Six-component micro-seismic single-measuring-point stratigraphic structure inversion method

Abstract

The invention discloses a six-component micro-seismic single-measuring-point stratigraphic structure inversion method, which comprises the following steps of: 1) selecting a measuring point in a target area, and installing a six-component seismograph at the measuring point; 2) collecting and storing the six-component seismic signal of the measuring point, and synchronously updating the quality coefficient of the collected data; if the quality coefficient is not greater than the set quality coefficient threshold, continuing to collect the seismic signal and synchronously updating the quality coefficient of the collected data; 3) setting a time window width according to the minimum time length required by each slice, and carrying out time domain windowing slicing on the six-component seismic data; 4) extracting a phase velocity frequency dispersion curve corresponding to the slice data, and obtaining a quantitative relation between the phase velocity and the frequency; and 5) according to the correlation coefficient of each slice data, performing weighted average on the phase velocity corresponding to the slice data to obtain a final phase velocity frequency dispersion curve of the measuring point, establishing a mapping relation between the phase velocity and depth, obtaining structure information of different depths of the stratum at the measuring point, and performing inversion on the stratum phase velocity profile.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of seismic exploration, and particularly relates to a six-component microseismic single-measuring-point stratum structure inversion method based on a quality coefficient control. BACKGROUND

[0002] In the fields of geological exploration, mineral development and geophysics, accurate estimation of stratum structure is the basis for understanding the properties of underground media and guiding engineering decisions. Traditional stratum structure inversion relies on seismic exploration, electrical logging and other means, but in deep complex geological environments or strong interference backgrounds, these methods often face problems such as insufficient resolution, difficulty in data acquisition and high cost. Microseismic monitoring, as a technology for passively acquiring underground wave information, has been gradually applied to stratum structure detection research in recent years. Compared with active sources, microseismic signals have lower frequencies and can reflect geological characteristics at deeper depths. At the same time, since microseismic events occur naturally underground, their wave fields can reflect real geological structures and medium changes, providing new information sources and constraint methods for inversion, with characteristics of high temporal resolution, long-period observation and low interference, showing great potential in stratum imaging applications.

[0003] The stratum structure inversion method based on microseismic data mainly uses microseismic waveform information to extract seismic wave propagation paths and velocity models, thereby inferring the spatial distribution and physical property differences of underground media. Its principle generally includes three key steps: first, a network of multiple receivers is used to record microseismic events occurring in the region in real time; second, source location and travel time picking are performed to construct seismic wave propagation paths and an initial velocity model through inversion algorithms; finally, the velocity structure is iteratively optimized using ray tracing or wave equation numerical simulation to obtain a layered or three-dimensional distribution of underground media. Especially in faulted structural zones or mining affected areas, the microseismic wave propagation characteristics correspond to the discontinuity and heterogeneity of the medium, providing important evidence for revealing underground structure mutations.

[0004] In recent years, with the improvement of microseismic monitoring technology and seismic data processing capabilities, the stratum structure inversion method based on microseismic data has been continuously developed and improved. On the one hand, inversion algorithms have become increasingly diverse, from traditional least squares travel time inversion to advanced methods based on full waveform fitting, deep learning and Bayesian inference, making model construction more refined and reliable. On the other hand, receiver hardware has also been continuously developed, from traditional single-component receivers to three-component receivers and even six-component receivers. The development of hardware technology not only improves the accuracy of source location, but also enhances the ability to perceive the anisotropy of the medium and wave propagation mechanisms. At the same time, joint inversion that integrates multi-source data (such as microseismic, borehole seismic and surface exploration data) has gradually become a research hotspot, improving the resolution and stability of stratum structure identification.

[0005] However, microseismic data stratigraphic structure inversion still faces some key technical bottlenecks. First, microseismic events are usually low in energy and poor in signal-to-noise ratio, resulting in great uncertainty in waveform identification and travel time picking. Second, the uncertainty of the focal mechanism introduces positioning errors, thereby affecting the accuracy of the velocity model inversion. Limited by the uneven distribution of microseismic events, the inversion result is prone to have blind areas, especially in the area where the seismic sources are sparse or skewed, which restricts the implementation effect of the inversion. Especially in the process of monitoring microseismic signals by using a six-component seismograph, there are still many problems in the processing of six-component microseismic data. SUMMARY

[0006] In view of the problem that the phase velocity dispersion curve obtained from the six-component microseismic data often has a low amplitude in the process of extracting the frequency dispersion curve of the microseismic signal and the stratigraphic structure inversion by using a single-station six-component seismograph, the application provides a new six-component microseismic data processing method based on mass coefficient control. On the basis of realizing the single-station underground structure inversion relying on the relationship between the translational and rotational components, the application redesigns a complete set of data processing scheme, which can solve the problem of spectral leakage caused by the application of non-ideal band-pass filter in the traditional method and the estimation error caused by the low correlation between the components. When processing long-period seismic signals, the scheme can take into account the high spectral resolution and processing speed, more flexibly adjust the spectral resolution of the dispersion curve for different application scenarios, and more accurately obtain the corresponding stratigraphic structure profile information under the measuring point.

[0007] In addition to the data processing algorithm level, at the hardware level, the stratigraphic structure inversion method of the application only uses a single-station six-component seismograph. Compared with the traditional data processing method, the application can process the real-time collected seismic data by adjusting the parameters and output the mass coefficient of the collected data. For different application scenarios, the mass coefficient threshold can be preset according to the information to be measured, and the mass coefficient of the collected data is also updated synchronously in the process of real-time data collection. Once the mass coefficient meets the threshold requirement, the data collection at the point can be ended. Compared with the traditional method of collecting data for a specific time and then analyzing offline, the application can effectively solve the problems of data quality and collection efficiency that the traditional method may face.

[0008] The technical solution adopted by the application to solve the technical problems is:

[0009] The traditional single-station six-component stratigraphic structure inversion method first arranges a six-component seismograph at the measuring point to collect data. The duration of data collection is usually determined according to the depth to be measured as a standard period. At the same time, in order to prevent insufficient signal quality in a single period, five to six standard periods of data are usually collected at the measuring point. The arrangement and collection process in the multi-measuring-point scenario is as follows Figure 1The application improves the method by designing a new data slicing scheme to ensure the correlation between different components in each data slice in the time domain. While determining the slicing scheme, the correlation between different components in each slice under the slicing scheme is saved as the reliability index of the slice. When collecting data, the reliability analysis result of the data, i.e., the quality coefficient, can be given while collecting and recording the data in real time. Once the quality coefficient of the data at the measuring point meets the threshold value, the data collection at the measuring point can be ended. Therefore, compared with the traditional scheme, the application can reduce the collection time at any measuring point to one fifth to one sixth of the original scheme, realize efficient collection of data at the measuring point, and be more suitable for large-scale underground structure exploration and mineral exploration fields.

[0010] The single-station six-component seismograph layout and data collection flowchart based on the quality coefficient control provided by the application is shown in Figure 2 On the basis of ensuring correct installation of the instrument, the effective seismic signals collected are processed, specifically including the following steps:

[0011] S1: install the six-component seismograph, align the six components of the instrument with the respective directions in reality according to the measurement requirements, wherein the positive direction of the z component of the six-component seismograph points to the direction opposite to the gravitational acceleration at the measuring point, i.e., the celestial direction, and ensure that the seismograph is tightly coupled with the ground, estimate the minimum spectral resolution of the measurement signal according to the detection depth requirement of the measuring point, and calculate the corresponding data collection time length and quality coefficient threshold value;

[0012] S2: collect and save the six-component seismic signals of the selected measuring point, and update the quality coefficient of the collected data synchronously, if the quality coefficient is not greater than the set quality coefficient threshold value, continue to collect the seismic signals and update the quality coefficient of the collected data synchronously, if the quality coefficient is greater than the quality coefficient threshold value, output a prompt information to remind the experimental personnel that the collection at the measuring point can be ended, and decide whether to collect redundant data to improve the data reliability according to the actual needs;

[0013] S3: pre-process the collected seismic signals, including removing typical abnormal data in the time domain, aligning the data clock, removing data trends, etc., filtering in the frequency domain to remove the influence of noise, and performing time-frequency analysis on the processed data to ensure that the obtained six-component seismic data contain the frequency band of the target signal; the frequency band of the target signal is a frequency band greater than the minimum spectral resolution but less than 0.5 times the sampling frequency of the six-component seismograph; there is a signal obviously higher than the instrument self-noise in this frequency band, and in the case of collecting seismic noise, this step S3 mainly ensures that the signal is collected, avoiding the situation that the collected signal is all zero or only the instrument self-noise due to hardware failure or the like;

[0014] S4: slicing the pre-processed six-component seismic data time domain windowed by S3, the time window width is set according to the detection depth requirement of the measuring point, the minimum time length required by each slice is calculated, the time window center time is determined according to the correlation coefficient of the translational component and the rotational component in the time window, that is, the time when the correlation coefficient of the translational component and the rotational component in the time window is the largest is used as the center time of the time window, for the case that the time length of the six-component seismic data is greater than the minimum time length corresponding to the detection depth, all possible slicing schemes of the data are traversed, all slicing data and their correlation coefficients whose correlation coefficients of the translational component and the rotational component in the slice are greater than the quality coefficient threshold are saved;

[0015] S5: in the data analysis stage, the data of each slice saved is analyzed one by one, in each slice, the ratio of the translational component to the rotational component in the orthogonal direction at a single frequency point is used as the phase velocity value at the corresponding frequency point of the measuring point, the phase velocity dispersion curve corresponding to the slice data is extracted respectively, and the quantitative relationship between the phase velocity and the frequency is obtained;

[0016] S6: on the basis of obtaining the phase velocity dispersion curves of each slice, the phase velocities corresponding to each slice are weighted and averaged according to the correlation coefficients of each slice, the weighted and averaged phase velocity is taken as the final phase velocity dispersion curve estimation result of the measuring point, according to the final phase velocity dispersion curve, the mapping relationship between the phase velocity and the depth can be established, the structure information of different depths of the stratum at the measuring point is obtained, the underground structure detection of a single measuring point is completed, and the stratum phase velocity profile is inverted.

[0017] Preferably, in step S1, the six components of the instrument are aligned with the directions in reality, including aligning the three translational components with the north-south, east-west and up-down directions, and aligning the three rotational components with the north-south, east-west and up-down directions, wherein the x components of the translational and rotational directions are aligned with the north-south direction and the east-west direction respectively, and the z components of the translational and rotational directions are aligned with the direction opposite to the gravitational acceleration.

[0018] Preferably, the quality coefficient threshold in step S1 needs to determine two parts, the time length of the collected data and the quality coefficient threshold, wherein the quality coefficient threshold is used to control the correlation coefficient between the components, and only when the quality coefficient between the components of a signal reaches this threshold, the six-component signal is considered to be qualified, and the next step of analysis can be performed, the setting of the quality coefficient threshold is determined according to the actual application scene.

[0019] Preferably, the synchronous updating of the seismic signal quality coefficient in step S2 is specifically correlation analysis of the collected seismic data horizontal component and the rotation component, if the length of the collected data is less than the required data length, the correlation analysis is directly performed on the collected data, the amplitude of the correlation coefficient is taken as the amplitude of the quality coefficient, and the sign of the quality coefficient is uniformly set as negative to avoid affecting the subsequent cutoff condition judgment, if the length of the collected data is greater than or equal to the required data length, a sliding time window with a width of the required data length is applied to the collected data, the absolute value of the correlation coefficient of the data in the sliding time window is calculated, and the maximum amplitude of the obtained correlation coefficient amplitude is taken as the amplitude of the quality coefficient, and the sign of the quality coefficient is set as positive.

[0020] Preferably, the output prompt information in step S2 reminds the experimental personnel to end the acquisition of the measuring point, and the prompt information is output only when the cutoff condition is met, that is, the obtained quality coefficient is greater than the threshold value, if the experimental personnel does not manually end the acquisition of the measuring point data, the data acquisition is continued, otherwise the collected data is saved, and other parameters are initialized for the next data acquisition.

[0021] Preferably, the time window width in step S4 is calculated according to the required detection depth of the measuring point to obtain the minimum time length required by each slice, that is, the minimum spectral resolution of data processing is estimated according to the required detection depth, and the reciprocal of the minimum spectral resolution is taken as the minimum time length of each slice, in order to reduce the error caused by data processing, in actual application, two to three times of the reciprocal of the minimum spectral resolution is usually taken as the minimum time length of each slice.

[0022] Preferably, the time window center time in step S4 is determined according to the correlation coefficient of the horizontal component and the rotation component in the time window, specifically, the correlation coefficient of the horizontal component and the rotation component in the time window is calculated by sliding the sliding time window according to a given step, if the correlation coefficient of the horizontal component and the rotation component in the time window is greater than the maximum correlation coefficient value calculated previously, the correlation coefficient value is taken as the new maximum correlation coefficient value, the center time and the time window width of the slice are saved, and the next time window is continuously slid to repeat the above steps.

[0023] Preferably, the weighted average of the phase velocity corresponding to each slice according to the correlation coefficient of each slice in step S6 is that the correlation coefficient is normalized to be taken as the weight of the phase velocity, and the phase velocities corresponding to each slice are weighted and averaged, so as to avoid the influence of the correlation coefficient value on the estimated value of the phase velocity amplitude.

[0024] The characteristics of the present application are:

[0025] A new data quality evaluation system is established, which can collect data in real time and give the quality coefficient of the data, and a new data collection scheme and data processing scheme are designed in combination with the data quality coefficient, feedback prompt information is given according to the quality coefficient of the collected data whether it can be ended, and in the data processing stage, the correlation coefficient is introduced as the weight, the velocity estimation values of multiple time slices are weighted and averaged as the final result to reduce the estimation error introduced due to the poor correlation between data components, and the stability of the output result is greatly improved.

[0026] Compared with the prior art, the positive effects of the present application are:

[0027] The six-component microseismic single-point formation structure inversion method based on the quality coefficient control can improve the problem that the traditional method cannot analyze data quality in real time and adjust the experimental scheme, and can only measure fixed time data and then analyze and process the data separately, and can monitor the quality change of the collected data while collecting data by using a single station six-component seismograph, and the experimental controllability and experimental efficiency are obviously improved compared with the traditional method, and the data processing method introduced in the present application can effectively reduce the estimation error introduced due to the poor correlation between data components by introducing the correlation coefficient as the weight, weighting and averaging the estimation results of multiple time slices as the final result, greatly improving the stability of the output result, and realizing effective inversion of the formation profile information. BRIEF DESCRIPTION OF DRAWINGS

[0028] Figure 1 is a single station six-component seismograph layout and data collection process example diagram in a multi-point scene.

[0029] Figure 2 is a single station six-component seismograph layout and data collection process diagram based on the quality coefficient control proposed in the present application.

[0030] Figure 3 is a six-component microseismic single-point data time slicing operation process diagram based on the quality coefficient control proposed in the present application.

[0031] Figure 4 is a specific implementation effect diagram of the formation velocity profile structure inversion of the single-point corresponding position in a certain experiment of the present application.

[0032] (a) P-wave velocity structure diagram at the measuring point, (b) S-wave velocity structure diagram at the measuring point. DETAILED DESCRIPTION

[0033] The six-component seismic data collection process diagram of the single station six-component seismograph is as shown in Figure 2 The flowchart of the data time slicing operation in the formation structure inversion method of the single station six-component seismograph is as shown in Figure 3As shown, on the basis of ensuring correct installation of the instrument, the effective seismic signal collected is processed, specifically including the following steps:

[0034] S1: Install the six-component seismograph instrument, align the six components of the instrument with the respective directions in reality according to the measurement requirements, wherein the positive direction of the z component of the six-component seismograph points to the direction opposite to the gravitational acceleration at the measurement point, i.e., the celestial direction, and ensure that the seismograph is tightly coupled with the ground, according to the maximum detection depth d of the measurement point max Estimate the minimum spectral resolution Δf of the measurement signal, calculate the corresponding minimum data collection time T0 = 1 / Δf, and predefine a quality coefficient threshold ρ thre ;

[0035] S2: Collect and save the six-component seismic signal, denoted as a x (t), a y (t), a z (t), r x (t), r y (t), r z (t), wherein a(t) is the translational component of the seismic data, r(t) is the rotational component of the seismic data, and the subscript indicates that the variable is along the direction corresponding to the subscript (for example, a x (t) is the translational component of the seismic data in the x direction), and the quality coefficient ρ of the collected data is updated synchronously, if the quality coefficient ρ is not greater than the threshold ρ thre , continue to collect the seismic signal and update the quality coefficient of the collected data synchronously, if the quality coefficient ρ is greater than the threshold ρ thre , output a prompt information to remind the experimenter that the collection of the measurement point can be ended, and whether to collect redundant data to improve the reliability of the data is determined according to actual needs;

[0036] S3: Preprocess the collected seismic signal [a x (t), a y (t), a z (t), r x (t), r y (t), r z (t)], including removing typical abnormal data in the time domain, aligning the data clock, removing data trends, etc., filtering in the frequency domain to remove the influence of noise, and performing time-frequency analysis on the processed data [a′ x (t), a′ y (t), a′ z (t), r′ x (t), r′ y (t), r ′ z (t)], to ensure that the obtained six-component seismic data contains the frequency band of the target signal;

[0037] S4: slice the pre-processed data time domain with windowing, the time window width T is calculated according to the detection depth requirement d of the measuring point, the minimum time length T0 required for each slice is set as 1 / Δf, and Δf is the minimum spectral resolution under the detection depth requirement d, and the time window center time t i According to the correlation coefficient p of the horizontal translational component a' T (t) and the rotational vertical component r ′ z (t) in the time window, that is, the time t is adopted to make the correlation coefficient p of the horizontal translational component and the rotational component in the time window maximum i As the center time of the time window, for the data time length T total greater than the minimum time length T0 corresponding to the detection depth, all possible slice schemes S={T, t i |i=1, 2, 3, …, n} of the data are traversed, n is the number of slices contained in the final slice scheme, and all slice schemes S(p(t thre )>p i ) and the correlation coefficients c thre of the horizontal translational component and the rotational vertical component divided by each slice are saved, where i corr(a' T (t), r ′ z (t))

[0038] S5: in the data analysis stage, the data is divided according to the saved slice scheme, and the data of each slice is analyzed one by one In each slice, the phase velocity dispersion curve corresponding to the slice data is extracted by using the component data characteristics and the specific relationship characteristics between the components, and the quantitative relationship v i (f) of phase velocity and frequency is obtained, and the method for extracting the phase velocity dispersion curve includes but is not limited to the damped least square method, the genetic algorithm, the Bayesian algorithm and the single-point interference method, etc. Taking the single-point interference method as an example, the calculation formula of the dispersion curve v(f) is:

[0039]

[0040] S6: on the basis of obtaining the phase velocity dispersion curves v i (f) of each slice, the correlation coefficients c i of each slice are used to weight average the phase velocities corresponding to each slice, and the weighted average phase velocity v(f) is taken as the final phase velocity dispersion curve estimation result of the measuring point, and the depth d can be calculated from the frequency f by using the half-wavelength theory:

[0041]

[0042] where c is wave number, f is frequency, depth d is just half of the wavelength λ, so the mapping relationship v(d) between phase velocity v(f) and depth d can be established, the structure information of different depths at the measuring point is obtained, the underground structure detection of single measuring point is completed, and the formation phase velocity profile is inverted.

[0043] Preferably, in step S1, the six components of the instrument are aligned with the respective directions in reality, including pointing the positive direction of the translational x component to the north, pointing the positive direction of the translational y component to the east, pointing the positive direction of the translational z component to the direction opposite to the gravitational acceleration, pointing the positive direction of the rotational x component to the north, pointing the positive direction of the rotational y component to the east, and pointing the positive direction of the rotational z component to the direction opposite to the gravitational acceleration.

[0044] Preferably, the quality coefficient threshold ρ in step S1 is thre The quality coefficient threshold ρ is used to represent whether the length of the collected data meets the minimum data collection length T0, and the negative sign means that the length of the collected data is less than the minimum data collection length T0, and the positive sign means that the length of the collected data is greater than or equal to the minimum data collection length T0. The amplitude of the correlation coefficient threshold is determined according to the actual application scenario. thre The quality coefficient threshold ρ is used to represent whether the length of the collected data meets the minimum data collection length T0, and the negative sign means that the length of the collected data is less than the minimum data collection length T0, and the positive sign means that the length of the collected data is greater than or equal to the minimum data collection length T0. The amplitude of the correlation coefficient threshold is determined according to the actual application scenario.

[0045] Preferably, in step S2, the synchronous updating of the seismic signal quality coefficient is specifically correlation analysis on the collected translational components and rotational components, if the length of the collected data is less than the minimum data collection length T0, the correlation analysis is directly performed on the collected data, the amplitude of the correlation coefficient is taken as the amplitude of the quality coefficient, and the sign of the quality coefficient is uniformly set to negative to avoid affecting the subsequent cutoff condition judgment, if the length of the collected data is greater than or equal to the minimum data collection length T0, a sliding time window with a width of the minimum data collection length T0 is applied to the collected data, the absolute value of the correlation coefficient is calculated, the amplitude of the obtained correlation coefficient with the maximum amplitude is taken as the amplitude of the quality coefficient, and the sign of the quality coefficient is set to positive. Taking the length of the collected data length(x(t)) of any component x(t) as an example, the expression is:

[0046]

[0047] Preferably, in step S2, the output prompt information reminds the experimental personnel that the measuring point collection can be ended, and the prompt information is output only when the cutoff condition is met, i.e., the obtained quality coefficient ρ is greater than the threshold ρ thre , if the experimental personnel do not manually end the measuring point data collection, the six-component seismic data continues to be collected, otherwise, the collected data is saved, recorded as [a x (t), ay (t),a z (t),r x (t),r y (t),r z (t)] and other parameters are initialized for the next data acquisition.

[0048] Preferably, the time window width T in step S4 is calculated according to the detection depth requirement d of the measuring point, that is, the minimum time length T0 required for each slice is estimated according to the detection depth requirement d, the minimum spectral resolution Af of data processing is estimated, and the reciprocal of the minimum spectral resolution Af is taken as the minimum time length of each slice. In order to reduce the error caused by data processing, in actual application, three times of the reciprocal of the minimum spectral resolution is usually taken as the time length of each slice, that is, the time window width T is calculated according to the formula:

[0049] Preferably, the time window center time t in step S4 is calculated according to the detection depth requirement d of the measuring point, that is, the minimum time length T0 required for each slice is estimated according to the detection depth requirement d, the minimum spectral resolution Af of data processing is estimated, and the reciprocal of the minimum spectral resolution Af is taken as the minimum time length of each slice. i According to the correlation coefficient of the translational component and the rotational component in the time window, specifically, the correlation coefficient corr(a T (t),r ′ z (t)) is calculated by sliding the time window according to a given step length. If the correlation coefficient of the translational component and the rotational component in the time window is greater than the maximum correlation coefficient value corr max , the value of the correlation coefficient is taken as the new maximum correlation coefficient value corr The center time t i and the time window width T of the slice are saved, and then the above steps are repeated by sliding to the next time window.

[0050] Preferably, in step S6, the phase velocity corresponding to each slice is weighted and averaged according to the correlation coefficient of each slice, that is, after normalization, the correlation coefficient is taken as the weight of the phase velocity, and the phase velocity corresponding to each slice is weighted and averaged, and the expression is:

[0051]

[0052] By using the correlation coefficient c i , the phase velocity v i (f) corresponding to each slice is weighted and averaged, which can increase the weight of the frequency dispersion curve obtained from the data with high correlation in the final result, so as to avoid deviation and error when estimating the phase velocity amplitude by using the data with relatively low correlation coefficient.

[0053] Figure 4 ​The schematic diagram of the inversion results of the stratum structure velocity profile corresponding to the measuring point position obtained by using the microseismic six-component data collected in the experiment is drawn, and the rock interface at the depth of 107 meters can be obviously seen in the diagram and is highly consistent with the results of the drilling sampling.

[0054] The above has carried out the detailed explanation to the present application, but obviously the specific implementation form of the present application is not limited to this. For the general technical personnel of the present technical field, the various obvious changes to it without departing from the spirit of the method and the scope of the claims of the present application are within the protection scope of the present application.

Claims

1. A six-component microseismic single-station formation structure inversion method, comprising the steps of: 1) selecting a station in a target area and installing a six-component seismometer at the station; wherein the positive direction of the z-component of the six-component seismometer points to a direction opposite to the gravitational acceleration at the station; estimating the minimum spectral resolution of the measured signal according to the detection depth requirement of the station, calculating the corresponding data acquisition time length and setting a quality coefficient threshold; 2) collecting and saving the six-component seismic signals of the station and synchronously updating the quality coefficient of the collected data; if the quality coefficient is not greater than the set quality coefficient threshold, continue collecting the seismic signals and synchronously updating the quality coefficient of the collected data, and if the quality coefficient is greater than the quality coefficient threshold, end the collection; 3) preprocessing the collected seismic signals and performing time-frequency analysis on the processed data to ensure that the obtained six-component seismic data contain the target signal frequency band; the target signal frequency band is a frequency band greater than the minimum spectral resolution but less than 0.5 times the sampling frequency of the six-component seismometer; 4) calculating the minimum time length required for each slice according to the detection depth requirement of the station, and then setting the time window width to perform time-domain windowed slicing on the six-component seismic data processed in step 3) according to the minimum time length; wherein the time when the correlation coefficient of the translational component and the rotational component of the six-component seismic data in the time window is maximum is taken as the center time of the time window, all slice schemes are traversed, and the slice data and the correlation coefficient thereof whose correlation coefficient of the translational component and the rotational component is greater than the quality coefficient threshold are saved; 5) for each slice data, extracting the corresponding phase velocity dispersion curve of the slice data to obtain the quantitative relationship between phase velocity and frequency; 6) according to the correlation coefficient of each slice data, weightedly averaging the phase velocities corresponding to each slice data, taking the weightedly averaged phase velocity as the final phase velocity dispersion curve of the station, establishing the mapping relationship between phase velocity and depth according to the final phase velocity dispersion curve, obtaining the structure information of different depths of the station, and inverting the formation phase velocity profile.

2. The method of claim 1, wherein, The method for synchronously updating the quality coefficient of the collected data is: if the length of the collected data is less than the set data length, directly performing correlation analysis on the translational component and the rotational component of the collected six-component seismic data, taking the correlation coefficient amplitude of the translational component and the rotational component as the amplitude of the quality coefficient, and uniformly setting the sign of the quality coefficient as negative; if the length of the collected data is greater than or equal to the set data length, applying a sliding time window with a width of the set data length to the collected data, calculating the absolute value of the correlation coefficient of the data in the sliding time window, and taking the maximum correlation coefficient amplitude obtained as the amplitude of the quality coefficient, and setting the sign of the quality coefficient as positive.

3. The method of claim 1, wherein, The method for calculating the minimum time length is: estimating the minimum spectral resolution of data processing according to the detection depth requirement, and taking the reciprocal of the minimum spectral resolution as the minimum time length of each slice.

4. The method of claim 3, wherein, The minimum time length is taken as two to three times the reciprocal of the minimum spectral resolution.

5. The method according to claim 1 or 2 or 3, characterized in that, For each slice data, a phase velocity value of a corresponding frequency point at the measuring point is generated by taking a ratio of a translational component in a quadrature direction to a rotational component at each single frequency point in the slice data, and a phase velocity dispersion curve corresponding to the slice data is obtained to obtain a quantitative relationship between the phase velocity and the frequency.

6. The method according to claim 1 or 2 or 3, characterized in that, According to the maximum probing depth d of the measuring point max The lowest spectral resolution Δf of the measurement signal is estimated, the minimum data acquisition time T0 = 1 / Δf.

7. The method according to claim 1 or 2 or 3, characterized in that, Depth is calculated using frequency f A mapping relationship v(d) between the weighted average phase velocity v(f) and the depth d is then established to obtain structural information of the formation at different depths at the measurement point; where c is the wave number and f is the frequency.

8. The method of claim 1, wherein, The method for extracting the phase velocity dispersion curve includes, but is not limited to, a damped least square method, a genetic algorithm, a Bayesian algorithm and a single-point interference method.

Images