Static correction method based on multi-horizon constraint and computer readable storage medium
By using a multi-layer bit constraint static correction method in seismic data processing, the multi-layer bit mutual correlation coefficient is calculated to determine the remaining static correction amount, which solves the problem of excessive information caused by single-layer bit calculation and improves the accuracy of static correction results.
Patent Information
- Application Number
- CN202311418372.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-10-30
- Publication Date
- 2025-05-02
AI Technical Summary
When calculating the static correction amount of seismic data, the prior art relies only on the information of a single layer, resulting in excessive impact of local information on the entire data and affecting the accuracy of the results.
The static correction method based on multi-layer bit constraints is adopted. By calculating the mutual correlation coefficient on each layer bit and fusing it to obtain the multi-layer bit mutual correlation coefficient, the time shift corresponding to the maximum multi-layer bit mutual correlation coefficient is finally used as the residual static correction amount.
This method can more comprehensively consider the information of multiple layers, improve the accuracy of static correction results, and overcome the defect of excessive local information caused by single-layer calculation.
Smart Images
Figure CN119916477A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of seismic data processing, and in particular relates to a static correction method based on multi-layer bit constraints and a computer-readable storage medium. Background Art
[0002] In seismic data processing, more accurate static correction calculation results play a very important role in underground structure imaging and the accuracy of seismic wave velocity. There are many methods for calculating residual static correction, including correlation method, generalized linear inversion method, nonlinear inversion method, etc., but these methods have various defects. The maximum energy method is a commonly used method in seismic data processing. Its basic principle is that after applying the correct static correction, the stacking energy of the seismic profile reaches a maximum value. According to this principle, the calculation of surface consistency residual static correction is considered as an optimization problem. Wu Bo et al. took into account the correlation between the reflection waves from adjacent CMP data sets in the same layer, so they proposed a weighted mixed wave model.
[0003] When using the maximum energy method to calculate the static correction, Pan Shulin et al. adopted a calculation time window, which tracks along the continuous reflection phase axis. That is, a layer of seismic data is first selected, and the calculation time window is expanded according to the layer. This method can fully utilize the information provided by the same axis to calculate the static correction. However, if the static correction of all seismic data channels is calculated based on the information in only one layer, the local information of the layer will have too much influence on the entire data. Summary of the invention
[0004] The purpose of the present invention is to solve the above-mentioned problems existing in the prior art and to provide a static correction method based on multi-layer position constraints.
[0005] The present invention is implemented by the following technical scheme: A static correction method based on multi-layer position constraints comprises the following steps:
[0006] At each layer, the mutual correlation coefficient between each data in the gather and the corresponding standard trace at different time shifts is calculated; the mutual correlation coefficients at the same time shift at each layer are fused to obtain the multi-layer mutual correlation coefficient; the time shift corresponding to the maximum multi-layer mutual correlation coefficient is taken as the residual static correction.
[0007] Furthermore, the construction method of the standard track includes the following steps: intercepting each layer on the track set through a time window and picking up the sampling points of each data track in the time window to construct a track set within the window; constructing a superimposed track according to the track set within the window; and constructing a corresponding standard track for each data track in the track set within the window according to the superimposed track.
[0008] Furthermore, for each horizon, the calculation formula of the stack trace is as follows:
[0009]
[0010] In the formula, data stack [k] represents the data point on the stacked trace; m represents the number of traces in the window; data j [k] represents the kth sampling point on the jth channel, k = 0, 1, ..., (t2-t1), t2-t1 represents the time window length.
[0011] Furthermore, for each data track, the calculation steps of the standard track include:
[0012] Calculate the difference: data diff [k] = data stack [k]-data[k]
[0013] Find the maximum error data Max =max{data diff [k], k = 0, 1, ..., (t2-t1)};
[0014] Calculation standard:
[0015] In the formula, data prefer [k] represents the data point on the standard track; data[k] represents the kth sampling point on a data track in the window, k = 0, 1, ..., (t2-t1). diff [k] represents the track error; data stack Indicates the data points on the superposition track; data Max Indicates the maximum track error.
[0016] Furthermore, the in-window gather is constructed according to the CMP gather, and the cross-correlation coefficient between each data in the in-window gather and the corresponding standard data at different time shifts is calculated, and the cross-correlation coefficients of the same shot point at the same time shift and the cross-correlation coefficients of the same detection point at the same time shift are accumulated to obtain the accumulated cross-correlation coefficients of the same shot point and the accumulated cross-correlation coefficients of the same detection point.
[0017] Furthermore, the calculation formula of the cumulative correlation coefficient of the same shot point is as follows:
[0018] scramp[srcid][i]+=coef[i]
[0019] The calculation formula of the cumulative correlation coefficient of the same detection point is as follows:
[0020] recamp[recid][i]+=coef[i]
[0021] Where i represents the time shift, i∈[-range,range], i is zero when there is no time shift, is negative when the time shift is in the direction of time decrease, and is positive when the time shift is in the direction of time increase; coef[i] represents the mutual correlation coefficient under time shift i; scramp[srcid][i] represents the cumulative sum of the mutual correlation coefficients of the shot point number srcid under time shift i, that is, the cumulative mutual correlation coefficients of the same shot point; recamp[recid][i] represents the cumulative sum of the mutual correlation coefficients of the detection point number srcid under time shift i, that is, the cumulative mutual correlation coefficients of the same detection point.
[0022] Furthermore, the accumulated mutual correlation coefficients of the same shot points at different levels are added together to obtain the multi-layer mutual correlation coefficients of the common shot points: the accumulated mutual correlation coefficients of the same detection points at different levels are added together to obtain the multi-layer mutual correlation coefficients of the common detection points.
[0023] Furthermore, it is used to perform static correction on the post-stack data. Each data in the post-stack data is formed by stacking a CMP gather. The sampling iterative algorithm solves the optimal residual static correction amount, and the first convergence condition is set as:
[0024] amp[iter]-amp[iter-1] <s,
[0025] Where s represents the threshold; amp[iter] and amp[iter-1] represent the accumulated energy of all superimposed traces at the iter-th iteration and iter-1-th iteration, respectively;
[0026] The calculation formula of the cumulative energy of all superimposed channels is:
[0027] amp+=data stack [k]*data stack [k]
[0028] In the formula, data stack [k] represents the data point on the superimposed trace;
[0029] If the iterth iteration satisfies the first convergence condition, the iteration is stopped and the time shift corresponding to the maximum multi-layer correlation coefficient of the iteration is taken as the optimal residual static correction amount.
[0030] Furthermore, a second convergence condition is set: a maximum number of iterations is set. If the number of iterations reaches the maximum number of iterations and still cannot meet the first convergence condition, the iteration is stopped, and a maximum value is selected from the maximum multi-layer correlation coefficients of each iteration, and the time shift corresponding to the maximum value is used as the optimal residual static correction amount.
[0031] The present invention also provides a computer-readable storage medium, which stores at least one computer-executable program. When the at least one program is executed by the computer, the computer executes the steps in the static correction method based on multi-layer bit constraints described in the present invention.
[0032] Compared with the prior art, the beneficial effects of the present invention include:
[0033] The present invention utilizes the information of multiple layers and the correlation between channels. The constraints of multiple layers (phase axes) enable the calculated residual static correction to fully consider the information of multiple layers and better reflect the underground structure, thereby overcoming the defect that the static correction of all seismic data channels is calculated based on the information in only one layer, which will make the local information of the layer have too great an impact on the entire data, and ultimately plays a role in improving the accuracy of the static correction results. BRIEF DESCRIPTION OF THE DRAWINGS
[0034] Figure 1 A schematic diagram of intercepting a time window on a CMP gather;
[0035] Figure 2 This is the structural principle diagram of the superimposed channel;
[0036] Figure 3 Schematic diagram for determining the residual static correction amount based on the maximum multi-layer potential correlation coefficient;
[0037] Figure 4 Schematic diagram of extracting CMP gathers from post-stack data. DETAILED DESCRIPTION
[0038] The present invention utilizes the information of multiple layers and the correlation between channels. The constraints of multiple layers (phase axes) enable the calculated static correction to fully consider the information of multiple layers and better reflect the underground structure, thereby overcoming the defect that the static correction of all seismic data channels is calculated based on the information in only one layer, which will make the local information of the layer have too great an impact on the entire data, and ultimately plays a role in improving the accuracy of the static correction results.
[0039] The present invention is further described in detail below in conjunction with the accompanying drawings.
[0040] A static correction method based on multi-layer position constraints comprises the following steps:
[0041] At each layer, the mutual correlation coefficient between each data in the gather and the corresponding standard trace at different time shifts is calculated; the mutual correlation coefficients at the same time shift at each layer are fused to obtain the multi-layer mutual correlation coefficient; the time shift corresponding to the maximum multi-layer mutual correlation coefficient is taken as the residual static correction.
[0042] By manually picking layers with clear event axes, the residual static correction is constrained using information from multiple layers. For each layer, a time window is first extended upward and downward according to the layer, and then the mutual correlation coefficient is calculated in the time window. The mutual correlation coefficients under the simultaneous shifts in each layer are accumulated, and the time shift corresponding to the maximum mutual correlation coefficient is the static correction.
[0043] Example 1
[0044] This embodiment performs static correction on a CMP gather (common center point gather). Each data in the CMP gather corresponds to a shot point number and a detection point number. Different data may have the same shot point number but different detection point numbers; they may have the same detection point number but different shot point numbers; or they may have different shot point numbers and detection point numbers.
[0045] refer to Figure 1 As shown, the black curve represents the layer on the CMP gather (the user picks a clear layer), and the red curves above and below the black curve represent the equal time lengths t for upward and downward extensions, respectively. i The subsequent time window.
[0046] The construction method of the standard trace includes the following steps: intercepting each layer on the trace gather through the time window and picking up the sampling points of each trace data in the time window to construct the trace gather in the window; Figure 2 As shown, a stacked trace is constructed based on the trace gather in the window; and a corresponding standard trace is constructed for each data track in the trace gather in the window based on the stacked trace.
[0047] In this embodiment, the window gather is constructed based on the CMP gather, and the cross-correlation coefficient between each data track in the window gather and the corresponding standard track at different time shifts is calculated, and the cross-correlation coefficients of the same shot point at the same time shift and the cross-correlation coefficients of the same detection point at the same time shift are accumulated to obtain the accumulated cross-correlation coefficients of the same shot point and the accumulated cross-correlation coefficients of the same detection point.
[0048] For each horizon, the calculation formula of the stack trace is as follows:
[0049]
[0050] In the formula, data stack [k] represents the data point on the stacked trace; m represents the number of traces in the window; data j [k] represents the kth sampling point on the jth channel, k = 0, 1, ..., (t2-t1), t2-t1 represents the time window length.
[0051] For each data track, the calculation steps of the standard track include:
[0052] Calculate the difference: datadiff [k] = data stack [k]-data[k]
[0053] Find the maximum error data Max =max{data diff [k], k = 0, 1, ..., (t2-t1)};
[0054] Calculation standard:
[0055] In the formula, data prefer [k] represents the data point on the standard track; data[k] represents the kth sampling point on a data track in the window, k = 0, 1, ..., (t2-t1). diff [k] represents the track error; data stack Indicates the data points on the superposition track; data Max Indicates the maximum track error.
[0056] The calculation formula of the cumulative correlation coefficient of the same shot point is as follows:
[0057] scramp[srcid][i]+=coef[i]
[0058] The calculation formula of the cumulative correlation coefficient of the same detection point is as follows:
[0059] recamp[recid][i]+=coef[i]
[0060] Where i represents the time shift, i∈[-range,range], i is zero when there is no time shift, is negative when the time shift is in the direction of time decrease, and is positive when the time shift is in the direction of time increase; coef[i] represents the mutual correlation coefficient under time shift i; scramp[srcid][i] represents the cumulative sum of the mutual correlation coefficients of the shot point number srcid under time shift i, that is, the cumulative mutual correlation coefficients of the same shot point; recamp[recid][i] represents the cumulative sum of the mutual correlation coefficients of the detection point number srcid under time shift i, that is, the cumulative mutual correlation coefficients of the same detection point.
[0061] The accumulated mutual correlation coefficients of the same shot points on different layers are added together to obtain the multi-layer mutual correlation coefficients of the common shot points; the accumulated mutual correlation coefficients of the same detection points on different layers are added together to obtain the multi-layer mutual correlation coefficients of the common detection points.
[0062] refer to Figure 3 As shown, for a shot point srcid, the multi-layer potential correlation coefficient coef srcid [i] is obtained by adding the cumulative correlation coefficients of the same shot points in different layers, as follows:
[0063] coef srcid [i] = coef srcid1 [i]+coef srcid2 [i]
[0064] The residual static correction a of the shot point srcid satisfies cofe[a]=max(coef srcid [i]), which is the time shift corresponding to the maximum multi-layer bit correlation coefficient.
[0065] For a detection point recid, the multi-layer bit correlation coefficient coef recid [i] is obtained by adding the cumulative correlation coefficients of the same detection points in different layers, as follows:
[0066] coef recid [i] = coef recid1 [i]+coef recid2 [i]
[0067] The residual static correction b of the detection point recid satisfies cofe[b]=max(coef recid [i]), which is the time shift corresponding to the maximum multi-layer bit correlation coefficient.
[0068] After each iteration, find the time shift corresponding to the maximum value of the multi-layer potential correlation coefficient under different time shifts at the same shot point, and the time shift corresponding to the maximum value of the multi-layer potential correlation coefficient under different time shifts at the same detection point. This time shift is the residual static correction of this iteration.
[0069] The iteration cutoff condition can be set to the maximum number of iterations. When the maximum number of iterations is reached, the iteration is stopped, and the maximum value is selected from the maximum multi-layer correlation coefficients of each iteration, and the time shift corresponding to the maximum value is used as the optimal residual static correction value.
[0070] This embodiment further provides a computer-readable storage medium, which stores at least one computer-executable program. When the at least one program is executed by the computer, the computer executes the steps of the static correction method based on multi-layer bit constraints described in this embodiment.
[0071] Example 2
[0072] This embodiment performs static correction on post-stack data. Each trace in the post-stack data is formed by stacking a CMP gather. The user picks a clear layer on the post-stack data and records the layer as lay. j , j = 1, 2, ..., n, where n is the total number of layers. This embodiment generates multiple stack traces, each stack trace corresponds to a CMP gather, and the same operation is performed on each CMP gather.
[0073] The more specific steps of this embodiment are as follows:
[0074] (1) For a layer, the time length t is extended upward and downward in the layer, and the data of the time window is intercepted. The trace data used in the subsequent calculations are all the data of the time window.
[0075] refer to Figure 4 As shown in the figure, after picking the horizon, each data in the post-stack data has a time sampling point where the horizon is located. In the CMP gather corresponding to the data, a time window is obtained by extending upward and downward from the time sampling point. The data in the time window is intercepted, and the subsequent processing is performed on the time window on the CMP gather, that is, on the window gather.
[0076] (2) Circulate the CMP data set and calculate the correlation coefficient between each data trace in the CMP data set and the standard trace at different time shifts. And accumulate the correlation coefficients of other traces with the same shot number or detection point number calculated at the same time shift.
[0077] Each layer on the data set is intercepted through the time window and the sampling points of each data in the time window are picked up to construct the data set in the window; the stacking track is constructed according to the data set in the window; and the corresponding standard track is constructed for each data in the data set in the window according to the stacking track.
[0078] Calculate the cross-correlation coefficient between each data track in the window gather and the corresponding standard track at different time shifts, and accumulate the cross-correlation coefficients of the same shot point at the same time shift, and accumulate the cross-correlation coefficients of the same detection point at the same time shift to obtain the accumulated cross-correlation coefficients of the same shot point and the accumulated cross-correlation coefficients of the same detection point.
[0079] For each horizon, the calculation formula of the stack trace is as follows:
[0080]
[0081] In the formula, data stack [k] represents the data point on the stacked trace; m represents the number of traces in the window; data j [k] represents the kth sampling point on the jth channel, k = 0, 1, ..., (t2-t1), t2-t1 represents the time window length.
[0082] For each data track, the calculation steps of the standard track include:
[0083] Calculate the difference: data diff [k] = data stack [k]-data[k]
[0084] Find the maximum error dataMax =max{data diff [k], k = 0, 1, ..., (t2-t1)};
[0085] Calculation standard:
[0086] In the formula, data prefer [k] represents the data point on the standard track; data[k] represents the kth sampling point on a data track in the window, k = 0, 1, ..., (t2-t1). diff [k] represents the track error; data stack Indicates the data points on the superposition track; data Max Indicates the maximum track error.
[0087] (3) After all CMP gathers are processed, this iteration ends. The correlation coefficients of the shot numbers and receiver numbers at different layers with the same shift are accumulated. The time shift corresponding to the maximum correlation coefficient is the static correction obtained in this iteration. This static correction is applied to the CMP gathers for the next iteration.
[0088] The calculation formula of the cumulative correlation coefficient of the same shot point is as follows:
[0089] scramp[srcid][i]+=coef[i]
[0090] The calculation formula of the cumulative correlation coefficient of the same detection point is as follows:
[0091] recamp[recid][i]+=coef[i]
[0092] Where i represents the time shift, i∈[-range,range], i is zero when there is no time shift, is negative when the time shift is in the direction of time decrease, and is positive when the time shift is in the direction of time increase; coef[i] represents the mutual correlation coefficient under time shift i; scramp[srcid][i] represents the cumulative sum of the mutual correlation coefficients of the shot point number srcid under time shift i, that is, the cumulative mutual correlation coefficients of the same shot point; recamp[recid][i] represents the cumulative sum of the mutual correlation coefficients of the detection point number srcid under time shift i, that is, the cumulative mutual correlation coefficients of the same detection point.
[0093] (4) If the iteration reaches the convergence condition, the iteration is completed, and the static correction value for each shot point and receiver point obtained in the last iteration is the static correction value finally calculated.
[0094] The sampling iterative algorithm is used to solve the optimal residual static correction, and the first convergence condition is set as:
[0095] amp[iter]-amp[iter-1] <s,
[0096] Where s represents the threshold; amp[iter] and amp[iter-1] represent the accumulated energy of all superimposed traces at the iter-th iteration and iter-1-th iteration respectively;
[0097] The calculation formula of the cumulative energy of all superimposed channels is:
[0098] amp+=data stack [k]*data stack [k]
[0099] In the formula, data stack [k] represents the data point on the superimposed trace;
[0100] If the iterth iteration satisfies the first convergence condition, the iteration is stopped and the time shift corresponding to the maximum multi-layer correlation coefficient of the iteration is taken as the optimal residual static correction amount.
[0101] Set the second convergence condition: set the maximum number of iterations. If the number of iterations reaches the maximum number of iterations and still cannot meet the first convergence condition, stop the iteration and select the maximum value from the maximum multi-layer correlation coefficients of each iteration. The time shift corresponding to the maximum value is used as the optimal residual static correction amount.
[0102] This embodiment further provides a computer-readable storage medium, which stores at least one computer-executable program. When the at least one program is executed by the computer, the computer executes the steps of the static correction method based on multi-layer bit constraints described in this embodiment.
[0103] The above technical scheme is only a specific implementation method of the present invention. For those skilled in the art, it is easy to make various types of improvements or modifications based on the principles disclosed in the present invention, and it is not limited to the technical scheme described in the above specific embodiments of the present invention. Therefore, the above description is only preferred and does not have a restrictive meaning.
Claims
1. A static correction method based on multi-layer position constraints, characterized in that: The following steps are involved: At each layer, the mutual correlation coefficient between each data in the gather and the corresponding standard trace at different time shifts is calculated; the mutual correlation coefficients at the same time shift at each layer are fused to obtain the multi-layer mutual correlation coefficient; the time shift corresponding to the maximum multi-layer mutual correlation coefficient is taken as the residual static correction.
2. The static correction method based on multi-layer position constraints according to claim 1 is characterized in that: The construction method of the standard track includes the following steps: intercepting each layer on the track set through a time window and picking up the sampling points of each data track in the time window to construct the track set in the window; constructing a superimposed track according to the track set in the window; and constructing a corresponding standard track for each data track in the track set in the window according to the superimposed track.
3. The static correction method based on multi-layer position constraints according to claim 1 is characterized in that: For each horizon, the calculation formula of the stack trace is as follows: In the formula, data stack [k] represents the data point on the stacked trace; m represents the number of traces in the window; data j [k] represents the kth sampling point on the jth channel, k = 0, 1, ..., (t2-t1), t2-t1 represents the time window length.
4. The static correction method based on multi-layer position constraints according to claim 1 is characterized in that: For each data track, the calculation steps of the standard track include: Calculate the difference: data diff [k] = data stack [k]-data[k] Find the maximum error data Max =max{data diff [k], k = 0, 1, ..., (t2-t1)}; Calculation standard: In the formula, data prefer [k] represents the data point on the standard track; data[k] represents the kth sampling point on a data track in the window, k = 0, 1, ..., (t2-t1). diff [k] represents the track error; data stack Indicates the data points on the superposition track; data Max Indicates the maximum track error.
5. The static correction method based on multi-layer position constraints according to claim 1 is characterized in that: The in-window gather is constructed according to the CMP gather, and the mutual correlation coefficient between each data track in the in-window gather and the corresponding standard track at different time shifts is calculated, and the mutual correlation coefficients of the same shot point at the same time shift and the mutual correlation coefficients of the same detection point at the same time shift are accumulated to obtain the accumulated mutual correlation coefficients of the same shot point and the accumulated mutual correlation coefficients of the same detection point.
6. The static correction method based on multi-layer position constraints according to claim 1 is characterized in that: The calculation formula of the cumulative correlation coefficient of the same shot point is as follows: scramp[srcid][i]+=coef[i] The calculation formula of the cumulative correlation coefficient of the same detection point is as follows: recamp[recid][i]+=coef[i] Where i represents the time shift, i∈[-range,range], i is zero when there is no time shift, is negative when the time shift is in the direction of time decrease, and is positive when the time shift is in the direction of time increase; coef[i] represents the mutual correlation coefficient under time shift i; scramp[srcid][i] represents the cumulative sum of the mutual correlation coefficients of the shot point number srcid under time shift i, that is, the cumulative mutual correlation coefficients of the same shot point; recamp[recid][i] represents the cumulative sum of the mutual correlation coefficients of the detection point number srcid under time shift i, that is, the cumulative mutual correlation coefficients of the same detection point.
7. The static correction method based on multi-layer position constraints according to claim 6 is characterized in that: The cumulative mutual correlation coefficients of the same shot points on different layers are added together to obtain the multi-layer mutual correlation coefficients of the common shot points: the cumulative mutual correlation coefficients of the same detection points on different layers are added together to obtain the multi-layer mutual correlation coefficients of the common detection points.
8. The static correction method based on multi-layer position constraints according to claim 3 is characterized by: It is used to perform static correction on post-stack data. Each data in the post-stack data is formed by stacking a CMP gather. The sampling iterative algorithm solves the optimal residual static correction amount. The first convergence condition is set as: amp[iter]-amp[iter-1] <s, Where s represents the threshold; amp[iter] and amp[iter-1] represent the accumulated energy of all superimposed traces at the iter-th iteration and iter-1-th iteration, respectively; The calculation formula of the cumulative energy of all superimposed channels is: amp+=data stack [k]*data stack [k] In the formula, data stack [k] represents the data point on the superimposed trace; If the iterth iteration satisfies the first convergence condition, the iteration is stopped and the time shift corresponding to the maximum multi-layer correlation coefficient of the iteration is taken as the optimal residual static correction amount.
9. The static correction method based on multi-layer position constraints according to claim 8 is characterized in that: Set the second convergence condition: set the maximum number of iterations. If the number of iterations reaches the maximum number of iterations and still cannot meet the first convergence condition, stop the iteration and select the maximum value from the maximum multi-layer correlation coefficients of each iteration. The time shift corresponding to the maximum value is used as the optimal residual static correction amount.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one computer-executable program, and when the at least one program is executed by the computer, the computer executes the steps of the static correction method based on multi-layer bit constraints according to any one of claims 1 to 9.