Fiber seismic wave velocity monitoring method, system and medium using vehicle vibration signals

Abstract

The application discloses a fiber seismic wave velocity monitoring method and system using vehicle vibration signals, and a medium, relates to the technical fields of fiber sensing and vehicle-road cooperation, and comprises the following steps: step one, detecting vehicle signals recorded by each fiber channel; step two, sequentially calculating the cross spectrum, coherence and weight after S transforming the vehicle signals, and calculating the delay by weighting; step three, repeating step two to obtain a delay sequence set corresponding to all vehicle signals; and step four, taking the sign inversion of the travel time difference obtained by bidirectional propagation of the vehicle signals between two fiber channels as prior information, screening the delay sequence set, calculating the delay change of each vehicle signal between the two fiber channels, and thus obtaining a velocity change sequence; the monitoring method realizes high-resolution time-varying monitoring of the wave velocity of underground media.

Description

Technical Field

[0001] This invention relates to the fields of fiber optic sensing and vehicle-road cooperative technology, and in particular to a fiber optic seismic wave velocity monitoring method, system, and medium utilizing vehicle vibration signals. Background Technology

[0002] Subsurface wave velocity variation is an important indicator in geophysics characterizing the dynamic evolution of the physical state of subsurface media, and is widely used in fields such as geological disaster early warning (e.g., landslides, subsidence), groundwater monitoring, oil and gas reservoir monitoring, and urban underground space safety. Traditional methods mainly repeat seismic events and active sources, but repeating earthquakes is infrequent and their spatial locations are not fixed. Using artificial seismic sources (e.g., blasting, air guns) is costly and also cannot achieve large-scale continuous monitoring.

[0003] The most widely used continuous monitoring technique at present is the background noise wakewave interferometry method. The implementation scheme of this technique is typically as follows:

[0004] (1) Use a seismograph to continuously record background seismic noise signals generated by nature, human activities, etc.

[0005] (2) Cross-correlate the records from the two stations to extract the signal containing multiple scattered waves, i.e. the wake wave part.

[0006] (3) The relative change of medium wave velocity is calculated by calculating the phase change of the wake signal (usually using the stretching method, moving window cross spectrum method and wavelet cross spectrum method).

[0007] The disadvantages of the above-mentioned background noise wake wave interferometry method are:

[0008] 1) This method requires highly sensitive monitoring equipment; otherwise, it is difficult to reliably recover the weak scattered coma wave signal. However, compared to professional seismographs, DAS typically has a lower signal-to-noise ratio. The coma wave signal itself has weak energy and is more difficult to record effectively by fiber optic systems compared to the direct waves generated by vehicle vibrations.

[0009] 2) It is necessary to ensure that the noise source is evenly distributed and that the noise signal has sufficient scattering. However, in practical applications, especially when optical fibers are laid along highways, the recorded seismic wave signals do not meet the above conditions, and the noise source mainly comes from vehicle vibrations in a specific direction (along the highway).

[0010] 3) Low spatiotemporal resolution; a reliable cross-correlation function can only be obtained by using noise signals of a certain duration and the superposition of cross-correlation results of multiple channels, thus sacrificing spatiotemporal resolution. Summary of the Invention

[0011] Based on the technical problems existing in the background technology, this invention proposes a method, system and medium for monitoring seismic wave velocity using vehicle vibration signals, so as to realize high-resolution time-varying monitoring of underground medium wave velocity.

[0012] The fiber optic seismic wave velocity monitoring method utilizing vehicle vibration signals proposed in this invention includes:

[0013] Step 1: Detect the vehicle signal recorded in each fiber optic channel;

[0014] Step 2: After performing S-transform on the vehicle signal, calculate the cross spectrum, coherence, and weights in sequence, and calculate the delay accordingly.

[0015] Step 3: Repeat step 2 to obtain the set of delay sequences corresponding to all vehicle signals;

[0016] Step 4: Using the sign reversal of the travel time difference obtained from the bidirectional propagation of the vehicle signal between the two optical fiber channels as prior information, the delay sequence set is filtered to calculate the delay change of the vehicle signal between the two optical fiber channels, thereby obtaining the speed change sequence.

[0017] Furthermore, in step one, detecting the vehicle signal recorded by each fiber optic channel specifically involves:

[0018] From the The envelope template of the vehicle signal is calculated by selecting a portion of the vehicle vibration signal from the records of each fiber optic channel.

[0019] Set the search window and step size, with the envelope template at the 1st... The system performs a matching search within the records of each fiber channel, and selects a search window whose correlation coefficient between the envelope template and the envelope of the searched vehicle signal is greater than a set threshold as the vehicle signal window, and then corrects the vehicle signal window.

[0020] Using the corrected vehicle signal window from the first The fiber optic channel and from the first Vehicle signals were extracted from the records of each fiber optic channel;

[0021] Perform a Fast Fourier Transform on the captured vehicle signal to ensure the frequency range under study. Vehicle signal present. These are the lower and upper frequency limits, respectively.

[0022] Furthermore, the aforementioned from the first The envelope template of the vehicle signal is calculated by selecting a portion of the vehicle vibration signal from the records of each fiber optic channel. Specifically:

[0023] From the M typical vehicle vibration signals are selected from the records of each fiber optic channel, with a duration of T1, where M is an integer;

[0024] The envelope of the selected vehicle vibration signal is calculated and averaged using Hilbert transform to obtain the envelope template of the vehicle signal.

[0025] Furthermore, the calibration of the vehicle signal window specifically involves:

[0026] The cross-correlation function is used to correct the deviation between the envelope and the envelope template of the vehicle signal in the vehicle signal window. The maximum value of the cross-correlation function is taken as the deviation time shift to correct the vehicle signal window.

[0027] Furthermore, the use of the corrected vehicle signal window from the first The fiber optic channel and from the first Vehicle signals were extracted from the records of each fiber optic channel, specifically as follows:

[0028] Using the corrected vehicle signal window from the first The vehicle signal set is obtained by re-matching and searching the records of each fiber optic channel. , The number of vehicle signals selected. For the first The first fiber optic channel One vehicle signal;

[0029] Arranged in chronological order of vehicle signal occurrence, and synchronously utilizing the corrected vehicle signal window from the [number]th [signal]... Vehicle signal sets were intercepted from the records of each fiber optic channel. , For the first The first fiber optic channel Vehicle signal.

[0030] Furthermore, the step of using the sign reversal of the travel time difference obtained from the bidirectional propagation of the vehicle signal between the two optical fiber channels as prior information to filter the set of delay sequences specifically involves:

[0031] Set the vehicle signal delay sequence that has not undergone sign inversion as an empty vector.

[0032] Furthermore, the calculation of the delay change of each vehicle signal between the two fiber optic channels specifically involves:

[0033] Calculate each time window All vehicle signals on the road are subject to time difference. The standard deviation of the delay is used to select time windows with stable delay.

[0034] Regarding the first Each vehicle signal will stabilize within the time window. Take the absolute values ​​and average them to get the first... The fiber optic channel and from the first Between fiber optic channels and within a frequency range of Time delay ,in, These are the lower frequency limit and the upper frequency limit, respectively.

[0035] Calculate the delay change:

[0036] ;

[0037] in, It is a delay sequence calculated by processing all vehicle signals sequentially. for The mean.

[0038] Furthermore, the calculation of the delay change of each vehicle signal between the two fiber optic channels to obtain the speed change sequence is specifically as follows:

[0039] Based on the negative correlation between velocity change and time delay change, the velocity change sequence is calculated using the time delay change:

[0040] ;

[0041] in, The velocity of the underground medium between the two fiber optic channels.

[0042] A computer system includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the method described above.

[0043] A computer-readable storage medium storing a plurality of classification programs, the plurality of classification programs being invoked by a processor to execute the method described above.

[0044] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as ROM, RAM, magnetic disk, or optical disk.

[0045] The advantages of the fiber optic seismic wave velocity monitoring method, system, and medium utilizing vehicle vibration signals provided by this invention are as follows:

[0046] Actively utilize vehicle signals as a reliable vibration source: abandon the traditional reliance on a uniform background noise field, and explicitly regard the vibration signals of vehicles along the road as a high-quality source with a fixed orientation and a clear propagation path. Each vehicle event is equivalent to a controllable experiment, avoiding the problems caused by non-uniform noise sources.

[0047] Simplified data processing workflow: The travel time difference between two fiber optic channels is directly calculated through time-frequency analysis (S-transform), eliminating the need to calculate the cross-correlation function, thus reducing computational complexity and improving processing efficiency.

[0048] High spatiotemporal resolution monitoring: Utilizing vehicle signals with high signal-to-noise ratio recorded by two fiber optic channels, reliable inter-channel travel time information can be obtained for single or a few vehicle events, enabling near real-time (e.g., hourly) and high spatial resolution (depending on the distance between the two channels) monitoring of underground medium wave velocity changes. Attached Figure Description

[0049] Figure 1 This is a schematic diagram of the process of the present invention;

[0050] Figure 2 For fiber optic channels Correspondence curve between rainfall and precipitation; Detailed Implementation

[0051] The technical solution of the present invention will now be described in detail through specific embodiments. Many specific details are set forth in the following description to provide a thorough understanding of the invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0052] like Figure 1 and Figure 2 As shown, the fiber optic seismic wave velocity monitoring method using vehicle vibration signals proposed in this invention includes:

[0053] Step 1: Detect the vehicle signal recorded in each fiber optic channel;

[0054] Step 2: After performing S-transform on the vehicle signal, calculate the cross spectrum, coherence, and weights in sequence, and calculate the delay accordingly.

[0055] Step 3: Repeat step 2 to obtain the set of delay sequences corresponding to all vehicle signals;

[0056] Step 4: Using the sign reversal of the travel time difference obtained from the bidirectional propagation of the vehicle signal between the two optical fiber channels as prior information, the delay sequence set is filtered to calculate the delay change of the vehicle signal between the two optical fiber channels, thereby obtaining the speed change sequence.

[0057] Based on steps one through four, this embodiment has the following advantages:

[0058] Actively utilize vehicle signals as a reliable vibration source: abandon the traditional reliance on a uniform background noise field, and explicitly regard the vibration signals of vehicles along the road as a high-quality source with a fixed orientation and a clear propagation path. Each vehicle event is equivalent to a controllable experiment, avoiding the problems caused by non-uniform noise sources.

[0059] Simplified data processing workflow: The travel time difference between two fiber optic channels is directly calculated through time-frequency analysis (S-transform), eliminating the need to calculate the cross-correlation function, thus reducing computational complexity and improving processing efficiency.

[0060] High spatiotemporal resolution monitoring: Utilizing vehicle signals with high signal-to-noise ratio recorded by two optical fiber channels, reliable inter-track travel time information can be obtained for single or a few vehicle events, enabling near real-time (e.g., hourly) and high spatial resolution (depending on the distance between the two optical fiber channels) monitoring of underground medium wave velocity changes.

[0061] Clear depth sensitivity: Based on the dispersion characteristics of surface waves, by analyzing the travel time changes of different frequency components, the wave velocity changes of the medium at different depths can be inferred semi-quantitatively, thus improving the depth resolution capability of the monitoring.

[0062] The core of this embodiment lies in directly calculating the seismic surface wave in the fiber optic channel based on the vehicle vibration signal recorded by DAS using time-frequency analysis technology. and The time difference between them enables high-resolution time-varying monitoring of underground media wave velocity, among which, and For the first One fiber optic channel Each fiber channel can be an adjacent channel or a non-adjacent channel. In this embodiment, it is preferred that the two are adjacent channels. DAS (Distributed Acoustic Sensing) is a distributed fiber optic sensing technology.

[0063] In addition, the fiber optic channel in this embodiment refers to a virtual segment of the optical fiber, representing different spatial locations.

[0064] In one embodiment, step one, detecting the vehicle signal recorded by each fiber optic channel, specifically involves:

[0065] (a1) Sensor fiber data is transmitted through the sensor fiber laid on both sides of the road and collected by the data acquisition card. The sensor fiber data is then preprocessed by downsampling, removing the mean and removing the trend.

[0066] (a2) From the first Fiber optic channels Calculate the envelope template of the vehicle signal by selecting a portion of the vehicle vibration signal from the DAS record. ;

[0067] This embodiment is from M typical vehicle vibration signals with duration T1 are selected from the records, where M is an integer; the envelope of the selected vehicle vibration signals is calculated using Hilbert transform and averaged to obtain the envelope template of the vehicle signals.

[0068] For example, from Ten typical vehicle vibration signals, each lasting 10 seconds, were selected from the records. The envelopes of these signals were calculated using Hilbert transform and averaged to obtain the vehicle signal envelope template. Among them, the characteristic of vehicle signals is that the amplitude gradually increases and then decreases. If the vehicle is traveling at a constant speed and the road has no obvious curves, the envelope of the vehicle signal will be approximately symmetrical about the center. Ten typical vehicle vibration signals were then selected.

[0069] (a3) Set the search window and step size (e.g., search window is 10 seconds, step size is 1 second), and use the envelope template in the first... The system performs a matching search within the records of each fiber channel, and selects the search window whose correlation coefficient between the envelope template and the envelope of the searched vehicle signal is greater than a set threshold (e.g., the set threshold is 0.65) as the vehicle signal window, and then corrects the vehicle signal window.

[0070] In the process of calibrating the vehicle signal window, the cross-correlation function is used to correct the deviation between the envelope and the envelope template of the vehicle signal in the vehicle signal window, and the maximum value of the cross-correlation function is taken as the deviation time shift to calibrate the vehicle signal window.

[0071] The cross-correlation function is:

[0072] ；

[0073] in, For envelope template, It is a cross-correlation function. For time delay, For the envelope of the vehicle signal, This is a time-based sliding window, not an absolute time. For example, if the duration of a vehicle signal is 10 seconds, then... It is a discrete time point variable ranging from 0 to 10 seconds.

[0074] Next, we need to find the peak position of the cross-correlation function to obtain the accurate bias time shift. :

[0075] .

[0076] (a4) Using the corrected vehicle signal window from the first The fiber optic channel and from the first Vehicle signals were extracted from the records of each fiber optic channel;

[0077] After calibration, the retrieved vehicle signals are considered to reflect the same or opposite (vehicles approaching from the opposite direction) motion. The calibrated vehicle signal window is then used to... Re-match and search within the records to obtain the vehicle signal set. , The number of vehicle signals selected. for The Middle One vehicle signal. Arranged in chronological order of their occurrence, and synchronously utilizing the corrected vehicle signal window from... Vehicle signal sets were intercepted from the records. , for The Middle Vehicle signal.

[0078] (a5) Perform a Fast Fourier Transform on the captured vehicle signal to ensure the frequency range under study. Vehicle signal present. These are the lower and upper frequency limits, respectively.

[0079] From (a1) to (a5), the following can be clearly stated:

[0080] Vehicle signals: refer to data from vibration sensors (such as...) or The vehicle signal windows extracted from the data are identified as containing vibration characteristics generated when a vehicle passes by. Specifically, the vehicle signals are discrete signal segments obtained after Hilbert transform envelope extraction, template matching, and bias correction, for example... and ,in Indicates the first The vehicle signal window also corresponds to the first vehicle signal window. The vehicle signals are processed using Hilbert transform to better filter the vehicle signal window. The vehicle signals extracted from the corresponding fiber optic channels using the corrected vehicle signal window are the original vibration signals without transformation.

[0081] Vehicle events refer to physical events in which a vehicle actually passes through sensors, i.e., the behavior of a vehicle passing through a monitored location at a specific point in time or within a time period. Each vehicle event corresponds to a vehicle signal window, and the occurrence time of the vehicle event is determined by the time and location of the vehicle signal window.

[0082] The specific relationships between vehicle signals and vehicle events are shown in (b1) to (b4):

[0083] (b1) Correspondence: Each vehicle signal window represents a vehicle event. That is, when the system detects a vehicle signal (i.e., the correlation coefficient between the envelope and the envelope template is greater than a set threshold), it considers a vehicle event to have occurred. The vehicle signal is the representation of the vehicle event in the sensor data.

[0084] (b2) Time sequence: Vehicle events are arranged in chronological order of occurrence, which is directly determined by the time sequence of the vehicle signal windows. That is, By index The arrangement reflects the time sequence of vehicle events.

[0085] (b3) Multi-channel synchronization: The same vehicle event will be synchronized across multiple fiber optic channels (e.g., and Vehicle signals are generated simultaneously from [the source]. For example, from [the source]. Extracted from and from Extracted from Since they correspond to the same vehicle event, vehicle signals are used to synchronously identify vehicle events across multiple channels.

[0086] (b4) Deviation correction: Time delay correction is performed on the envelope and envelope template of the vehicle signal using cross-correlation technology. This ensures that the vehicle signal is aligned with the envelope template.

[0087] Since all vehicle signals within the same sliding window need to be processed subsequently, deviation correction is used to ensure that each vehicle signal reflects the same or opposite (vehicle approaching from the opposite direction) motion process as much as possible. For example, the exact time that the vehicle passes the detector (channel) is recorded at the very center of each vehicle signal window.

[0088] In one embodiment, step two involves performing an S-transform on the vehicle signal and then sequentially calculating the cross-spectrum, coherence, and weights, based on which the delay is calculated using weighted averages, specifically as (c1) to (c3):

[0089] (c1) S-transform;

[0090] Regarding the first The first vehicle incident (i.e., the first (each vehicle signal), respectively for as well as Perform S-transform:

[0091] ;

[0092] ;

[0093] in, and Yes as well as Vehicle signal after S-transformation For time sliding window, It is a Gaussian window function. For being in The frequency between This is the integral variable that slides along the time axis.

[0094] in The calculation formula is as follows:

[0095] ;

[0096] in, The width of the adjustable Gaussian window.

[0097] (c2) Calculation of cross-spectrum, coherence, and weights;

[0098] calculate and cross spectrum between :

[0099] ;

[0100] in, This indicates the complex conjugate. Subsequently, to reduce random fluctuations, a two-dimensional smoothing (smoothing in both the frequency and time directions) is performed on the cross-spectrum and power spectrum to obtain the cross-spectrum. , No. Power spectrum of each fiber channel , No. Power spectrum of each fiber channel .

[0101] Next, the coherence and weights in the time and frequency domains are calculated to delay the subsequent weighted calculation.

[0102] Coherence:

[0103] ;

[0104] in, For signal and Coherence in the time-frequency domain.

[0105] Coherence reflects the similarity of two signals within the vehicle signal window; only points with coherence greater than a set coherence threshold are selected for subsequent calculations. Furthermore, lower energy time-frequency points often have lower signal-to-noise ratios, leading to decreased reliability of the calculated delay; therefore, cross-spectral amplitude is chosen for weighting.

[0106] ;

[0107] in, These are the weighted values.

[0108] (c3) Phase processing and delay calculation;

[0109] Given the same time delay, the phase of the vehicle signal is approximately linear with respect to frequency. (Regarding the cross-spectrum...) Its phase information (Right now and The phase difference is:

[0110] ;

[0111] For each time sliding window (i.e. each ), in the frequency Above, phase It can be represented as:

[0112] ;

[0113] in The delay is the value to be solved. The delay is calculated using the weighted least squares method:

[0114] ;

[0115] in, The number of frequency points. For each time window... If the weight is 0, the key point The number is greater than If the time sliding window is not found to be reliable, then the fitting reliability is considered insufficient. Set to null.

[0116] In one embodiment, step four involves using the sign reversal of the travel time difference obtained from the bidirectional propagation of the vehicle signal between the two fiber optic channels as prior information to filter the set of delay sequences, thereby calculating the delay change of the vehicle signal between the two fiber optic channels and obtaining the speed change sequence. Specifically:

[0117] By repeating step two above for all vehicle events, a set of delay sequences corresponding to all vehicle signals can be obtained. , For the first The time difference corresponding to each vehicle signal This represents the number of vehicle signals selected. For a fixed pair of channels... and Vehicle signal from spread to and from spread to Measured travel time difference A sign inversion will occur. Based on this prior information, this embodiment applies the measured set of delay sequences. Further filtering involves setting vehicle event delay sequences without sign reversal as empty vectors. This avoids introducing erroneous delay sequences due to potential vehicle signal recognition errors or poor vehicle signal quality, which could then affect the calculation of subsequent speed change sequences.

[0118] Finally, calculate each time window. All vehicle signals on (i.e.) (Each vehicle signal) corresponds to a different time zone. Based on the standard deviation, select time sliding windows with stable delays (standard deviation less than a threshold). For the first... Each vehicle signal will be within a stable time window. The channel can be obtained by averaging the absolute values. and Between, within the frequency band Time delay .

[0119] Then we can obtain the time delay changes:

[0120] ;

[0121] in, It is a delay sequence calculated by processing all vehicle signals sequentially. The variable is ,For example For the first The delay is calculated for each vehicle (arranged in chronological order). for The mean, which is the average delay calculated from all vehicle signals.

[0122] It is understood that in this embodiment... For time sliding window, This represents the travel time difference between the two fiber optic channels within the time window.

[0123] Based on the negative correlation between velocity change and time delay change, the velocity change sequence is calculated using the time delay change:

[0124] ;

[0125] in, The velocity of the underground medium between the two fiber optic channels.

[0126] Since the delay measurement using a single vehicle event has a large error, averaging all speed changes over an hour can yield hourly resolution speed change results.

[0127] like Figure 2 As shown, based on The comparison chart of (two-day smoothing) and rainfall shows good consistency. Increased rainfall leads to a decrease in seismic wave velocity in the medium, indicating that the results obtained in this embodiment... It accurately reflects the time-varying information of the underground medium and can be used for real-time monitoring.

[0128] This embodiment has the following advantages:

[0129] 1) Independent of Uniform Noise Field: Existing technologies require noise sources to be uniformly distributed and have strong scattering, which is difficult to meet in real-world scenarios such as roadsides. This embodiment abandons the reliance on an idealized, uniform background noise field and actively utilizes vehicle signals from vehicles traveling on the road, avoiding the limitation of uneven noise source distribution. The vibration source is more clearly defined and controllable. That is, vehicle signals are not regarded as interference, but as high-quality human activity sources with fixed orientations and clear propagation paths. Each vehicle event is equivalent to a controllable experiment with a known path, thus avoiding the problem of requiring uniform noise source distribution in background noise wake wave interferometry methods.

[0130] 2) Data processing is simpler and more efficient: It realizes direct and efficient calculation from the original vibration signal to the velocity change. That is, this embodiment directly calculates the travel time difference through time-frequency analysis. The steps are simple, the physical meaning is intuitive, and the computational resources and time consumption are reduced.

[0131] 3) Improved Spatiotemporal Resolution: Existing technologies require long-term noise recording (e.g., hours to days) and multi-channel result superposition to obtain reliable cross-correlation functions, resulting in low spatiotemporal resolution. This embodiment utilizes high signal-to-noise ratio vehicle signals, enabling reliable inter-channel travel time information to be obtained from the measurement of single or a few vehicle events. It supports near real-time inter-channel monitoring on hourly or even shorter timescales, making high-resolution near real-time monitoring of underground media possible.

[0132] 4) Provides a monitoring method with clear depth sensitivity: In the prior art, the depth sensitivity of wake interference is ambiguous and it is difficult to correlate frequency and depth; This embodiment is based on the surface wave dispersion characteristics (i.e., signals of different frequencies are sensitive to different depths). By analyzing the travel time changes of different frequencies, the wave velocity changes at different depths can be inferred semi-quantitatively. Moreover, the frequency of vehicle signals is generally higher, which can effectively detect the velocity changes of shallow media and improve the depth resolution capability of monitoring.

[0133] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A fiber optic seismic wave velocity monitoring method utilizing vehicle vibration signals, characterized in that, include: Step 1: Detect the vehicle signal recorded in each fiber optic channel; Step 2: After performing S-transform on the vehicle signal, calculate the cross spectrum, coherence, and weights in sequence, and calculate the delay accordingly. Step 3: Repeat step 2 to obtain the set of delay sequences corresponding to all vehicle signals; Step 4: Using the sign reversal of the travel time difference obtained from the bidirectional propagation of vehicle signals between the two fiber optic channels as prior information, the set of delay sequences is filtered to calculate the delay change of each vehicle signal between the two fiber optic channels, thereby obtaining the speed change sequence, specifically: Calculate each time window All vehicle signals on the road are subject to time difference. The standard deviation of the delay is used to select time windows with stable delay. Regarding the first Each vehicle signal will stabilize within the time window. Take the absolute values ​​and average them to get the first... The fiber optic channel and from the first Between fiber optic channels and within a frequency range of Time delay ,in, These are the lower frequency limit and the upper frequency limit, respectively. Calculate the delay change: in, It is a delay sequence calculated by processing all vehicle signals sequentially. for The mean; Based on the negative correlation between velocity change and time delay change, the velocity change sequence is calculated using the time delay change: in, The velocity of the underground medium between the two fiber optic channels.

2. The monitoring method according to claim 1, characterized in that, Step 1: Detect the vehicle signal recorded in each fiber optic channel, specifically as follows: From the The envelope template of the vehicle signal is calculated by selecting a portion of the vehicle vibration signal from the records of each fiber optic channel. Set the search window and step size, with the envelope template at the 1st... The system performs a matching search within the records of each fiber channel, and selects a search window whose correlation coefficient between the envelope template and the envelope of the searched vehicle signal is greater than a set threshold as the vehicle signal window, and then corrects the vehicle signal window. Using the corrected vehicle signal window from the first The fiber optic channel and from the first Vehicle signals were extracted from the records of each fiber optic channel; Perform a Fast Fourier Transform on the captured vehicle signal to ensure the frequency range under study. Vehicle signal present. These are the lower and upper frequency limits, respectively.

3. The monitoring method according to claim 2, characterized in that, The from the first The envelope template of the vehicle signal is calculated by selecting a portion of the vehicle vibration signal from the records of each fiber optic channel. Specifically: From the M typical vehicle vibration signals are selected from the records of each fiber optic channel, with a duration of T1, where M is an integer; The envelope of the selected vehicle vibration signal is calculated and averaged using Hilbert transform to obtain the envelope template of the vehicle signal.

4. The monitoring method according to claim 2, characterized in that, The specific steps for correcting the vehicle signal window are as follows: The cross-correlation function is used to correct the deviation between the envelope and the envelope template of the vehicle signal in the vehicle signal window. The maximum value of the cross-correlation function is taken as the deviation time shift to correct the vehicle signal window.

5. The monitoring method according to claim 2, characterized in that, The corrected vehicle signal window is used from the first The fiber optic channel and from the first Vehicle signals were extracted from the records of each fiber optic channel, specifically as follows: Using the corrected vehicle signal window from the first The vehicle signal set is obtained by re-matching and searching the records of each fiber optic channel. , The number of vehicle signals selected. For the first The first fiber optic channel One vehicle signal; Arranged in chronological order of vehicle signal occurrence, and synchronously utilizing the corrected vehicle signal window from the [number]th [signal]... Vehicle signal sets were intercepted from the records of each fiber optic channel. , For the first The first fiber optic channel Vehicle signal.

6. The monitoring method according to claim 1, characterized in that, The step of using the sign reversal of the travel time difference obtained from the bidirectional propagation of vehicle signals between two fiber optic channels as prior information to filter the set of delay sequences is as follows: Set the vehicle signal delay sequence that has not undergone sign inversion as an empty vector.

7. A computer system comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the method according to any one of claims 1-6.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a plurality of classification programs, which are used by a processor to execute the method as described in any one of claims 1-6.

Images