the generalized quadratic cross-correlation function;

[0115] S45: Then perform peak detection to finally obtain the time-delay estimation value TD_2.

[0116] Step Five: Add the time-delay estimation value TD_2 obtained in Step Four to the end-point offset Dev obtained in Step Two to obtain the final time-delay estimation value TD. According to the obtained final time-delay estimation value TD, the positioning formula can be obtained from the above fiber optic sensing positioning system and method as follows:

[0117]

[0118] where x is the position of the perturbation point from the second optical coupler 8, n is the refractive index of the sensing optical fiber, L is the total length of the sensing optical fiber, τ is the time delay between the two signals, that is, the final time-delay estimation value TD, and c is the speed of light in vacuum.

[0119] Verify according to the above distributed fiber optic sensing broadband positioning method. When the optical cable is tapped at 100 m of the optical cable and measured using the above distributed fiber optic sensing positioning system, as Figure 6As shown, after the double-endpoint detection, the first effective signal g_1 and the second effective signal g_2 are obtained. It can be seen that the double-endpoint detection reduces the difference between the signals for subsequent signal processing. Theoretically, if only one endpoint is detected, the high-frequency starting positions of the two intercepted effective signals should differ by 0.441 ms. After the double-endpoint detection, the intercepted effective signals have a higher correlation.

[0120] Figure 7 The envelope signals extracted from the two high-frequency disturbance signals are shown. Figure 8 The envelope signals extracted from the two low-frequency disturbance signals are not shown. The two envelope signals extracted after the iterative variational mode decomposition have extremely high similarity, which is more conducive to subsequent time-delay estimation.

[0121] The positioning value is determined by the deviation amount of the double-endpoint detection and the peak position of the time-delay estimation. Figure 9 The final time-delay estimation peak of the method for the actual signal is shown. Among them, the solid line corresponds to the higher-frequency signal, and the dashed line corresponds to the lower-frequency signal. Both show obvious peaks.

[0122] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some or all of the technical features. These modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention. The non-essential improvements, adjustments or replacements made by those skilled in the art according to the content of this specification all fall within the scope claimed by the present invention.

Claims

1. A distributed optical fiber sensing and positioning system based on an asymmetric dual Mach-Zehnder interferometer, comprising an optical fiber sensing module and a disturbance positioning module; the optical fiber sensing module includes a first light source, a second light source, an asymmetric dual Mach-Zehnder interferometer, a data acquisition card (13), and an industrial control computer (14); two beams of light emitted by the first light source (1) and the second light source (2) propagate along the clockwise and counterclockwise directions respectively in the sensing loop, and interfere at the couplers at the opposite ends in the interferometer to form two interference signals, and the data acquisition card (13) acquires the two interference signals, converts them into electrical signals, and sends them to the industrial control computer (14) for signal processing; characterized in that, The disturbance positioning module is used to obtain the positioning result, in which a dual-endpoint detection method based on median filtering is adopted for endpoint detection, iterative variational mode decomposition is used for extracting time-frequency variation features, and envelope extraction and generalized cross-correlation time-delay estimation algorithm are used for time-delay estimation and positioning.

2. The distributed optical fiber sensing and positioning system according to claim 1, wherein The optical path of the light emitted by the first light source is the first optical circulator (3) - the first polarization controller (5) - the asymmetric dual Mach-Zehnder interferometer - the second polarization controller (6) - the second optical circulator (4) - the second dense wavelength division multiplexer (10) - the second photodetector (12); The optical path of the light emitted by the second light source (2) is the second optical circulator (4) - the second polarization controller (6) - the asymmetric dual Mach-Zehnder interferometer - the first polarization controller (5) - the first optical circulator (3) - the first dense wavelength division multiplexer (9) - the first photodetector (11).

3. The distributed optical fiber sensing broadband positioning method of the distributed optical fiber sensing and positioning system according to claim 1, comprising: Step 1: Based on the optical fiber sensing and positioning system, the industrial control computer (14) receives two voltage signals acquired by the data acquisition card (13), and obtains two numerical signals through analog-to-digital conversion, namely the first numerical signal s_1 and the second numerical signal s_2; Step 2: Perform normalization and median filtering on the first numerical signal s_1 and the second numerical signal s_2 obtained in Step 1 in sequence, and then perform dual-endpoint detection to obtain the endpoint offset Dev, and based on the theoretical maximum time delay TDMax, obtain the first effective signal g_1 and the second effective signal g_2 corresponding to the two signals; Step 3: Perform iterative variational mode decomposition on the first effective signal g_1 and the second effective signal g_2 obtained in Step 2; Step 4: Perform envelope extraction and generalized cross-correlation time-delay estimation on the decomposed first-layer sub-signals imf_1 and imf_2 obtained in Step 3 to obtain the time-delay estimation value; Step 5: Add the time-delay estimation value obtained in Step 4 to the endpoint offset Dev obtained in Step 2 to obtain the final time-delay estimation value.

4. The distributed optical fiber sensing wideband positioning method according to claim 3, characterized in that, The said Step 2 further includes: S21: Synchronously perform normalization and median filtering on the two original signals in sequence to obtain the numerical signals ss_1 and ss_2, and obtain the high-pass filtered results Sss_1 and Sss_2 by subtracting the normalized numerical signal from the median-filtered signal; S22: Preset a threshold Thr, and respectively find the positions Pos_1 and Pos_2 of the first values greater than the threshold Thr in Sss_1 and Sss_2. Denote the subtraction of the two positions Pos_1 and Pos_2 as the endpoint offset Dev; S23: Determine whether the absolute value of the endpoint offset Dev is greater than 1.2 - 2 times the theoretical maximum value TDMax of the time delay. If it is greater, execute step S24; otherwise, execute step S25; S24: In the first numerical signal s_1 and the second numerical signal s_2, extract the first effective signal g_1 and the second effective signal g_2 of a preset time length from 1.2 times the theoretical maximum time delay TDMax before Pos_1, and normalize again; set the endpoint offset Dev = 0; S25: In the first numerical signal s_1, extract the first effective signal g_1 of a preset time length from 1.2 times the theoretical maximum time delay TDMax before Pos_1, and normalize again; in the second numerical signal s_2, extract the second effective signal g_2 of a preset time length from 1.2 times the theoretical maximum time delay TDMax before Pos_2, and normalize again, and output the endpoint offset Dev.

5. The distributed optical fiber sensing broadband positioning method according to claim 4, wherein The theoretical maximum value TDMax of the time delay is obtained by the following formula: where n is the refractive index of the sensing optical fiber, c is the speed of light in vacuum, and L is the total length of the sensing optical fiber.

6. The distributed optical fiber sensing wideband positioning method according to claim 3, wherein, Step three further includes: S31: Calculate the 99% power occupancy bandwidths B_1 and B_2 of the two-way effective signals g_1 and g_2; S32: Preset the decomposition layer number K = 2, initialize the quadratic penalty factor α, and set the iterative update step size UL; S33: Perform variational mode decomposition and determine whether to complete the iteration according to condition 1 and condition 2; where condition 1: The central frequency of the first layer component of the variational mode decomposition is close to the bandwidth, and the central frequencies of the remaining modes are within the bandwidth; Condition 2: The central frequency of the first layer mode of the variational mode decomposition is 1.2 times that of the second layer mode; Determine whether both condition 1 and condition 2 are satisfied. If so, perform iterative variational mode decomposition and output the first layer sub-signals imf_1 and imf_2 of the two-way effective signal decomposition; Otherwise, update the quadratic penalty factor α according to the following formula When one of the central frequencies of the first layer components of the variational mode decomposition of the two-way effective signals g_1 and g_2 is much smaller than B_1 and B_2, execute formula (2); when one of them is much larger than B_1 and B_2, execute formula (1); when condition 2 is not satisfied, execute formula (1); Update the quadratic penalty factor α, and perform variational mode decomposition on the two-way effective signals g_1 and g_2 again; Until both condition 1 and condition 2 are satisfied, complete the iterative variational mode decomposition, and output the first layer sub-signals imf_1 and imf_2 of the two-way effective signal decomposition.

7. The distributed optical fiber sensing broadband positioning method according to claim 3, characterized in that Step four further includes: S41: Perform Hilbert transform on the first layer sub-signals imf_1 and imf_2 obtained after decomposition in step four to extract the envelope signals, denoted as the first envelope signal z1 and the second envelope signal z2; S42: Perform autocorrelation on the first envelope signal z1 to obtain R1, perform cross-correlation on the first envelope signal z1 and the second envelope signal z2 to obtain R2, and perform second-order cross-correlation on R1 and R2 to obtain R3; S43: Multiply R3 by the generalized weighting function G to obtain the generalized second-order cross-power spectrum, and then perform inverse Fourier transform on the obtained generalized second-order cross-power spectrum to obtain the generalized second-order cross-correlation function; S44: Then perform peak detection to finally obtain the time delay estimation value TD_2.

8. The distributed optical fiber sensing wideband positioning method according to claim 3, characterized in that, Step five further includes obtaining the actual disturbance point position according to the obtained final time delay estimation value by using the following formula: where x is the position of the disturbance point from the first optical coupler or the second optical coupler, n is the refractive index of the sensing optical fiber, L is the total length of the sensing optical fiber, τ is the final time delay estimation value obtained in step five, and c is the speed of light in vacuum.

Images