Time back projection earthquake positioning system and method based on distributed optical fiber sensing

Abstract

The invention discloses a time back projection earthquake positioning system and method based on distributed optical fiber sensing. The time back projection earthquake positioning system comprises a distributed optical fiber sensing module and a data processing module, wherein the distributed optical fiber sensing module comprises optical fibers arranged in the circumferential direction of an active source observation area and an optical fiber demodulator connected with the optical fibers, and the data processing module is in communication connection with the optical fiber demodulator; the optical fiber demodulator is used for acquiring optical fiber vibration signals of a plurality of preset monitoring points on the optical fiber in real time; and the data processing module is used for acquiring optical fiber vibration signals of a plurality of monitoring points in real time to obtain a dynamic space-time image, a seismic source position, a seismic moment and an energy characteristic. According to the invention, a full real-time flow of data acquisition and data processing of an earthquake in a micro scale is realized, and the earthquake can be quickly positioned.

Description

Technical Field

[0001] The present invention relates to the technical field of seismic positioning, and in particular, to a time-reversed projection seismic positioning system and method based on distributed optical fiber sensing. Background Art

[0002] Seismic positioning is a basic problem in seismology, and its main task is to accurately locate the source position of an earthquake. In the seismic positioning of large-scale regions, fast and reliable seismic positioning is the primary issue determining the success or failure of earthquake early warning. And seismic positioning in small and micro-scale regions is equally important, such as security monitoring, i.e., real-time and fast positioning of the vibration position within the monitoring area; microseismic monitoring, i.e., real-time positioning of the seismic positions generated during fracturing production and guiding the production according to the positioning positions at the same time. The traditional seismic positioning method is to use the detected seismic event to inversely calculate information such as the excitation time and source position of the earthquake by using the arrival time of its seismic phases.

[0003] However, the traditional seismic inverse position inversion method still has the following problems when applied to the real-time positioning of microseisms in small-scale regions: (1) The existing seismic positioning method is limited by the price of seismographs and it is difficult to deploy them at high density to cover a good azimuth angle, so the positioning accuracy is difficult to be very high; (2) During the multi-source excitation process, whether it is manual identification or using AI algorithms, it is difficult to correlate the seismic phases between each signal, so it is very difficult to locate the simultaneously excited sources; (3) Since the data collected by traditional seismographs are stored in the instrument's own storage, it is necessary to download the data after detection and it cannot be processed quasi-real-time. Summary of the Invention

[0004] To solve the technical problems in the background art, the present invention proposes a time-reversed projection seismic positioning system and method based on distributed optical fiber sensing.

[0005] A time-reversed projection seismic positioning system based on distributed optical fiber sensing proposed by the present invention includes: a distributed optical fiber sensing module and a data processing module; wherein, the distributed optical fiber sensing module includes an optical fiber arranged circumferentially along the active source observation area and an optical fiber demodulator connected to the optical fiber, and the data processing module is communicatively connected to the optical fiber demodulator;

[0006] The optical fiber demodulator is used to obtain the optical fiber vibration signals of a plurality of preset monitoring points on the optical fiber in real time; wherein, the duration of the optical fiber vibration signal is a preset duration;

[0007] The data processing module is used to obtain the optical fiber vibration signals of a plurality of monitoring points obtained by the optical fiber demodulator in real time;

[0008] Construct a velocity model;

[0009] Divide the active source observation area into a plurality of grid cells; wherein, each grid cell represents a possible source rupture point;

[0010] Using the velocity model, calculate the theoretical travel time from each grid cell to each monitoring point when the grid cell is regarded as a single seismic source;

[0011] According to the theoretical travel time from each grid cell to each monitoring point when the grid cell is regarded as a single seismic source, delay and translate the fiber optic vibration signals collected at each monitoring point when each grid cell is regarded as a single seismic source, to obtain the intermediate fiber optic vibration signals collected at each monitoring point when each grid cell is regarded as a single seismic source;

[0012] Weightedly superimpose the intermediate fiber optic vibration signals collected at each monitoring point when each grid cell is regarded as a single seismic source, to obtain the superimposed value of the fiber optic vibration signals when each grid cell is regarded as a single seismic source;

[0013] Using the back-projection method, obtain the dynamic spatio-temporal image based on the superimposed value of the fiber optic vibration signals when each grid cell is regarded as a single seismic source;

[0014] Perform seismic source identification on the dynamic spatio-temporal image to obtain the seismic source location, earthquake occurrence time, and energy characteristics.

[0015] Preferably, the data processing module includes multiple GPU servers, and the multiple GPU servers are used for parallel computing.

[0016] Preferably, it further includes network attached storage and a storage server. The fiber optic demodulator is respectively connected to the data processing module and the network attached storage, and the data processing module is connected to the storage server.

[0017] Preferably, after using the velocity model to calculate the theoretical travel time from each grid cell to each monitoring point when the grid cell is regarded as a single seismic source, it further includes:

[0018] Perform secondary correction on the theoretical travel time from each grid cell to each monitoring point when the grid cell is regarded as a single seismic source.

[0019] Preferably, during the delay and translation process, the delay and translation value for each monitoring point when each grid cell is regarded as a single seismic source is the theoretical travel time from the grid cell to each monitoring point when the grid cell is regarded as a single seismic source;

[0020] The phases of the intermediate fiber optic vibration signals collected at each monitoring point when each grid cell is regarded as a single seismic source are consistent, that is, the phases of the intermediate fiber optic vibration signals collected at the corresponding monitoring points of the same grid cell are consistent.

[0021] Preferably, before respectively performing delay and translation on the fiber optic vibration signals collected at each monitoring point according to the theoretical travel time from each grid cell to each monitoring point, it further includes:

[0022] Preprocess the fiber optic vibration signals collected at each monitoring point;

[0023] Among them, the preprocessing includes waveform annihilation and noise suppression, detrending and de - meaning, filtering, waveform amplitude compensation, absolute value processing, and envelope processing.

[0024] Preferably, using the back - projection method, a dynamic spatio - temporal image is obtained based on the superposition value of the fiber optic vibration signals when each grid cell is used as a single seismic source, specifically including:

[0025] Extract the superposition value of the fiber optic vibration signals at each time point within a preset time period for each grid cell from the superposition values of the fiber optic vibration signals when each grid cell is used as a single seismic source;

[0026] Based on the superposition value of the fiber optic vibration signals at each time point within a preset time period for each grid cell, obtain the superposition value of the fiber optic vibration signals of all grid cells at each time point;

[0027] Based on the superposition value of the fiber optic vibration signals of all grid cells at each time point, obtain the two - dimensional energy distribution map at each time point;

[0028] Connect the two - dimensional images at each time point in chronological order to obtain a dynamic spatio - temporal image.

[0029] Preferably, perform seismic source identification on the dynamic spatio - temporal image to obtain the seismic source location, earthquake occurrence time, and energy characteristics, specifically including:

[0030] Analyze the energy distribution in the dynamic spatio - temporal image, identify the location where the energy peak is located, and use the location where the energy peak is located as the possible seismic source rupture point;

[0031] Utilize the continuity characteristics of the energy distribution and the local extreme value method to cluster and separate the possible seismic source rupture points to identify and distinguish multiple simultaneously excited seismic events;

[0032] Determine the initial earthquake occurrence time of multiple seismic events according to the energy enhancement moments of multiple seismic events.

[0033] Preferably, after performing seismic source identification on the dynamic spatio - temporal image to obtain the seismic source location, earthquake occurrence time, and energy characteristics, it further includes:

[0034] Organize the information of the dynamic spatio - temporal image, seismic source location, earthquake occurrence time, and energy characteristics to form an event catalog.

[0035] In a second aspect, the present invention also proposes a time - reversed projection seismic positioning method based on distributed fiber optic sensing, which is applied to the time - reversed projection seismic positioning system based on distributed fiber optic sensing described in any item of the first aspect, including:

[0036] Obtain the optical fiber vibration signals of multiple monitoring points in real time; among them, the duration of the optical fiber vibration signals is a preset duration

[0037] Construct a velocity model;

[0038] Divide the active source observation area into multiple grid cells; among them, each grid cell represents a possible seismic source rupture point;

[0039] Respectively use the velocity model to calculate the theoretical travel times from each grid cell to each monitoring point when each grid cell is used as a single seismic source;

[0040] According to the theoretical travel times from each grid cell to each monitoring point when each grid cell is used as a single seismic source, respectively perform delay translation on the optical fiber vibration signals collected at each monitoring point when each grid cell is used as a single seismic source, and obtain the intermediate optical fiber vibration signals collected at each monitoring point when each grid cell is used as a single seismic source;

[0041] Respectively perform weighted superposition on the intermediate optical fiber vibration signals collected at each monitoring point when each grid cell is used as a single seismic source, and obtain the optical fiber vibration signal superposition value when each grid cell is used as a single seismic source;

[0042] Use the back-projection method to obtain a dynamic spatio-temporal image according to the optical fiber vibration signal superposition value when each grid cell is used as a single seismic source;

[0043] Perform seismic source identification on the dynamic spatio-temporal image to obtain the seismic source position, earthquake occurrence time, and energy characteristics

[0044] In the present invention, the proposed time-reversed projection seismic positioning system and method based on distributed optical fiber sensing utilize multiple monitoring points on the optical fiber in the distributed optical fiber sensing module to realize high-density and real-time backhaul acquisition of optical fiber vibration signals, and perform delay superposition and back-projection on the optical fiber vibration signals through the data processing module to obtain a dynamic spatio-temporal image, perform seismic source identification on the dynamic spatio-temporal image, obtain the seismic source position, earthquake occurrence time, and energy characteristics, realize the full real-time process of seismic data acquisition and data processing within a small scale, can overcome the deficiencies of traditional seismographs in spatial resolution, can quickly locate earthquakes, do not require the identification of seismic signals and the picking of travel times, greatly reduce the errors introduced by manual operations into the data, and can overcome the problem that traditional methods cannot locate simultaneously excited seismic sources. Description of the Drawings

[0045] Figure 1 It is a schematic flowchart of the time-reversed projection seismic positioning method based on distributed optical fiber sensing in an embodiment proposed by the present invention.

[0046] Figure 2Schematic diagram of a test site in an embodiment proposed by the present invention; where (a) represents the test site and (b) represents the container in the test site.

[0047] Figure 3 Schematic diagram of a test site and a dynamic spatio-temporal image in an embodiment proposed by the present invention; where (a) represents the test site and (b) represents the dynamic spatio-temporal image. Specific implementation manners

[0048] It should be noted that, without conflict, the implementation manners and features in the implementation manners of the present invention can be combined with each other. The present invention will be described in detail below with reference to the drawings and in combination with the implementation manners.

[0049] In a first aspect, the present invention also proposes a time reversal projection seismic positioning system based on distributed optical fiber sensing, including: a distributed optical fiber sensing module and a data processing module; where the distributed optical fiber sensing module includes an optical fiber arranged circumferentially along the active source observation area and an optical fiber demodulator connected to the optical fiber, and the data processing module is communicatively connected to the optical fiber demodulator;

[0050] The optical fiber demodulator is used to obtain in real time the optical fiber vibration signals of a plurality of preset monitoring points on the optical fiber; where the duration of the optical fiber vibration signal is a preset duration;

[0051] The data processing module is used to obtain in real time the optical fiber vibration signals of a plurality of monitoring points obtained by the optical fiber demodulator;

[0052] Construct a velocity model;

[0053] Divide the active source observation area into a plurality of grid cells; where each grid cell represents a possible source rupture point;

[0054] Respectively use the velocity model to calculate the theoretical travel times from each grid cell to each monitoring point when each grid cell is used as a single source;

[0055] According to the theoretical travel times from each grid cell to each monitoring point when each grid cell is used as a single source, respectively perform delay translation on the optical fiber vibration signals collected at each monitoring point when each grid cell is used as a single source, to obtain the intermediate optical fiber vibration signals collected at each monitoring point when each grid cell is used as a single source;

[0056] Respectively perform weighted superposition on the intermediate optical fiber vibration signals collected at each monitoring point when each grid cell is used as a single source, to obtain the optical fiber vibration signal superposition value when each grid cell is used as a single source;

[0057] Use the backprojection method to obtain a dynamic spatio-temporal image according to the optical fiber vibration signal superposition value when each grid cell is used as a single source;

[0058] Perform source identification on dynamic spatio-temporal images to obtain the source location, earthquake occurrence time, and energy characteristics.

[0059] The present invention uses multiple monitoring points on the optical fiber in the distributed optical fiber sensing module to achieve high-density and real-time backhaul acquisition of optical fiber vibration signals, and performs delay stacking and backprojection on the optical fiber vibration signals through the data processing module to obtain dynamic spatio-temporal images. By performing source identification on the dynamic spatio-temporal images to obtain the source location, earthquake occurrence time, and energy characteristics, a full real-time process of seismic data acquisition and data processing within a small scale is realized. It can overcome the deficiencies of traditional seismographs in spatial resolution, quickly locate earthquakes, without the need for identification of seismic signals and picking of travel times, greatly reducing the errors introduced by manual operations into the data, and can overcome the problem that traditional methods cannot locate simultaneously excited seismic sources.

[0060] It should be noted that distributed optical fiber sensing technology can use standard communication optical fibers to achieve continuous and real-time monitoring of seismic waves. This technology relies on the Rayleigh scattering principle in optical fibers, and captures acoustic wave activities along the optical fiber path by analyzing minute changes in scattered light, thereby converting them into seismic wave information.

[0061] The main advantage of optical fiber sensors is their ability to provide extremely high spatial resolution and coverage. A single optical fiber line can be used as thousands of sensing points, and the distance between each sensing point can be as small as dozens of centimeters, which provides unprecedented details and accuracy for seismic monitoring. In addition, optical fiber sensors have the characteristic of being resistant to electromagnetic interference and are suitable for various environmental conditions, including underwater and high-temperature scenarios.

[0062] In the present invention, distributed optical fiber sensors are deployed in preset active source observation areas, such as urban infrastructure, industrial facilities, and geological disaster-prone areas. These optical fiber sensors can continuously monitor seismic activities and capture various seismic waves from minor vibrations to large-scale earthquakes. Due to the high-density deployment of optical fiber sensors, the present invention can achieve high-precision positioning of seismic events, especially showing excellent performance in monitoring microseismic activities in small-scale areas.

[0063] In addition, the economic benefits of distributed optical fiber sensing technology are also very significant. Compared with traditional seismographs, the deployment and maintenance costs of optical fiber sensors are lower, and they are easy to deploy on a large scale. This makes the construction and expansion of seismic monitoring networks more economical and efficient.

[0064] In this embodiment, the data processing module includes multiple GPU servers, and the multiple GPU servers are used for parallel computing.

[0065] Since it is necessary to perform time shift superposition of waveforms for each grid point during the back-projection calculation, and the operation functions for each grid point are the same, it is very easy to write it as a CUDA kernel function. Multiple GPU servers in this embodiment can perform GPU parallel acceleration calculation, which can greatly increase the back-projection calculation speed so as to achieve the purpose of near-real-time calculation, enabling the event catalog to be given, and at the same time, the wave field image can be given.

[0066] In this embodiment, it further includes a network attached storage (NAS). The NAS is connected to the fiber optic demodulator. The NAS is used to store the fiber optic vibration signals obtained after distributed fiber optic demodulation in real time, facilitating the download of fiber optic vibration signals while detecting.

[0067] In this embodiment, it further includes a storage server. The storage server is connected to the data processing module. The storage server is used to store the seismic source identification results.

[0068] In this embodiment, it further includes a display module. The display module is connected to the storage server. The display module is used to display the dynamic spatio-temporal image, the seismic source location, the earthquake occurrence time, and the energy characteristics.

[0069] In this embodiment, the velocity model is a one-dimensional velocity model.

[0070] This embodiment can achieve the preset calculation accuracy using a one-dimensional velocity model in small-scale active sources.

[0071] In this embodiment, after calculating the theoretical travel times from each grid cell to each monitoring point when each grid cell is regarded as a single seismic source using the velocity model, it further includes:

[0072] Performing a secondary correction on the theoretical travel times from each grid cell to each monitoring point when each grid cell is regarded as a single seismic source.

[0073] During the secondary correction process, one of the monitoring points among all the monitoring points is set as the reference monitoring point, and the reference travel times from each grid cell to the reference monitoring point when each grid cell is regarded as a single seismic source are calculated respectively;

[0074] According to the reference travel times from each grid cell to the reference monitoring point when each grid cell is regarded as a single seismic source, the theoretical travel times from each grid cell to each monitoring point when each grid cell is regarded as a single seismic source are corrected.

[0075] This embodiment can effectively improve the accuracy of the theoretical travel times from each grid cell to each monitoring point when each grid cell is regarded as a single seismic source.

[0076] In this embodiment, during the delay translation process, when each grid cell is used as a single seismic source, the delay translation value of each monitoring point is the theoretical travel time t from the grid cell to each monitoring point when the grid cell is used as a single seismic source. kj , where k represents the k-th monitoring point and j represents the j-th grid cell.

[0077] In this embodiment, when each grid cell is used as a single seismic source, the phases of the intermediate fiber vibration signals collected at each monitoring point are consistent, that is, the phases of the intermediate fiber vibration signals collected at each monitoring point corresponding to the same grid cell are consistent.

[0078] It can be seen from this that in this embodiment, through delay translation, if a certain grid cell is exactly the real seismic source rupture position, the phases of the intermediate fiber vibration signals collected at all monitoring points corresponding to this grid cell will be superimposed after time alignment, and obvious energy enhancement will be presented.

[0079] In order to ensure that the phases of the data at each monitoring point are consistent during coherent superposition, in this embodiment, before performing delay translation on the fiber vibration signals collected at each monitoring point according to the theoretical travel time from each grid cell to each monitoring point, it further includes:

[0080] Preprocessing the fiber vibration signals collected at each monitoring point.

[0081] The preprocessing in this embodiment includes waveform annihilation and noise suppression, detrending and demeaning, filtering, waveform amplitude compensation, absolute value processing, and envelope processing.

[0082] Specifically, in this embodiment, the original waveform of the fiber vibration signal is denoised by signal processing methods (such as wavelet denoising, smoothing filtering, etc.) to remove the noise components not excited by earthquakes; then the fiber vibration signal is detrended and demeaned to eliminate the influence of instrument drift and background noise, and then a band-pass or high-pass filter is designed according to the frequency band characteristics of the target signal to filter out low-frequency or high-frequency interference in the fiber vibration signal, making the signal more focused on the earthquake excitation part; then, considering factors such as attenuation along the fiber and instrument response characteristics, the amplitude of the fiber vibration signal is compensated to ensure that the signal amplitude is within a reasonable range during subsequent superposition; then, since DAS observation is only the component along the radial direction of the fiber, when the fiber direction rotates, it will cause a polarity change in the recorded waveform. In this embodiment, the absolute value operation is performed on the fiber vibration signal to further ensure that the phases of the fiber vibration signals at each monitoring point are consistent during coherent superposition; finally, the envelope processing is performed on the fiber vibration signal, and only the maximum value of the envelope line is retained to improve the stability of low signal-to-noise ratio signals.

[0083] During the weighted superposition process, if a grid cell is located in the actual earthquake source rupture area, due to the consistent phases of the fiber optic vibration signals at each monitoring point, the weighted superposition result will be significantly enhanced; conversely, the superposition value will be weak due to phase cancellation. The superposition result for each grid cell can be calculated using Equation 1. Among them, Equation 1 is

[0084]

[0085] In the formula, S j (t) represents the weighted superposition result of the signals at all monitoring points generated by the j-th possible earthquake source rupture point, u k is the waveform data recorded at the k-th monitoring point, that is, the strain rate data observed by DAS, is the theoretical arrival time of the seismic wave (P wave) from the k-th monitoring point to the j-th possible earthquake source rupture point, and Δt is the time calculation correction amount calculated from u k and the reference monitoring point, and A k is the weight of each recorded data.

[0086] It should be noted that in small-scale observations, it can be considered that the active source signal propagates in an approximately homogeneous medium. Therefore, the correction term Δt in the above Equation 1 can be omitted, that is, Equation 2 is used for calculation. Among them, Equation 2 is

[0087]

[0088] In this embodiment, the backprojection method is used to obtain the dynamic spatio-temporal image according to the superposition value of the fiber optic vibration signals when each grid cell is used as a single earthquake source, specifically including:

[0089] Extract the superposition value of the fiber optic vibration signal at each time point in the preset time period from each grid cell when it is used as a single earthquake source;

[0090] According to the superposition value of the fiber optic vibration signal at each time point in the preset time period for each grid cell, obtain the superposition value of the fiber optic vibration signals of all grid cells at each time point;

[0091] According to the superposition value of the fiber optic vibration signals of all grid cells at each time point, obtain the two-dimensional energy distribution map at each time point;

[0092] Connect the two-dimensional images at each time point in chronological order to obtain the dynamic spatio-temporal image.

[0093] Among them, the back-projection method is essentially a time superposition method. The back-projection method requires some prior data in traditional seismology applications, such as the source mechanism of the earthquake to be studied, the approximate area of the source rupture, and the calculation model is segmented according to this rupture area. However, in the process of small-scale active source research in this embodiment, the seismogenic mechanism of the active source and the active source observation area are already clear, greatly reducing the uncertainty introduced by the prior information error to the final result.

[0094] This embodiment uses the back-projection method to obtain a dynamic spatio-temporal image based on the superposition value of the fiber optic vibration signals when each grid cell is used as a single source, effectively ensuring the accuracy of the result.

[0095] Compared with the traditional back-projection method, the back-projection method of the present invention can directly use the wave field propagation characteristics for inversion because the monitoring points based on distributed fiber optic sensing are evenly distributed and the waveforms are relatively continuous, thus eliminating the need for separate near-field arrival time correction, and can flexibly select any position along different fibers within a wide area for monitoring without additional installation of special seismic stations, making the distribution of monitoring points on the fiber optic more operable.

[0096] In this embodiment, source identification is performed on the dynamic spatio-temporal image to obtain the source position, the earthquake occurrence time, and the energy characteristics, specifically including:

[0097] Analyze the energy distribution in the dynamic spatio-temporal image, identify the position where the energy peak is located, and use the position where the energy peak is located as a possible source rupture point;

[0098] For the small-scale multi-source excitation environment, use the continuity characteristics of the energy distribution and the local extreme value method to cluster and separate the possible source rupture points to identify and distinguish multiple seismic events that are simultaneously excited;

[0099] Combine the energy enhancement moments of multiple seismic events to determine their initial earthquake occurrence times to ensure the accuracy of time positioning.

[0100] It should be noted that the energy characteristics mainly refer to the key parameters used to describe the energy changes during the occurrence and propagation of seismic events, including the energy peak, the continuity of the energy distribution, the local extreme points, and the energy enhancement moment (the time point of sudden energy increase).

[0101] In this embodiment, after source identification is performed on the dynamic spatio-temporal image to obtain the source position, the earthquake occurrence time, and the energy characteristics, it further includes:

[0102] Organize the information of the dynamic spatio-temporal image, the source position, the earthquake occurrence time, and the energy characteristics to form an event catalog for subsequent rapid retrieval and statistics.

[0103] Second aspect, asFigure 1 As shown in Figure 1 , a time-reversal projection seismic positioning method based on distributed optical fiber sensing proposed by the present invention includes:

[0104] Real-time acquisition of optical fiber vibration signals at multiple monitoring points; wherein, the duration of the optical fiber vibration signal is a preset duration;

[0105] Construct a velocity model;

[0106] Divide the active source observation area into multiple grid cells; wherein, each grid cell represents a possible source rupture point;

[0107] Respectively use the velocity model to calculate the theoretical travel time from each grid cell to each monitoring point when each grid cell is used as a single source;

[0108] According to the theoretical travel time from each grid cell to each monitoring point when each grid cell is used as a single source, respectively perform delay translation on the optical fiber vibration signals collected at each monitoring point when each grid cell is used as a single source, to obtain the intermediate optical fiber vibration signals collected at each monitoring point when each grid cell is used as a single source;

[0109] Respectively perform weighted superposition on the intermediate optical fiber vibration signals collected at each monitoring point when each grid cell is used as a single source, to obtain the superposition value of the optical fiber vibration signals when each grid cell is used as a single source;

[0110] Use the back-projection method to obtain a dynamic spatio-temporal image according to the superposition value of the optical fiber vibration signals when each grid cell is used as a single source;

[0111] Perform source identification on the dynamic spatio-temporal image to obtain the source location, origin time, and energy characteristics.

[0112] In this embodiment, the velocity model is a one-dimensional velocity model.

[0113] In this embodiment, using a one-dimensional velocity model can achieve a preset calculation accuracy in small-scale active sources.

[0114] In this embodiment, after respectively using the velocity model to calculate the theoretical travel time from each grid cell to each monitoring point when each grid cell is used as a single source, it further includes:

[0115] Perform secondary correction on the theoretical travel time from each grid cell to each monitoring point when each grid cell is used as a single source.

[0116] During the secondary correction process, set one of the monitoring points as a reference monitoring point, and respectively calculate the reference travel time from each grid cell to the reference monitoring point when each grid cell is used as a single source;

[0117] According to the reference travel times from each grid cell to the reference monitoring point when each grid cell is regarded as a single seismic source, the theoretical travel times from each grid cell to each monitoring point when each grid cell is regarded as a single seismic source are corrected.

[0118] This embodiment can effectively improve the accuracy of the theoretical travel times from each grid cell to each monitoring point when each grid cell is regarded as a single seismic source.

[0119] In this embodiment, during the delay translation process, the delay translation values of each monitoring point when each grid cell is regarded as a single seismic source are the theoretical travel times t from each grid cell to each monitoring point when each grid cell is regarded as a single seismic source, where k represents the k-th monitoring point and j represents the j-th grid cell. kj , k represents the k-th monitoring point, and j represents the j-th grid cell.

[0120] In this embodiment, the phases of the intermediate optical fiber vibration signals collected by each monitoring point when each grid cell is regarded as a single seismic source are consistent, that is, the phases of the intermediate optical fiber vibration signals collected by each monitoring point corresponding to the same grid cell are consistent.

[0121] It can be seen from this that in this embodiment, through delay translation, if a certain grid cell is exactly the real seismic source rupture position, the phases of the intermediate optical fiber vibration signals collected by all monitoring points corresponding to this grid cell are superimposed after time alignment, and obvious energy enhancement will be presented.

[0122] In order to ensure that the phases of the data at each monitoring point are consistent during coherent superposition, in this embodiment, before performing delay translation on the optical fiber vibration signals collected by each monitoring point according to the theoretical travel times from each grid cell to each monitoring point, it further includes:

[0123] Preprocessing the optical fiber vibration signals collected by each monitoring point.

[0124] The preprocessing in this embodiment includes waveform annihilation and noise suppression, detrending and demeaning, filtering, waveform amplitude compensation, absolute value processing, and envelope processing.

[0125] In specific implementation, this embodiment performs noise reduction on the original waveform of the fiber optic vibration signal through signal processing methods (such as wavelet denoising, smoothing filtering, etc.), and eliminates the noise components not excited by earthquakes; then, detrending and de-meaning processing are performed on the fiber optic vibration signal to eliminate the influence of instrument drift and background noise. Next, a band-pass or high-pass filter is designed according to the frequency band characteristics of the target signal to filter out low-frequency or high-frequency interference in the fiber optic vibration signal, making the signal more focused on the earthquake excitation part; then, considering factors such as attenuation along the fiber optic and instrument response characteristics, amplitude compensation processing is performed on the fiber optic vibration signal to ensure that the signal amplitude is within a reasonable range during subsequent superposition; then, since the DAS observation is only the component in the radial direction of the fiber optic, when the fiber optic rotates in direction, it will cause a polarity change in the recorded waveform. This embodiment further ensures that the data at each monitoring point is in phase during coherent superposition by performing an absolute value operation on the fiber optic vibration signal; finally, envelope processing is performed on the fiber optic vibration signal, and only the maximum value of the envelope line is retained to improve the stability of the low signal-to-noise ratio signal.

[0126] During the weighted superposition process, if a grid cell is located in the actual earthquake source rupture area, due to the consistent phase of the signals at each monitoring point, the weighted superposition result will be significantly enhanced; conversely, the superposition value will be weak due to phase cancellation. The superposition result for each grid cell can be calculated using Equation 1. Among them, Equation 1 is

[0127]

[0128] In the formula, S j (t) represents the weighted superposition result of the fiber optic vibration signals at all monitoring points generated by the j-th possible earthquake source rupture point, u k is the waveform data recorded at the k-th monitoring point, that is, the strain rate data observed by DAS, is the theoretical arrival time of the seismic wave (P wave) from the k-th monitoring point to the j-th possible earthquake source rupture point, and Δt is the time calculation correction amount calculated from u k and the reference monitoring point, and A k is the weight of each recorded data.

[0129] It should be noted that in small-scale observations, it can be considered that the active source signal propagates in an approximately homogeneous medium. Therefore, the correction term Δt in Equation 1 above can be omitted, that is, Equation 2 is used for calculation. Among them, Equation 2 is

[0130]

[0131] In this embodiment, the back-projection method is used to obtain a dynamic spatio-temporal image based on the superposition value of the fiber optic vibration signals when each grid cell is used as a single earthquake source, specifically including:

[0132] From the superimposed values of the fiber optic vibration signals when each grid cell is used as a single seismic source, the superimposed values of the fiber optic vibration signals at each time point within a preset time duration for each grid cell are extracted separately;

[0133] Based on the superimposed values of the fiber optic vibration signals at each time point within a preset time duration for each grid cell, the superimposed values of the fiber optic vibration signals of all grid cells at each time point are obtained;

[0134] Based on the superimposed values of the fiber optic vibration signals of all grid cells at each time point, a two-dimensional energy distribution map at each time point is obtained;

[0135] The two-dimensional images at each time point are concatenated in chronological order to obtain a dynamic spatio-temporal image.

[0136] Among them, the back-projection method is essentially a time superposition method. The back-projection method requires some prior data in traditional seismology applications, such as the seismic source mechanism of the earthquake to be studied, the approximate area of the seismic source rupture, and the calculation model is segmented according to this rupture area. However, in the process of small-scale active source research in this embodiment, the seismogenic mechanism of the active source and the active source observation area are already clear, greatly reducing the uncertainty introduced by the prior information error to the final result.

[0137] This embodiment uses the back-projection method to obtain a dynamic spatio-temporal image based on the superimposed values of the fiber optic vibration signals when each grid cell is used as a single seismic source, effectively ensuring the accuracy of the result.

[0138] Compared with the traditional back-projection method, in the back-projection method of the present invention, since the monitoring points on the optical fiber in the distributed fiber optic sensing module are evenly distributed and the waveforms are relatively continuous, the wave field propagation characteristics can be directly used for inversion, thus eliminating the need for separate near-field arrival time correction, and different arbitrary positions along the optical fiber can be flexibly selected for monitoring within a wide area without additional installation of special seismic monitoring stations, making the distribution of monitoring points more operable.

[0139] In this embodiment, source identification is performed on the dynamic spatio-temporal image to obtain the source location, the earthquake occurrence time, and the energy characteristics, specifically including:

[0140] Analyze the energy distribution in the dynamic spatio-temporal image, identify the position where the energy peak is located, and take the position where the energy peak is located as a possible seismic source rupture point;

[0141] Utilize the continuity characteristics of the energy distribution and the local extreme value method to cluster and separate the possible seismic source rupture points to identify and distinguish multiple simultaneously excited seismic events;

[0142] Based on the energy enhancement moments of multiple seismic events, determine the initial earthquake occurrence times of multiple seismic events.

[0143] In this embodiment, after identifying the seismic source for the dynamic spatio-temporal image to obtain the seismic source location, the earthquake occurrence time, and the energy characteristics, the following steps are further included:

[0144] Organize the information of the dynamic spatio-temporal image, the seismic source location, the earthquake occurrence time, and the energy characteristics to form an event catalog for subsequent rapid retrieval and statistics.

[0145] The effects of the present invention can be further illustrated by the following specific embodiments.

[0146] Embodiment 1

[0147] Use the time-reversal projection seismic positioning system based on distributed optical fiber sensing proposed in this application to conduct a drop hammer test. The test site is as shown in (a) of Figure 2 . In this embodiment, a communication optical fiber about 300 m long is laid in the lawn, and the landfill depth is 30 cm. Among them, the distributed optical fiber sensing device samples at a sampling frequency of 5 KHz, the station spacing is 1 m, the gauge length is 10 m, and there are 253 channels in total.

[0148] To simulate the single-force source active source signal generated by the shell landing, in this embodiment, a 10-kg lead ball is freely dropped from a container about 3 m high in the northern part of the lawn. The container is as shown in (b) of Figure 2 .

[0149] In this embodiment, the test site is divided into 1 m × 1 m cells.

[0150] The actual excitation point in this embodiment is as shown in (a) of Figure 3 . In this embodiment, multiple optical fiber vibration signals sampled by the distributed optical fiber sensing device are preprocessed, delay stacked, and back-projected. The obtained dynamic spatio-temporal image is as shown in (b) of Figure 3 . The difference between the actual excitation point and the possible seismic source rupture point is 1 m, which is equivalent to the accuracy of the grid cell.

[0151] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and all should be covered within the protection scope of the present invention.

Claims

1. A time-reverse projection seismic positioning system based on distributed optical fiber sensing, characterized in that: include: A distributed optical fiber sensing module and a data processing module; wherein the distributed optical fiber sensing module comprises an optical fiber arranged along the circumference of the active source observation area and an optical fiber demodulator connected to the optical fiber, and the data processing module is communicatively connected to the optical fiber demodulator; The optical fiber demodulator is used to obtain optical fiber vibration signals of multiple preset monitoring points on the optical fiber in real time; wherein the duration of the optical fiber vibration signal is a preset duration; The data processing module is used to obtain in real time the optical fiber vibration signals of multiple monitoring points obtained by the optical fiber demodulator; Construct velocity model; The active source observation area is divided into multiple grid cells, where each grid cell represents a possible earthquake source rupture point; The velocity model is used to calculate the theoretical travel time from each grid unit to each monitoring point when each grid unit is used as a single earthquake source. According to the theoretical travel time from each grid unit to each monitoring point when each grid unit is used as a single seismic source, the optical fiber vibration signal collected from each monitoring point when each grid unit is used as a single seismic source is delayed and translated to obtain the intermediate optical fiber vibration signal collected from each monitoring point when each grid unit is used as a single seismic source; The intermediate optical fiber vibration signals collected at each monitoring point when each grid unit is used as a single earthquake source are weighted and superimposed to obtain the optical fiber vibration signal superposition value when each grid unit is used as a single earthquake source; The back-projection method is used to obtain the dynamic space-time image based on the superposition value of the optical fiber vibration signal when each grid unit is used as a single seismic source; The earthquake source is identified on the dynamic space-time image to obtain the earthquake source location, earthquake occurrence time and energy characteristics.

2. The time-reverse projection seismic positioning system based on distributed optical fiber sensing according to claim 1 is characterized in that: The data processing module includes multiple GPU servers, and the multiple GPU servers are used for parallel computing.

3. The time-reverse projection seismic positioning system based on distributed optical fiber sensing according to claim 1, characterized in that: It also includes a network attached storage and a storage server. The optical fiber demodulator is connected to the data processing module and the network attached storage respectively, and the data processing module is connected to the storage server.

4. The time-reverse projection seismic positioning system based on distributed optical fiber sensing according to claim 1, characterized in that: After using the velocity model to calculate the theoretical travel time from each grid unit to each monitoring point when each grid unit is used as a single earthquake source, it also includes: When each grid unit is used as a single earthquake source, a secondary correction is made to the theoretical travel time from the grid unit to each monitoring point.

5. The time-reverse projection seismic positioning system based on distributed optical fiber sensing according to claim 1, characterized in that: In the delay translation process, the delay translation value of each monitoring point when each grid unit is used as a single earthquake source is the theoretical travel time from the grid unit to each monitoring point when each grid unit is used as a single earthquake source; When each grid unit is used as a single seismic source, the phases of the intermediate optical fiber vibration signals collected at each monitoring point are consistent.

6. The time-reverse projection seismic positioning system based on distributed optical fiber sensing according to claim 1, characterized in that: Before delaying and translating the optical fiber vibration signal collected from each monitoring point according to the theoretical travel time from each grid unit to each monitoring point, the method further includes: Pre-process the optical fiber vibration signals collected at each monitoring point; Among them, preprocessing includes waveform annihilation and noise suppression, detrending and deaveraging, filtering, waveform amplitude compensation, absolute value processing and envelope processing.

7. The time-reverse projection seismic positioning system based on distributed optical fiber sensing according to claim 1, characterized in that: The back-projection method is used to obtain the dynamic spatiotemporal image based on the superposition value of the optical fiber vibration signal when each grid unit is used as a single seismic source, including: Extracting the superposition value of the optical fiber vibration signal of each grid unit at each time point in the preset time length from the superposition value of the optical fiber vibration signal when each grid unit is used as a single seismic source; According to the superposition value of the optical fiber vibration signal of each grid unit at each time point in the preset time length, the superposition value of the optical fiber vibration signal of all grid units at each time point is obtained; According to the superposition value of the optical fiber vibration signal of all grid cells at each time point, a two-dimensional energy distribution diagram at each time point is obtained; The two-dimensional images at each time point are connected in series in chronological order to obtain a dynamic space-time image.

8. The time-reverse projection seismic positioning system based on distributed optical fiber sensing according to claim 1, characterized in that: The dynamic spatiotemporal image is used to identify the earthquake source, and the earthquake source location, earthquake occurrence time and energy characteristics are obtained, including: Analyze the energy distribution in the dynamic space-time image, identify the location of the energy peak, and use the location of the energy peak as the possible earthquake source rupture point; Using the continuity characteristics of energy distribution and the local extreme value method, possible source rupture points are clustered and separated to identify and distinguish multiple earthquake events excited simultaneously. The initial occurrence times of the multiple seismic events are determined according to the energy enhancement moments of the multiple seismic events.

9. The time-reverse projection seismic positioning system based on distributed optical fiber sensing according to claim 1, characterized in that: After the earthquake source identification is performed on the dynamic spatiotemporal image and the earthquake source location, earthquake occurrence time and energy characteristics are obtained, the following steps are also included: The dynamic space-time images, source locations, earthquake occurrence times and energy characteristics are organized to form an event catalog.

10. A time-reverse projection earthquake positioning method based on distributed optical fiber sensing, applied to the time-reverse projection earthquake positioning system based on distributed optical fiber sensing as claimed in any one of claims 1 to 9, characterized in that: include: Obtaining optical fiber vibration signals of multiple monitoring points in real time; wherein the duration of the optical fiber vibration signal is a preset duration; Construct velocity model; The active source observation area is divided into multiple grid cells, where each grid cell represents a possible earthquake source rupture point; The velocity model is used to calculate the theoretical travel time from each grid unit to each monitoring point when each grid unit is used as a single earthquake source. According to the theoretical travel time from each grid unit to each monitoring point when each grid unit is used as a single seismic source, the optical fiber vibration signal collected from each monitoring point when each grid unit is used as a single seismic source is delayed and translated to obtain the intermediate optical fiber vibration signal collected from each monitoring point when each grid unit is used as a single seismic source; The intermediate optical fiber vibration signals collected at each monitoring point when each grid unit is used as a single earthquake source are weighted and superimposed to obtain the optical fiber vibration signal superposition value when each grid unit is used as a single earthquake source; The back-projection method is used to obtain the dynamic space-time image based on the superposition value of the optical fiber vibration signal when each grid unit is used as a single seismic source; The earthquake source is identified on the dynamic space-time image to obtain the earthquake source location, earthquake occurrence time and energy characteristics.

Images