Three-component data processing system and method for vector seismic source directivity evaluation
By combining a multi-station three-component geophone array with a data processing terminal, and employing vertical excitation orientation reference conditions and circumferential geometric mean normalization, the stability problem of quantitative evaluation of vector source directionality was solved, achieving high reliability and repeatability evaluation under complex environments.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHINA UNIV OF MINING & TECH
- Filing Date
- 2026-01-23
- Publication Date
- 2026-05-12
AI Technical Summary
Existing technologies struggle to achieve stable and repeatable quantitative evaluation of vector source directionality under multi-station three-component observation conditions, and lack a unified processing procedure to effectively mitigate the effects of site effects and system differences, resulting in insufficient comparability of directionality results between different operating conditions or test batches.
By combining a multi-station three-component geophone array with a data processing terminal, and by introducing a vertical excitation orientation reference condition, combined with circumferential geometric mean normalization and robust statistical processing, the interference of non-directional factors is weakened, and the objective quantification of the vector source directionality is achieved.
It significantly improves the reliability and repeatability of directional evaluation results, provides physical basis at the frequency domain level, is suitable for lateral comparative analysis under different excitation conditions and observation conditions, and enhances the detection effect and interpretation accuracy under complex geological conditions.
Smart Images
Figure CN122018039A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of geophysical exploration technology, specifically a three-component data processing system and method for vector source directionality assessment. Background Technology
[0002] Seismic exploration is a key technology in geology and geophysics used to detect underground structures. It involves generating seismic waves at the surface or underground and recording their propagation response in the subsurface medium to infer the properties, structural morphology, and distribution of anomalies in the subsurface rock layers. Traditional seismic exploration methods typically rely on artificial seismic sources such as explosives and impact tests to actively acquire seismic data. However, in engineering scenarios such as mine tunnels, chambers, and highly heterogeneous media, anomaly characteristic signals are prone to aliasing due to factors such as wave field scattering, multiple waves, and background noise, leading to increased uncertainty in exploration interpretation and limiting imaging accuracy and target identification capabilities.
[0003] To improve detection performance in complex environments, new types of seismic sources with directional excitation capabilities, such as vector sources, have emerged in recent years. Vector sources can concentrate energy in a specific orientation by controlling the excitation direction, thereby enhancing the effective illumination of targets to the side or at depth, and have the potential to improve the response characteristics of anomalies and increase detection range. However, the practical application effect of vector sources is highly dependent on their radiation directionality: on the one hand, energy distribution varies significantly under different excitation directions; on the other hand, influenced by site conditions, propagation path differences, station coupling, and installation conditions, the observed amplitude variations often include both directional and non-directional factors, easily leading to deviations in directionality judgment, which in turn affects the selection of the dominant excitation direction and the optimization of the detection scheme.
[0004] Current methods for evaluating the directionality of vector seismic sources often rely on amplitude comparisons from single stations or a small number of measurement points, or on empirical qualitative interpretations. These methods struggle to achieve stable and repeatable quantitative evaluations under multi-station three-component observation conditions. Furthermore, the lack of a unified processing procedure to effectively mitigate the influence of site effects and system differences results in insufficient comparability of directionality results across different operating conditions or test batches. Therefore, this invention aims to provide a new quantitative method for evaluating directionality that, while ensuring full utilization of multi-station three-component data, can suppress interference from non-directional factors and objectively quantify the directional intensity and dominant excitation direction of vector seismic sources. Summary of the Invention
[0005] To address the problems existing in the prior art, this invention provides a three-component data processing system and method for vector source directionality assessment. While ensuring full utilization of three-component data from multiple stations, it can suppress interference from non-directional factors and objectively quantify the directional intensity and dominant excitation direction of the vector source, providing data support for subsequent use.
[0006] To achieve the above objectives, the technical solution adopted by the present invention is: a three-component data processing system for vector source directionality assessment, comprising a vector source, a multi-station three-component geophone array, a synchronous acquisition unit, and a data processing terminal.
[0007] The vector source is used to generate seismic waves in the corresponding orientation under different excitation orientation conditions.
[0008] The multi-station three-component geophone array is arranged around the vector seismic source in a preset orientation to form an equidistant circumferential observation system.
[0009] The synchronous acquisition unit is used to synchronously control the timing of data acquisition by the three-component detector array of multiple stations when the vector seismic source is in each excitation orientation condition, and output a unified time reference.
[0010] The data processing terminal is used to perform directional evaluation calculations on three-component seismic data collected from multiple stations under different excitation orientations and output the evaluation results.
[0011] Furthermore, the data processing terminal includes a preprocessing module, a coordinate rotation module, a time window determination module, a time-frequency energy analysis module, a multi-band feature extraction module, a station consistency correction module, a circumferential normalization module, a reference condition background suppression module, an abnormal station identification and robust statistics module, a directional function fitting module, a directional index and confidence assessment module, and a visualization output module. The preprocessing module is used to preprocess the three-component waveforms; the coordinate rotation module is used to rotate the three-component seismic data acquired by each station from instrument coordinates to radial-tangential-vertical coordinates; the time window determination module is used to determine the effective wavefield window required for directional evaluation; and the time-frequency energy analysis module is used to preprocess the three-component waveforms. The system performs time-frequency analysis on the waveform; a multi-band feature extraction module extracts energy / amplitude features within multiple preset frequency bands; a station consistency correction module compensates for station gain and coupling differences; a circumferential normalization module normalizes the data; a reference condition background suppression module introduces vertical excitation towards the reference condition to weaken non-directional effects; an abnormal station identification and robust statistics module identifies and removes abnormal stations; a directional function fitting module inverts the azimuth and intensity parameters of the directional main radiation; a directional index and confidence assessment module determines the directional index and provides confidence intervals; and a visualization output module outputs directional rose diagrams, azimuth scatter plots, and evaluation reports.
[0012] Furthermore, the different excitation orientation conditions include a first orientation excitation condition, a second orientation excitation condition, and a no-orientation reference excitation condition; wherein the excitation seismic waves in the first orientation excitation condition and the second orientation excitation condition have different orientations, which are used to generate wavefield responses with directional differences; the vertical excitation reference condition is used to excite seismic waves perpendicular to the vector excitation planes of the first and second conditions without introducing horizontal directional bias, thereby providing a reference wavefield response for background suppression.
[0013] Furthermore, the time window determination module adopts a fixed time window or an adaptive time window.
[0014] The evaluation method for the above three-component data processing system includes the following steps: Step 1: Based on the target area and deployment conditions, deploy a multi-station three-component geophone array around the vector source according to a preset orientation, and determine the azimuth angle of each station in the multi-station three-component geophone array relative to the source to form an equidistant circumferential observation system.
[0015] Step 2: Trigger vector sources to generate seismic waves under the first orientation excitation condition, the second orientation excitation condition, and the vertical excitation reference condition, respectively, and control the multi-station three-component geophone array through the synchronous acquisition unit to synchronously acquire the multi-station three-component seismic record dataset.
[0016] Step 3: Preprocess and time-correct the dataset obtained in Step 2 to obtain the three-component waveforms of each station under a unified time reference.
[0017] Step 4: Convert the three-component waveforms of each station to the radial-tangential-vertical coordinate system to obtain the R, T, and Z three-component waveforms and their composite vector components.
[0018] Step 5: Determine the time window for directional evaluation.
[0019] Step Six: Within the directional evaluation time window determined in Step Five, perform time-frequency analysis on the three-component waveforms and their composite vector components of each station under each excitation condition obtained in Step Four, in order to obtain the waveform energy distribution characteristics of different excitation methods in the frequency dimension.
[0020] Step 7: Based on the waveform energy distribution characteristics obtained in Step 6, extract the amplitude / energy index of each station in multiple frequency bands, then construct the circumferential amplitude vector and fusion index, and correct and normalize the station consistency.
[0021] Step 8: Use the normalized result of the vertical excitation orientation reference condition to suppress the ratio of the orientation condition, and obtain the effective directional circumferential vector to weaken the non-directional effects such as site effect, path difference and installation difference.
[0022] Step 9: Based on the excitation orientation, define the forward and backward sectors, perform abnormal station elimination and robust statistics on the effective directional circumferential vector, construct the directional index, and calculate the directional intensity.
[0023] Step 10: Fit the directional function to the circumferential distribution to obtain the directional main radiation azimuth and directional intensity parameters, and then weight and fuse the multi-component and multi-band results.
[0024] Step 11: Output the directionality index, main radiation azimuth, and visualization results to provide a basis for selecting the dominant excitation direction and optimizing the detection scheme.
[0025] Furthermore, the preprocessing in step three includes sequentially performing DC / desiccating processing, filtering processing, and noise reduction processing.
[0026] Furthermore, step four specifically involves converting the horizontal two components of each station's azimuth angle determined in step one into radial components. With tangential component and with vertical component Together they form a radial-tangential-vertical coordinate system, yielding the R, T, and Z three-component waveforms and their composite vector components; by constructing a unified physical coordinate system, the physical consistency of the three-component directionality evaluation is improved.
[0027] Furthermore, in step five, the directionality evaluation time window adopts either a fixed time window or an adaptive time window; wherein the adaptive time window determines the start and end times of the time window through AIC first arrival picking, envelope threshold, or STA / LTA triggering, so as to reduce the impact of first arrival drift and noise changes on the stability of the directionality index under different operating conditions.
[0028] Furthermore, in step seven, the amplitude / energy index of each station is extracted under multiple frequency bands. Specifically, the waveform energy distribution characteristics are divided into multiple frequency bands and the characteristics are extracted separately. The frequency band weight is determined by the signal-to-noise ratio, in-window energy, or repetitive excitation consistency to improve the adaptability of the directional evaluation to different frequency components.
[0029] Furthermore, in step nine, abnormal stations are removed and robust statistics are performed on the effective directional features. Specifically, the MAD / box line method is used to remove abnormal stations from the effective directional features, and the median, truncated mean, or Huber estimate is used to replace the ordinary mean in order to reduce the impact of individual station failures on the directional results.
[0030] Compared with the prior art, the present invention has the following advantages: 1. This invention employs a combination of multi-station three-component detector arrays and data processing terminals to acquire three-component seismic record data under different excitation orientation conditions and a vertical excitation orientation reference condition. By introducing the vertical excitation orientation reference condition as a background benchmark and combining circumferential geometric mean normalization and robust statistical processing, the influence of site effects, propagation path differences, and station coupling condition variations on the observed amplitude is effectively reduced. This achieves the separation and stable characterization of the vector source directional contribution, significantly improving the reliability and repeatability of the directional evaluation results.
[0031] 2. Within the directional evaluation time window, this invention not only extracts amplitude / energy indices from the three-component signals of each station, but also introduces time-frequency analysis to characterize the component energy distribution features in different frequency bands. By comparing the energy distribution differences in the frequency dimension between the vector excitation condition and the vertical excitation orientation reference condition, it reveals the manifestation of vector source directionality at the frequency domain and component coupling level, providing a physical basis at the frequency domain level for quantitative evaluation of directionality, and avoiding the problem of insufficient interpretation that may be caused by relying solely on a single time-domain amplitude index.
[0032] 3. This invention establishes forward and backward sector grouping based on excitation orientation and constructs a directional index in the form of logarithmic ratio to objectively quantify the directional intensity of vector seismic sources. The directional index can comprehensively reflect information from multiple stations, multiple components, and multiple frequency bands, and has the characteristics of scale independence, strong stability, and high comparability. It is suitable for horizontal comparative analysis under different excitation conditions, different test batches, and different observation conditions.
[0033] 4. This invention can output directional evaluation results for multiple components, including radial, tangential, vertical, and vector synthesis, and combine them with directional rose diagrams, azimuth scatter plots, and time-frequency energy distribution characteristics for comprehensive visualization. This makes the directional differences of vector sources under different azimuths, components, and frequencies more intuitive and clear. Thus, it provides quantitative basis for source orientation selection, observation system optimization, and operational parameter configuration in anomaly detection, improving the detection effect and interpretation accuracy under complex geological conditions. Attached Figure Description
[0034] Figure 1 This is a diagram of the equidistant circumferential observation system used in the data acquisition of this invention. Detailed Implementation
[0035] The present invention will be further described below.
[0036] like Figure 1As shown, a three-component data processing system for vector source directionality assessment includes a vector source, a multi-station three-component geophone array, a synchronous acquisition unit, and a data processing terminal. The vector source is used to generate seismic waves in corresponding orientations under different excitation orientation conditions. The different excitation orientation conditions include a first orientation excitation condition, a second orientation excitation condition, and a vertical excitation reference condition. The excitation seismic waves generated under the first orientation excitation condition and the second orientation excitation condition have different orientations, which are used to generate wavefield responses with directional differences. The vertical excitation reference condition is used to generate seismic waves perpendicular to the vector excitation planes of the first and second conditions without introducing horizontal directional bias, thereby providing a reference wavefield response for background suppression.
[0037] The multi-station three-component geophone array is arranged around the vector source according to a preset orientation, forming an equidistant circumferential observation system. Figure 1 As shown.
[0038] The synchronous acquisition unit is used to synchronously control the timing of data acquisition by the three-component detector array of multiple stations when the vector seismic source is in each excitation orientation condition, and output a unified time reference.
[0039] The data processing terminal is used to perform directional evaluation calculations on three-component seismic data acquired from multiple stations under different excitation orientations and output the evaluation results. It includes a preprocessing module, a coordinate rotation module, a time window determination module, a time-frequency energy analysis module, a multi-band feature extraction module, a station consistency correction module, a circumferential normalization module, a reference condition background suppression module, an abnormal station identification and robust statistics module, a directional function fitting module, a directional index and confidence assessment module, and a visualization output module. The preprocessing module is used to preprocess the three-component waveforms; the coordinate rotation module is used to rotate the three-component seismic data acquired from each station from instrument coordinates to radial-tangential-vertical coordinates; and the time window determination module is used to determine the effective wavefield window required for directional evaluation. The system includes a fixed time window or an adaptive time window; a time-frequency energy analysis module for time-frequency analysis of three-component waveforms; a multi-band feature extraction module for extracting energy / amplitude features within multiple preset frequency bands; a station consistency correction module for compensating for station gain and coupling differences; a circumferential normalization module for normalizing data; a reference condition background suppression module for introducing vertical excitation towards the reference condition to weaken non-directional effects; an abnormal station identification and robust statistics module for identifying and removing abnormal stations; a directional function fitting module for inverting the azimuth and intensity parameters of the directional main radiation; a directional index and confidence assessment module for determining the directional index and providing confidence intervals; and a visualization output module for outputting directional rose diagrams, azimuth scatter plots, and evaluation reports.
[0040] The evaluation method for the above three-component data processing system includes the following steps: Step 1: Based on the target area and deployment conditions, deploy an N-station three-component geophone array around the vector source according to a preset azimuth, and determine the azimuth angle of each station in the multi-station three-component geophone array relative to the source to form an equidistant circumferential observation system; in this embodiment, the set of azimuth angles of each station relative to the source is measured or calculated. Furthermore, it is preferable to satisfy equal angular intervals (in this embodiment, there are 8 units, with angles of 0°, 45°, ... 315°); and the acquisition parameters, including the sampling interval, are set. Recording duration T, preset evaluation time window Or an adaptive window strategy.
[0041] Step 2: Trigger the vector source to generate seismic waves under the first orientation excitation condition, the second orientation excitation condition, and the vertical excitation reference condition, respectively. Let k be the excitation condition number. In this embodiment, k=1 for the first orientation excitation condition, and the excitation orientation is... The second orientation excitation condition is k=2, and the excitation orientation is... Vertical excitation reference case k=ref (excitation direction perpendicular to the plane of the previous two vector excitations); under each case, vector source excitation is triggered, and the synchronous acquisition unit is started simultaneously to acquire three-component records from all stations, forming a three-component seismic record dataset (data matrix): Each station corresponds to three columns ( , , ), Where N is the number of sampling points, and N represents the total number of stations. In this embodiment, N=8. Repeated excitation can be performed multiple times, preferably M times for each operating condition, for subsequent consistency assessment and confidence calculation.
[0042] Step 3: Preprocess and time-correct the dataset obtained in Step 2 to obtain the three-component waveforms of each station under a unified time reference; in this embodiment, the discrete waveform of the c-th component of the i-th station is denoted as... The sampling interval is dt, n=0,1,… For each operating condition data implement: 1) DC removal: Mean removal is performed on the waveform of each component; in, This represents the mean waveform value of the c-th component of the i-th station under the k-th operating condition; This represents the waveform after DC removal.
[0043] 2) Linear detrending, least squares fitting and removal: Let the trend term be... ,in If the parameter is a linear trend term, then... in, This represents the waveform after detrending.
[0044] 3) Bandpass filtering: A bandpass filter is used according to the preset frequency band. Perform filtering, passband Then the filtered output is in convolution form: in, This represents the filtered waveform.
[0045] 4) Timing alignment: A unified time zero point is obtained based on the earthquake source trigger signal.
[0046] in, This represents the timing alignment offset for the k-th operating condition; This represents the waveform after timing alignment.
[0047] 5) Data Truncation: A uniform length is used for the three-component models under different operating conditions; the truncation length is standardized to [length to be specified]. Each sampling point is matched with the recording duration of the synchronous acquisition unit to ensure consistency in subsequent indexing.
[0048] Step 4: Based on the azimuth angles of each station determined in Step 1, convert their horizontal components into radial components. With tangential component and with vertical component Together they form a radial-tangential-vertical coordinate system; in this embodiment, the horizontal component of the detector is set as the instrument coordinate system. , ), then based on the azimuth angle Rotate it into radial and tangential components: and with the vertical component Together they constitute a unified physical coordinate system. , , The three-component waveforms of R, T, and Z and their composite vector components are obtained; the physical consistency of the three-component directionality evaluation is improved by constructing a unified physical coordinate system.
[0049] Step 5: Determine the directionality evaluation time window, which can be a fixed time window or an adaptive time window. The adaptive time window determines the start and end times of the time window through AIC first-arrival pickup, envelope threshold, or STA / LTA triggering, to reduce the impact of first-arrival drift and noise variations on the stability of the directionality index under different operating conditions. In this embodiment, if a fixed window method is used, a preset evaluation time window is directly adopted. If an adaptive window strategy is used, then the time window is determined for each station or jointly for all stations, and the three-component signal of the k-th operating condition of that station is denoted as... , , The window center time is defined as... Based on different algorithms, the following can be determined: (a) Adopt Pick up at the time; structure function: ;in, This represents any single-component seismic signal used for pickup at that time. Indicates the waveform length. Indicates variance; The minimum value corresponds to the starting point of the seismic phase: .
[0050] (b) Adopt (Short-time / long-time energy ratio) Pick-up method.
[0051] Define short window energy With long window energy : in These represent the length of the short window and the length of the long window, respectively.
[0052] Energy ratio trigger function: Take the seismic phase of the i-th station at time The earliest time that meets the following conditions is taken as the arrival time of the seismic phase: in This indicates the trigger threshold.
[0053] After determining the arrival time of the seismic phase using step (a) or (b), then... Centered on the left and right, set fixed half-widths. The effective evaluation window of the station is obtained: that is, the effective evaluation window. .
[0054] Finally, robust statistics are taken from all stations. In this embodiment, the median is used, and the specific formula is as follows: Take the robust summary of all valid evaluation windows of all stations as the directional evaluation time window. .
[0055] Step Six: Within the directional evaluation time window determined in Step Five, perform time-frequency analysis on the radial component R, tangential component T, vertical component Z, and their composite vector component to obtain the waveform energy distribution characteristics of different excitation methods in the frequency dimension. In this embodiment, let the c-th component signal of the i-th station under the k-th operating condition be: Within the directional evaluation time window determined in step five, the above component signals are subjected to time-frequency transformation or frequency domain processing to obtain the corresponding time-frequency energy spectrum: By integrating the time-frequency energy spectrum within the directionality evaluation time window, the energy distribution function at each frequency is obtained: Based on a preset frequency band division scheme (such as low frequency, mid frequency, and high frequency), the frequency axis is integrated piecewise to obtain the component energy index for each frequency band: Where b represents the b-th frequency band.
[0056] Finally, for all stations under the same operating condition, the energy distribution indicators of each component in each frequency band are statistically analyzed to form a four-dimensional energy distribution feature set of "operating condition-station-component-frequency band", which is used for subsequent circumferential normalization, reference suppression and directional analysis.
[0057] Step 7: Based on the waveform energy distribution characteristics obtained in Step 6, extract the amplitude / energy index of each station in multiple frequency bands. Specifically, divide the waveform energy distribution characteristics into multiple frequency bands and extract features separately. The frequency band weights are determined by the signal-to-noise ratio, in-window energy, or repetitive excitation consistency to improve the adaptability of directional evaluation to different frequency components. In this embodiment, for each working condition... Each station The amplitude index is calculated using RMS within the directional evaluation time window: in, Indicates the station number. Indicates the activation condition number. This represents the set of sampling points within the evaluation time window; This represents the amplitude index of the i-th station in the radial component under the k-th operating condition; This represents the time signal of the i-th station in the radial component; This represents the amplitude index of the i-th station in the tangential component under the k-th operating condition; This represents the time signal of the i-th station in the tangential component; This represents the amplitude index of the i-th station in the vertical component under the k-th operating condition; This represents the time signal of the i-th station in the vertical component, and then a vector synthesis index is constructed: The circumferential amplitude vector for each index is obtained: Where N represents the total number of valid stations participating in the directional evaluation; T represents the transpose operation.
[0058] Next, set the frequency band set. (e.g., multiple frequency bands including low frequency, mid frequency, and high frequency); perform steps three through six for each frequency band to obtain... ; Calculate the frequency band weight , The fusion index is determined by the signal-to-noise ratio, in-window energy, or repetitive excitation consistency. Finally, station consistency is corrected and circumferentially normalized, specifically: for each operating condition k, the circumferential amplitude vector... Correction is performed by calculating the geometric mean: The normalized result is obtained: By using geometric mean correction and normalization, the overall amplitude drift is eliminated, thereby reducing the overall energy fluctuation and station coupling differences.
[0059] Step 8: Utilize vertical excitation reference conditions normalization results The ratio suppression is applied to the orientation condition, and the specific formula is as follows: ( (The denominator should be a very small positive number to avoid it being zero or too small.) The above equation yields the effective directional circumferential vector. Used to mitigate non-directional effects such as site effects.
[0060] Step 9: Define the forward and backward sectors based on the excitation orientation, specifically: based on the excitation orientation... Define forward and backward sectors: Where F represents the set of station indexes within the forward angle range of the main excitation direction of the seismic source under the k-th excitation condition; B represents the set of station indexes within the reverse (180°) angle range of the main excitation direction of the seismic source under the k-th excitation condition. For station indexing, , where N represents the total number of valid stations participating in the directional evaluation. This represents the geometric azimuth angle from the epicenter location to the i-th station; This represents the azimuth angle of the main radiation direction (nozzle orientation) of the vector source in the geodetic coordinate system under the kth excitation condition; These are the half-angle widths of the forward sector and the backward sector, respectively. When the angle is 45°, the total width of the forward sector is 90°, and the same principle applies. .
[0061] Abnormal station removal: For valid directional circumferential vectors Outliers were removed using the MAD / box method.
[0062] For robust statistics, the median or truncated mean is used to replace the mean in the data within the sector; then the directional index (in logarithmic ratio form) is calculated: in, and These represent the statistical average values of the effective amplitude indices within the forward and backward station sets, respectively. The effective amplitude indices have been obtained through circumferential normalization and suppression under reference conditions. By taking the natural logarithm of the ratio of forward to backward amplitudes, the strength of the source directionality is quantitatively characterized.
[0063] The above formula outputs the corresponding R / T / Z / V indicators respectively. , , , .in, This indicates that the forward average amplitude is greater than the backward amplitude, suggesting that there is a forward-enhancing directionality in this operating condition; This indicates a stronger divergence; It indicates that the front and back are basically symmetrical; The larger it is, the stronger its directionality.
[0064] Step 10: Fit a directional function to the circumferential distribution. The specific formula is as follows: The azimuth of the main radiation can be calculated using the above formula. And directional intensity parameters; and perform confidence assessment as needed, specifically: perform bootstrap resampling on the repeated excitation data to obtain confidence interval And classify the significance level (significant / moderate / insignificant) according to the threshold.
[0065] Step 11: Output the directivity index, main radiation azimuth, confidence interval, and visualization results. In this embodiment, this includes: the directivity index. , , , Directional rose diagram, azimuth scatter plot, forward / backward sector labeling results; main radiation azimuth. The report includes the following: a selection of advantageous stimulation directions; an evaluation report containing parameters, excluded stations, and confidence intervals, providing a basis for selecting advantageous stimulation directions and optimizing detection schemes.
[0066] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.
Claims
1. A three-component data processing system for vector source directionality assessment, characterized in that, It includes a vector source, a multi-station three-component geophone array, a synchronous acquisition unit, and a data processing terminal; The vector source is used to generate seismic waves in the corresponding orientation under different excitation orientation conditions; The multi-station three-component geophone array is arranged around the vector seismic source in a preset orientation to form an equidistant circumferential observation system; The synchronous acquisition unit is used to synchronously control the timing of data acquisition by the three-component detector array of multiple stations when the vector seismic source is in each excitation orientation condition, and output a unified time reference. The data processing terminal is used to perform directional evaluation calculations on three-component seismic data collected from multiple stations under different excitation orientations and output the evaluation results.
2. The three-component data processing system for vector source directionality assessment according to claim 1, characterized in that, The data processing terminal includes a preprocessing module, a coordinate rotation module, a time window determination module, a time-frequency energy analysis module, a multi-band feature extraction module, a station consistency correction module, a circumferential normalization module, a reference condition background suppression module, an abnormal station identification and robust statistics module, a directional function fitting module, a directional index and confidence assessment module, and a visualization output module. The preprocessing module is used to preprocess the three-component waveforms; the coordinate rotation module is used to rotate the three-component seismic data acquired from each station from instrument coordinates to radial-tangential-vertical coordinates; the time window determination module is used to determine the effective wavefield window required for directional evaluation; and the time-frequency energy analysis module is used to preprocess the three-component waveforms. The system performs time-frequency analysis; a multi-band feature extraction module extracts energy / amplitude features within multiple preset frequency bands; a station consistency correction module compensates for station gain and coupling differences; a circumferential normalization module normalizes the data; a reference condition background suppression module introduces vertical excitation towards the reference condition to weaken non-directional effects; an abnormal station identification and robust statistics module identifies and removes abnormal stations; a directional function fitting module inverts the azimuth and intensity parameters of the directional main radiation; a directional index and confidence assessment module determines the directional index and provides confidence intervals; and a visualization output module outputs directional rose diagrams, azimuth scatter plots, and evaluation reports.
3. The three-component data processing system for vector source directionality assessment according to claim 2, characterized in that, The different excitation orientation conditions include the first orientation excitation condition, the second orientation excitation condition, and the vertical excitation reference condition; The first orientation excitation condition and the second orientation excitation condition have different orientations of the excitation seismic waves, which are used to generate wavefield responses with directional differences; the vertical excitation reference condition is used to excite seismic waves perpendicular to the vector excitation planes of the first and second conditions without introducing horizontal directional bias, thereby providing a reference wavefield response for background suppression.
4. The three-component data processing system for vector source directionality assessment according to claim 2, characterized in that, The time window determination module uses either a fixed time window or an adaptive time window.
5. An evaluation method for the three-component data processing system according to claim 3, characterized in that, The steps include the following: Step 1: Based on the target area and deployment conditions, deploy the multi-station three-component geophone array around the vector source according to the preset orientation, and determine the azimuth angle of each station in the multi-station three-component geophone array relative to the source to form an equidistant circumferential observation system. Step 2: Trigger the vector source to generate seismic waves under the first orientation excitation condition, the second orientation excitation condition, and the vertical excitation orientation reference condition, and control the multi-station three-component geophone array through the synchronous acquisition unit to synchronously acquire the multi-station three-component seismic record dataset. Step 3: Preprocess and time-correct the dataset obtained in Step 2 to obtain the three-component waveforms of each station under a unified time reference. Step 4: Convert the three-component waveforms of each station to the radial-tangential-vertical coordinate system to obtain the R, T, Z three-component waveforms and their composite vector components; Step 5: Determine the time window for directional evaluation; Step 6: Within the directional evaluation time window determined in Step 5, perform time-frequency analysis on the three-component waveforms and their composite vector components of each station under each excitation condition obtained in Step 4, in order to obtain the waveform energy distribution characteristics of different excitation methods in the frequency dimension. Step 7: Based on the waveform energy distribution characteristics obtained in Step 6, extract the amplitude / energy index of each station in multiple frequency bands, then construct the circumferential amplitude vector and fusion index, and correct and normalize the station consistency. Step 8: Use the normalized result of the vertical excitation orientation reference condition to suppress the ratio of the orientation condition, and obtain the effective directional circumferential vector. Step 9: Based on the excitation orientation, define the forward and backward sectors, perform abnormal station elimination and robust statistics on the effective directional circumferential vector, construct the directional index and calculate the directional intensity; Step 10: Fit the directional function to the circumferential distribution to obtain the directional main radiation azimuth and directional intensity parameters; Step 11: Output the directionality index, main radiation azimuth, and visualization results to provide a basis for selecting the dominant excitation direction and optimizing the detection scheme.
6. The evaluation method according to claim 5, characterized in that, The preprocessing in step three includes sequentially performing DC / desiccating processing, filtering processing, and noise reduction processing.
7. The evaluation method according to claim 5, characterized in that, Step four specifically involves converting the horizontal components of each station's azimuth angle determined in step one into radial components. With tangential component and with vertical component Together they form a radial-tangential-vertical coordinate system, yielding the R, T, and Z component waveforms and their composite vector components.
8. The evaluation method according to claim 5, characterized in that, In step five, the directionality evaluation time window can be a fixed time window or an adaptive time window; the adaptive time window determines the start and end times of the time window through AIC first arrival picking, envelope threshold, or STA / LTA trigger.
9. The evaluation method according to claim 5, characterized in that, In step seven, the amplitude / energy index of each station is extracted in multiple frequency bands. Specifically, the waveform energy distribution characteristics are divided into multiple frequency bands and the characteristics are extracted separately. The frequency band weight is determined by the signal-to-noise ratio, the energy within the window, or the consistency of repeated excitation.
10. The evaluation method according to claim 5, characterized in that, In step nine, abnormal stations are removed and robust statistics are performed on the effective directional features. Specifically, the MAD / box line method is used to remove abnormal stations from the effective directional features, and the median, truncated mean, or Huber estimate is used to replace the ordinary mean.