A method and system for reconstructing passive seismic records while drilling based on modal filtering

Abstract

The present disclosure provides a drilling passive seismic recording reconstruction method and system based on modal screening, relating to the field of seismic inversion technology, comprising: acquiring a plurality of short-time multi-channel records in the drilling and blasting tunnel construction process; performing energy-spectrum entropy joint calibration noise-assisted modal decomposition on each short-time record to obtain multi-scale modes; calculating the modal comprehensive score of each mode and constructing effective seismic signals; performing phase-amplitude mixed whitening and reference channel adaptive selection on the effective seismic signals, constructing a multi-reference weighted deconvolution seismic interference equation in the preset effective frequency band, and solving the virtual source response according to the multiple reference channels; performing multi-process processing on the virtual source response to obtain a pulsed virtual shot gather or a passive source reconstruction record, and outputting the reconstruction result after quality evaluation. The present disclosure improves the interpretability, stability and repeatability of passive source data for tunnel advanced detection.

Description

Technical Field

[0001] This disclosure relates to the field of seismic inversion technology, specifically to a method and system for reconstructing passive seismic records while drilling based on modal screening. Background Technology

[0002] The statements in this section are merely background information relating to this disclosure and do not necessarily constitute prior art.

[0003] Seismic inversion is one of the core technical means of tunnel advanced geological prediction. It uses the propagation law of seismic waves to infer the velocity structure of the rock mass in front of the tunnel face, thereby identifying adverse geological bodies such as fault fracture zones, weak interlayers, water-rich zones and lithological abrupt interfaces, providing key basis for construction safety control and support parameter design.

[0004] Traditional seismic inversion relies heavily on active source seismic data. However, tunnel construction itself generates continuous and abundant passive source vibration signals, including drilling noise from the drilling rig, vibrations from the drill rod and tools, post-blasting rock structure adjustments, mechanical disturbances, and vibrations from muck removal equipment. Compared to active sources, passive sources offer significant advantages such as not interrupting construction time, being acquired synchronously with the construction process, and allowing for continuous long-term data collection. The propagation of these construction disturbances within the surrounding rock carries crucial information about wave velocity, attenuation, and interface reflections of the medium surrounding the tunnel and in front of the tunnel face, theoretically providing a physical basis for velocity analysis.

[0005] However, passive source data has inherent complexity, and its direct application still has the following limitations: (i) The location of the earthquake source is not fixed, the time of onset is unknown, and the source wavelet is difficult to describe accurately, resulting in the recorded signal exhibiting strong non-stationary characteristics; (ii) The narrowband line spectrum interference generated by construction machinery has strong energy and can easily mask effective propagation-related transient signals; (iii) The intensity of construction activities varies significantly within different time windows, resulting in drastic dynamic changes in signal energy; In summary, the above factors together result in passive source data exhibiting extremely messy characteristics, making it unsuitable for direct use in traditional processing workflows such as velocity analysis and reflection imaging. Summary of the Invention

[0006] To address the aforementioned issues, this disclosure proposes a method and system for reconstructing passive seismic records during drilling based on modal screening. This method extracts propagation-related wavefields from continuous, non-stationary construction vibration records containing narrowband mechanical disturbances and reconstructs them into pulsed virtual source responses similar to active source shot gathers. Through record reconstruction, while preserving the advantages of continuous passive source observations, it transforms cluttered data into effective input data suitable for velocity analysis, reflection identification, migration imaging, and full waveform inversion, thereby fundamentally transforming construction disturbance data into usable engineering information.

[0007] According to some embodiments, the present disclosure adopts the following technical solutions: Modality-based passive seismic record reconstruction methods during drilling include: Acquire passive source seismic records during the drill-and-blast tunnel construction process, and convert continuous passive source seismic records into multiple short-time multichannel records; For each short-time record, perform noise-assisted mode decomposition with joint energy-spectral entropy calibration to obtain multi-scale modes; Based on the target frequency band contribution, narrowband spectral line penalty term, transient energy concentration and inter-channel coherence of each mode, the modal comprehensive score is calculated and an effective seismic signal is constructed. Phase-amplitude hybrid whitening and adaptive selection of reference traces are performed on the effective seismic signals. Multi-reference weighted deconvolution seismic interferometry equations are constructed within the preset effective frequency band, and the virtual source response is solved jointly based on multiple reference traces. The virtual seismic source response is processed through multiple processes to obtain pulsed virtual shot gathers or passive source reconstruction records. The reconstruction results are then evaluated for quality and output.

[0008] According to some embodiments, the present disclosure adopts the following technical solutions: A seismic record reconstruction system based on modal screening and virtual source response includes: The data acquisition module is used to acquire passive source seismic records during the drill-and-blast tunnel construction process; The preprocessing and candidate window filtering module is used to convert continuous passive source seismic records into multiple short-time multichannel records; The adaptive mode decomposition module is used to perform noise-assisted mode decomposition with joint energy-spectral entropy calibration on each short-time record to obtain multi-scale modes. The modal scoring and effective signal construction module is used to calculate the modal comprehensive score and construct the effective seismic signal based on the target frequency band contribution, narrowband spectral line penalty term, transient energy concentration and inter-trace coherence of each mode. The multi-reference robust interferometry module is used to perform phase-amplitude hybrid whitening and adaptive selection of reference traces on effective seismic signals, construct multi-reference weighted deconvolution seismic interferometry equations within a preset effective frequency band, and solve the virtual source response based on multiple reference traces. The passive source reconstruction module is used to perform multi-process processing on the virtual seismic source response to obtain pulsed virtual shot sets or passive source reconstruction records. The evaluation and result output module is used to evaluate the quality of the reconstruction results and then output the results.

[0009] According to some embodiments, the present disclosure adopts the following technical solutions: A non-transitory computer-readable storage medium is provided for storing computer instructions, which, when executed by a processor, implement the modal screening-based passive seismic record reconstruction method for drilling.

[0010] According to some embodiments, the present disclosure adopts the following technical solutions: An electronic device includes a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to implement the modal screening-based passive seismic record reconstruction method.

[0011] Compared with the prior art, the beneficial effects of this disclosure are as follows: This disclosed method for reconstructing passive seismic records while drilling, based on modal screening, systematically processes continuous, non-stationary, and heavily disturbed raw passive source records, reconstructing them into virtual shot gather records with response characteristics similar to active sources. Through record reconstruction, while preserving the advantages of continuous passive source observations, it transforms chaotic data into effective input data suitable for velocity analysis, reflection identification, migration imaging, and full waveform inversion, thereby fundamentally transforming construction disturbance data into usable engineering information.

[0012] The present invention discloses a passive seismic record reconstruction method based on modal screening. Through modal decomposition with joint calibration of energy and spectral entropy, it adapts to the characteristics of strong nonstationarity, large energy variation and prominent narrowband line spectrum in construction records. Through multi-criteria mode soft screening, it avoids the effective transient loss caused by fixed IMF sequence number or fixed bandpass filter.

[0013] The modal screening-based passive seismic record reconstruction method disclosed herein reduces the impact of single reference trace coupling anomalies, local spectral anomalies, and reference sensitivity on the reconstruction results through multi-reference weighted deconvolution interferometry.

[0014] The present invention discloses a modal screening-based passive seismic record reconstruction method for drilling, which suppresses ringing, tailing and false continuity events caused by narrowband mechanical disturbances through phase-amplitude hybrid whitening and condition number adaptive regularization. Attached Figure Description

[0015] The accompanying drawings, which form part of this disclosure, are used to provide a further understanding of this disclosure. The illustrative embodiments of this disclosure and their descriptions are used to explain this disclosure and do not constitute an undue limitation of this disclosure.

[0016] Figure 1 This is a flowchart of the passive seismic record reconstruction method based on modal screening according to an embodiment of the present disclosure. Detailed Implementation

[0017] The present disclosure will be further described below with reference to the accompanying drawings and embodiments.

[0018] It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of this disclosure. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains.

[0019] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this disclosure. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms “comprising” and / or “including” are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0020] Example 1 One embodiment of this disclosure provides a method for reconstructing passive seismic records while drilling based on modal screening, the method steps of which include: Step 1: Obtain passive source seismic records during the drill-and-blast tunnel construction process, and convert the continuous passive source seismic records into multiple short-time multichannel records; Step 2: Perform noise-assisted mode decomposition with joint energy-spectral entropy calibration on each short-time record to obtain multi-scale modes; Step 3: Based on the target frequency band contribution, narrowband spectral line penalty term, transient energy concentration and inter-channel coherence of each mode, calculate the modal comprehensive score and construct the effective seismic signal; Step 4: Perform phase-amplitude hybrid whitening and adaptive selection of reference traces on the effective seismic signals, construct multi-reference weighted deconvolution seismic interferometry equations within the preset effective frequency band, and solve the virtual source response based on multiple reference traces. Step 5: Perform multi-process processing on the virtual source response to obtain pulsed virtual shot gather or passive source reconstruction records, and output the reconstruction results after quality evaluation.

[0021] As one embodiment, the modal screening-based passive seismic record reconstruction method disclosed herein is used to extract propagation-related wavefields from continuous, non-stationary construction vibration records containing narrowband mechanical disturbances, and reconstruct them into pulsed virtual source responses similar to active source shot gathers. This method aims to improve the interpretability, stability, and repeatability of passive source data when used for tunnel advance detection. The specific implementation process is as follows: Step 1: Obtain passive source seismic records during the drill-and-blast tunnel construction process, and convert the continuous passive source seismic records into multiple short-time multichannel records.

[0022] Multiple single-component or three-component geophones are deployed at the rear side wall, side arch waists, or arch feet of the drill-and-blast tunnel face. The number of geophones ranges from 12 to 48, with a spacing of 1 to 3 meters. The geophones are stably coupled to the surrounding rock or initial support structure through anchoring, borehole installation, plaster coupling, or expansion bolt fixation. During drilling of advance holes, blast holes, or geological boreholes by the drilling rig, continuous recording of passive drilling noise sources is collected. The collection time can cover one borehole, one blast hole group, or one construction cycle, with the sampling frequency determined based on the geophone's dominant frequency and the construction noise frequency band.

[0023] Specifically, deploying within the tunnel The detector continuously samples to obtain the first... Original records of each detector ,in Continuous records are divided into time windows. The process involves dividing the data into segments and performing mean-removal, detrending, amplitude stabilization normalization, and band-limiting preprocessing on each window. The formula for window-based amplitude stabilization preprocessing is shown below:

[0024] in, This represents the band-limited operator within the operating frequency band. For window functions, This represents the absolute deviation of the median for that window. This is a stabilizing term. This formula is used to unify the amplitude scale of different time windows and channels, reducing the impact of occasional large amplitudes and coupling anomalies on subsequent processing.

[0025] To avoid costly processing of all consecutive records, candidate window scores are calculated first. The candidate window scoring formula is shown below:

[0026] in, This indicates the coherence between the windows. Indicates the effective frequency band energy percentage. Indicates the intensity of line spectrum interference. Indicates the proportion of saturated or abnormal channels. This represents the positive weight coefficients corresponding to each item (n=1-4).

[0027] When candidate window scoring When the value exceeds a preset threshold, the window proceeds to subsequent adaptive mode decomposition.

[0028] Step 2: Perform noise-assisted mode decomposition with energy-spectral entropy joint calibration on each short-time record to obtain multi-scale modes.

[0029] Passive source noise records collected during drill-and-blast tunnel construction were analyzed. Within each time window, noise-assisted mode decomposition (MAD) with energy-spectral entropy joint calibration was used to obtain multi-scale modes. Specifically, for each record... Noise-assisted mode decomposition with joint energy-spectral entropy calibration is performed to obtain multiple modal components and residual terms. To accommodate the strong nonstationarity of drill-and-blast construction records, this disclosure does not use a fixed noise injection amplitude, but instead determines the time-varying perturbation scale based on local energy and local spectral entropy.

[0030] First, the formula for local energy is as follows:

[0031] in, The length of the calculation window for local energy.

[0032] Calculate normalized spectral probability within the same sliding window Thus, the local spectral entropy is defined.

[0033] The formula for local spectral entropy is as follows:

[0034] in, For the first Local energy within a frequency band This represents the number of frequency bands. The lower the local spectral entropy, the stronger the narrowband line spectrum or quasi-periodic mechanical vibration.

[0035] After combining energy and spectral entropy, the first The standard deviation of noise injection for sub-set decomposition can be set as follows:

[0036] in, As the reference disturbance coefficient, This is the amplitude scale. The average energy inside the window. The energy regulation index, is the spectral entropy adjustment coefficient. The energy-spectral entropy joint noise injection formula provides stronger ensemble perturbation to low-energy or line-spectral-dominant segments to improve mode separation stability; while reducing perturbation to high-energy propagation transient segments to retain effective arrival times.

[0037] Furthermore, after adding time-varying perturbations, the first... The set perturbation record can be written in the following form, and the formula for the set perturbation record is:

[0038] in, A sequence of white or colored noise with zero mean and unit variance.

[0039] right A complete ensemble empirical mode decomposition is performed, yielding multiple modal components and residual terms. The ensemble mode decomposition formula is shown below:

[0040] in, Indicates the first q The decomposition yields the th... k Individual Modal Components (IMF) No. q The residual terms (trend or indivisible energy) after the sub-decomposition. This indicates the total number of modes decomposed from this channel.

[0041] The stable mode components are obtained by averaging across all sets. The formula for the set-averaged mode is shown below:

[0042] Ultimately, each window and each record can be approximately represented in the form of a stable mode reconstruction formula:

[0043] in, Indicates the total number of sets. Indicates to Q The stable modal components (final modes) after averaging the decomposition. This represents the stable residual term after averaging.

[0044] Step 3: Based on the target frequency band contribution, narrowband spectral line penalty term, transient energy concentration and inter-channel coherence of each mode, construct a comprehensive mode score, and construct an effective seismic signal through soft weighting or threshold screening.

[0045] Specifically, effective seismic signals are constructed based on a comprehensive score of frequency band contribution, narrowband spectral line penalty, transient energy concentration, and inter-channel coherence structural modes, and through soft weighting or threshold screening.

[0046] For each mode The target frequency band contribution, narrowband spectral line penalty term, transient energy concentration and interchannel coherence are calculated respectively.

[0047] (1) The formula for the contribution of the target frequency band is as follows:

[0048] in, This represents the Fourier spectrum of the mode. This represents the Nyquist frequency (half the sampling rate). Indicates the low-frequency cutoff frequency. Indicates the high-frequency cutoff frequency. This indicates the proportion of the energy of this mode within the operating frequency band. It is a stable term.

[0049] (2) The formula for the narrowband spectral line penalty term is as follows:

[0050] in, Indicates the concentration of spectral peaks within the operating frequency band. Indicates the low-frequency cutoff frequency. This indicates the high-frequency cutoff frequency.

[0051] (3) The formula for transient energy concentration is as follows:

[0052] in, The length of the transient energy concentration window. Indicates the concentration of transient energy. Indicates the start time of the sliding window.

[0053] (4) Let In order to be with the first The coherence formula between adjacent channels is shown below:

[0054] in, Indicates the relationship with the first i The set of adjacent channel indices. Indicates the time offset. Indicates the maximum allowed time-shift search range. This indicates that the mode is in the first... i Daohedi j The total energy of the Tao This indicates the average similarity between the mode and its adjacent channels.

[0055] Based on the above indicators, the modal comprehensive score is established as follows:

[0056] in, This is a penalty term for low-frequency trends or excessively long periods of sustained oscillation. , , , , The soft weights are constructed based on the scores, using the following parameters: Modal soft weight formula:

[0057] in, Control the steepness of the screening. As the scoring threshold, This indicates the retained soft weights for this mode.

[0058] The effective signal is obtained by weighted superposition of each mode, and the formula for constructing the effective signal is as follows:

[0059] When the number of modes that meet the conditions within a certain time window is too small, the mode with the highest score can be retained. Each modality serves as a fallback strategy to prevent valid signals from being empty.

[0060] Step 4: Perform phase-amplitude hybrid whitening and adaptive selection of reference traces on the effective seismic signal, construct multi-reference weighted deconvolution seismic interferometry equations within the preset effective frequency band, and solve the virtual source response based on multiple reference traces.

[0061] First, perform a Fourier transform on the effective signal to obtain the effective signal spectrum, as shown in the following formula:

[0062] in, Represents angular frequency. This represents the Fourier transform operator.

[0063] In the operating frequency band Phase-amplitude hybrid whitening is performed internally, and the phase-amplitude hybrid whitening formula is as follows:

[0064] in, The whitening index. When When, it is approximately phase-dominated reconstruction; when At that time, the original amplitude is retained. For drill-and-blast construction records with obvious narrow-band mechanical line spectra, the original amplitude can be retained. Approaching 1; for records with relatively reliable amplitude information, the whitening index can be reduced. .

[0065] Reference Dao Collection The selection is not fixed, but automatically determined based on a reference quality score. The reference mass can be defined in the following form:

[0066] in, The average coherence of this channel with other channels. For an effective in-window signal-to-noise ratio, For spatially dispersed reward items, For line spectrum dominance, This is a penalty term for coupling anomalies or saturation. (Selection) Higher and not overly concentrated in space Each reference channel consists of The formula for selecting the reference set is shown below:

[0067] Furthermore, the frequency domain signals of all effective windows are combined to form a receiver-realization matrix, the formula of which is:

[0068] in, Represents the field of complex numbers. Indicates the number of channels. Indicates a time window.

[0069] from Extracting reference set The corresponding rows form a reference submatrix. To reflect the quality differences of the reference track, a reference weight matrix is ​​set up. The formula for the reference weight matrix is ​​as follows:

[0070] The weighted reference matrix is ​​constructed as follows:

[0071] A multi-reference deconvolution interferometry model is established in the frequency domain, as shown below:

[0072] in, This represents the multi-reference virtual source response matrix to be determined. This is the residual term.

[0073] The solution is obtained using weighted Tikhonov regularization, and the multi-reference weighted MDD solution formula is shown below:

[0074] in, This indicates the conjugate transpose. The reference stabilization matrix can be an identity matrix, a reference energy diagonal matrix, or a spectral stability diagonal matrix. This is the frequency-related regularization parameter.

[0075] To suppress the influence of ill-conditioned frequencies on the reconstruction results, a condition number for the reference normal matrix is ​​defined. The formula for the condition number of the reference normal matrix is ​​shown below:

[0076] The regularization parameters can be set adaptively as follows, and the adaptive regularization formula is shown below:

[0077] in, This represents the median of the condition number within the operating frequency band. For reference set at frequency points The dominance index of the line spectrum at a given location. The condition number adaptive regularization formula enhances the regularization of ill-conditioned frequencies and line spectrum frequencies, thereby reducing ringing and tailing.

[0078] Step 5: Perform multi-process processing on the virtual source response to obtain pulsed virtual shot gather or passive source reconstruction records, and output the reconstruction results after quality evaluation.

[0079] Specifically, by performing frequency band window smoothing and inverse Fourier transform on the frequency domain virtual source response obtained above, the virtual shot gather recovery formula is shown below:

[0080] in, A smooth window for the working frequency band.

[0081] If both positive and negative time delay directions are retained, the causal and anti-causal components can be integrated based on energy and coherence. The causal / anti-causal balance integration formula is shown below:

[0082] in, This represents the integrated symmetric / balanced virtual set. This indicates time reversal of the reciprocal path. These represent the weighted fusion coefficients for the forward and reverse components, respectively.

[0083] When construction noise lighting has a clear directionality, the positive and negative weights can be automatically determined based on the energy of the positive and negative time delay reflection windows to improve the continuity of events in the reconstructed records.

[0084] Furthermore, the suitability of the reconstruction results for subsequent imaging or inversion is determined by calculating the energy salience of the reflection window, the continuity of adjacent channels, the consistency of the active source reference, and the apparent velocity.

[0085] (1) To determine whether the reconstruction results are suitable for subsequent imaging or inversion, the energy salience of the reflection window can be calculated. The formula for the energy salience of the reflection window is:

[0086] in, This indicates the energy spurious ratio of the reflection window relative to the earlier window. This indicates the calculation of the envelope of the earthquake record. This indicates the time window in which the reflected wave from the target arrives. This indicates early noise or a direct wave window.

[0087] (2) The formula for the continuity of adjacent channels is as follows:

[0088] in, Represents the cross-correlation coefficient. This indicates the search window for correcting the time difference between adjacent channels. Indicates the number of channels.

[0089] (3) When there is a hammer blow or a small amount of active source reference record When active source reference consistency is reached, the formula for active source reference consistency can be calculated as follows:

[0090] in, This represents the envelope vector of the active source reference record. Indicates time offset The passive source reconstructs the envelope vector. This indicates the maximum allowed time-shift search range.

[0091] (4) The apparent velocity is estimated using the least squares method. The formula for the apparent velocity estimation is as follows:

[0092] in, For the first Detector position When picked up, To pick up the confidence level.

[0093] The final output includes reconstructed virtual shot set and reflection coherence indices. Continuity indicators Active source consistency Apparent speed And event zones that can be used for advanced geological interpretation.

[0094] Example 2 One embodiment of this disclosure provides a method for reconstructing passive seismic records while drilling based on modal screening. Taking the passive source reconstruction of drilling noise in drill-and-blast tunnels as an example, the specific content includes: During ventilation and hazard removal or muck removal intervals after blasting, the surrounding rock may experience short-term micro-fractures and structural adjustment vibrations. The system selects time windows with moderate energy and high inter-channel coherence from continuous fretting records, prioritizing the removal of strong periodic vibrations generated by ventilation fans, air compressors, and transportation equipment. Because post-blasting fretting events may be highly transient, the inter-channel coherence score can be appropriately increased in the candidate window evaluation. and effective frequency band energy ratio The weights are determined to ensure that only propagation-related events enter the modality decomposition.

[0095] In this embodiment, the phase-amplitude hybrid whitening index can be taken as 0.3 to 0.7 to retain some amplitude differences. When the continuity of the reflection window is insufficient, the condition number in the adaptive regularization formula can be increased to improve the stability of the deconvolution solution. If reflection events exist on both sides of the positive and negative time delays, the positive and negative weights can be automatically adjusted using the causal / anticausal balance integration formula to fuse causal and anticausal responses, thereby improving the lateral continuity of late-arriving events.

[0096] This embodiment is applicable to tunnel sections where there is significant surrounding rock adjustment vibration shortly after blasting, but it is inconvenient to deploy active sources again or the repeatability of active source excitation is insufficient.

[0097] Example 3 One embodiment of this disclosure provides a method for reconstructing passive seismic records while drilling based on modal screening. Taking robust reconstruction under strong narrowband mechanical disturbance scenarios as an example, the specific content includes: When ventilation fans, air compressors, rock drilling equipment, slag removal equipment, or other construction machinery generate strong narrow-band line spectra, continuous horizontal stripes or stable periodic oscillations may appear in the original record. If cross-correlation or single-reference deconvolution is performed directly, the line spectrum frequency tends to dominate the virtual source response, leading to ringing, tailing, or spurious continuous events in the reconstructed record.

[0098] In this scenario, the following parameter adjustment method can be adopted: increase the spectral entropy adjustment coefficient in the energy-spectral entropy joint noise injection formula. This allows for stronger ensemble perturbations during periods when line spectra with lower spectral entropy are dominant, thereby improving mode separation; the narrowband penalty weight is increased in the mode synthesis scoring formula. Reduce the soft weight of the dominant mode of the line spectrum; in the phase-amplitude hybrid whitening formula, reduce the whitening index. Approaching 1 weakens amplitude-dominant line spectra; increasing the frequency regularization intensity coefficient of the line spectrum in the condition number adaptive regularization formula. This allows for stronger regularization suppression of line spectrum frequencies.

[0099] By combining the above methods, the dominant role of narrowband mechanical vibration in the virtual source response can be reduced, the traceability of late-arrival reflection events can be improved, and the reconstructed virtual gun set can be made closer to the pulse-type active source response.

[0100] Example 4 As one implementation, active source recordings do not need to be acquired every time; they are only used to calibrate the passive source reconstruction quality. The system can calculate the active source reference consistency according to the active source reference consistency formula. And correct the zero point of the passive source reconstruction record based on the arrival time of the direct wave from the active source.

[0101] when When the threshold is below a preset threshold, the system can automatically take one or a combination of the following measures: increase the candidate window screening threshold, reselect the reference channel set, increase the regularization parameter, narrow the working bandwidth, or change the whitening index. .when When the threshold is reached, the passive source reconstructed gather can be used as input for subsequent pre-stack migration, reverse time migration, scatterer localization, or full waveform inversion.

[0102] This embodiment can improve the credibility of interpretation by using a small number of active source records to perform engineering verification on the passive source reconstruction results while maintaining the continuity and low interference of the passive source method.

[0103] Example 5 One embodiment of this disclosure provides a seismic record reconstruction system based on modal screening and virtual source response, comprising: The data acquisition module is used to acquire passive source seismic records during the drill-and-blast tunnel construction process; The preprocessing and candidate window filtering module is used to convert continuous passive source seismic records into multiple short-time multichannel records; The adaptive mode decomposition module is used to perform noise-assisted mode decomposition with joint energy-spectral entropy calibration on each short-time record to obtain multi-scale modes. The modal scoring and effective signal construction module is used to calculate the modal comprehensive score and construct the effective seismic signal based on the target frequency band contribution, narrowband spectral line penalty term, transient energy concentration and inter-trace coherence of each mode. The multi-reference robust interferometry module is used to perform phase-amplitude hybrid whitening and adaptive selection of reference traces on effective seismic signals, construct multi-reference weighted deconvolution seismic interferometry equations within a preset effective frequency band, and solve the virtual source response based on multiple reference traces. The passive source reconstruction module is used to perform multi-process processing on the virtual seismic source response to obtain pulsed virtual shot sets or passive source reconstruction records. The evaluation and result output module is used to evaluate the quality of the reconstruction results and then output the results.

[0104] Example 6 One embodiment of this disclosure provides a non-transitory computer-readable storage medium for storing computer instructions. When these computer instructions are executed by a processor, they implement the modal-filter-based passive seismic record reconstruction method.

[0105] Example 7 One embodiment of this disclosure provides an electronic device, including a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to implement the modal screening-based passive seismic record reconstruction method.

[0106] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0107] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0108] While the specific embodiments of this disclosure have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of this disclosure. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of this disclosure are still within the scope of protection of this disclosure.

Claims

1. A method for reconstructing passive seismic records while drilling based on modal screening, characterized in that, include: Acquire passive source seismic records during the drill-and-blast tunnel construction process, and convert continuous passive source seismic records into multiple short-time multichannel records; For each short-time record, perform noise-assisted mode decomposition with joint energy-spectral entropy calibration to obtain multi-scale modes; Based on the target frequency band contribution, narrowband spectral line penalty term, transient energy concentration and inter-channel coherence of each mode, the modal comprehensive score is calculated and an effective seismic signal is constructed. Phase-amplitude hybrid whitening and adaptive selection of reference traces are performed on the effective seismic signals. Multi-reference weighted deconvolution seismic interferometry equations are constructed within the preset effective frequency band, and the virtual source response is solved jointly based on multiple reference traces. The virtual seismic source response is processed through multiple processes to obtain pulsed virtual shot gathers or passive source reconstruction records. The reconstruction results are then evaluated for quality and output.

2. The method for reconstructing passive seismic records while drilling based on modal screening as described in claim 1, characterized in that, The acquisition of passive source seismic records during the drill-and-blast tunnel construction process, and the conversion of continuous passive source seismic records into multiple short-time multichannel records, includes: Multiple detectors were installed inside the tunnel, and the raw records of the multiple detectors were obtained by continuous sampling. The continuous recordings are divided into multiple segments according to time windows, and each window is preprocessed by removing the mean, removing the trend, stabilizing the amplitude, and band-limiting. Calculate the candidate window score. When the candidate window score is greater than a preset threshold, the current window enters the subsequent adaptive modal decomposition process.

3. The method for reconstructing passive seismic records while drilling based on modal screening as described in claim 1, characterized in that, The noise-assisted mode decomposition, which involves joint energy-spectral entropy calibration for each short-time record, yields multi-scale modes, including: The time-varying perturbation scale is determined based on local energy and local spectral entropy, and the ensemble perturbation record is calculated. Complete ensemble empirical mode decomposition is performed on the ensemble disturbance record to obtain multiple modal components and residual terms; The stable modal components are obtained by averaging over all sets, and finally the stable modal reconstruction of each window and each record is obtained.

4. The method for reconstructing passive seismic records while drilling based on modal screening as described in claim 1, characterized in that, The modal comprehensive score is constructed based on the target frequency band contribution, narrowband spectral line penalty term, transient energy concentration, and inter-channel coherence of each mode. Effective seismic signals are then constructed using soft-weighting or threshold filtering methods, including: For each mode, calculate its target frequency band contribution, narrowband spectral line penalty term, transient energy concentration and interchannel coherence. A comprehensive modal score is constructed by integrating the target frequency band contribution, narrowband spectral line penalty term, transient energy concentration, and interchannel coherence index. Based on the scoring, soft weights are constructed, and effective seismic signals are obtained by weighted superposition of various modes.

5. The method for reconstructing passive seismic records while drilling based on modal screening as described in claim 1, characterized in that, The phase-amplitude hybrid whitening and adaptive reference trace selection of the effective seismic signal includes: Perform a Fourier transform on the effective seismic signal to obtain the effective signal spectrum; Phase-amplitude hybrid whitening is performed within the operating frequency band, and the reference channel set is automatically determined based on the reference quality score.

6. The method for reconstructing passive seismic records while drilling based on modal screening as described in claim 1, characterized in that, The construction of multi-reference weighted deconvolution seismic interferometry equations within a preset effective frequency band, and the solution of virtual source responses based on multiple reference traces, includes: The receiver-implementation matrix is ​​composed of the frequency domain signals of all effective windows. Extract the rows corresponding to the reference channel set from the receiver-implementation matrix to form a reference submatrix, and set the weighted reference matrix; A multi-reference deconvolution interferometry model is established in the frequency domain and solved using weighted Tikhonov regularization. To suppress the influence of ill-conditioned frequencies on the reconstruction results, the condition number of the reference normal matrix is ​​defined to obtain the virtual source response.

7. The method for reconstructing passive seismic records while drilling based on modal screening as described in claim 1, characterized in that, The process of performing multi-process processing on the virtual source response to obtain pulsed virtual shot gathers or passive source reconstruction records, and outputting the reconstruction results after quality evaluation, includes: The obtained frequency domain virtual source response is smoothed by a frequency band window and subjected to inverse Fourier transform to obtain the virtual shot gather recovery formula; The causal and anticausal components are integrated based on energy and coherence. When construction noise lighting has obvious directionality, the positive and negative weights are automatically determined based on the energy of the positive and negative time delay reflection windows. The suitability of the reconstruction results for subsequent imaging or inversion is determined by calculating the energy salience of the reflection window, the continuity of adjacent channels, and the reference consistency of the active source. The final output includes reconstructed virtual shot gathers, reflection coherence indices, continuity indices, active source consistency, apparent velocities, and event bands that can be used for advanced geological interpretation.

8. A seismic record reconstruction system based on modal screening and virtual source response, characterized in that, include: The data acquisition module is used to acquire passive source seismic records during the drill-and-blast tunnel construction process; The preprocessing and candidate window filtering module is used to convert continuous passive source seismic records into multiple short-time multichannel records; The adaptive mode decomposition module is used to perform noise-assisted mode decomposition with joint energy-spectral entropy calibration on each short-time record to obtain multi-scale modes. The modal scoring and effective signal construction module is used to calculate the modal comprehensive score and construct the effective seismic signal based on the target frequency band contribution, narrowband spectral line penalty term, transient energy concentration and inter-trace coherence of each mode. The multi-reference robust interferometry module is used to perform phase-amplitude hybrid whitening and adaptive selection of reference traces on effective seismic signals, construct multi-reference weighted deconvolution seismic interferometry equations within a preset effective frequency band, and solve the virtual source response based on multiple reference traces. The passive source reconstruction module is used to perform multi-process processing on the virtual seismic source response to obtain pulsed virtual shot sets or passive source reconstruction records. The evaluation and result output module is used to evaluate the quality of the reconstruction results and then output the results.

9. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium is used to store computer instructions, which, when executed by a processor, implement the modal screening-based passive seismic record reconstruction method for drilling as described in any one of claims 1-7.

10. An electronic device, characterized in that, include: The device includes a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to perform the modal screening-based passive seismic record reconstruction method as described in any one of claims 1-7.

Images