Method for quickly determining seismic horizon by using depth domain migration speed

Abstract

The invention discloses a method for quickly determining a seismic horizon by using a depth domain migration speed. According to the method, macroscopic seismic horizon information is rapidly extracted by analyzing the structural feature information of the depth domain seismic migration velocity, reliable seismic horizon basic data is provided for constructing a complex area, horizon recognition efficiency is remarkably improved, and the workload of interpretation personnel is greatly reduced. The method is characterized in that the horizon picking process has adaptive updating capability, macroscopic seismic horizon information of a work area can be rapidly and stably extracted, the depth change trend of an underground structure is accurately reflected, and the continuity of the seismic horizon in the transverse direction is ensured.

Description

Technical Field

[0001] The invention belongs to the field of oil and gas seismic exploration, and in particular relates to a method for rapidly determining seismic horizons by utilizing depth domain offset velocity. Background Art

[0002] As a key component and basic step of seismic interpretation, seismic horizon interpretation aims to identify and track stratigraphic interfaces for interpretation of stratigraphic structure and attributes, as well as to construct underground three-dimensional geological models. This process usually requires interpreters to manually pick up specific seismic reflection events from seismic data to track key stratigraphic information and form continuous stratigraphic data, which is a cumbersome and huge workload. In structurally complex areas, such as faults, folds, and salt domes, the difficulty of interpretation increases, requiring interpreters to have professional geological knowledge and rich practical experience to identify and distinguish complex seismic responses, making stratigraphic tracking results greatly affected by human subjective factors, thereby affecting the accuracy and reliability of subsequent reservoir interpretation.

[0003] Oil and gas seismic exploration and development has put forward higher and higher requirements for the accuracy and efficiency of seismic horizon interpretation. Rapidly obtaining macroscopic seismic horizon data of the work area to accurately reflect the spatial distribution and contact relationship of underground structures and strata is crucial for the subsequent description and accurate characterization of different scales (especially microscale) characteristics of strata and structures, and is the basis for reliably realizing the detailed description and development of oil and gas reservoirs. Therefore, the present invention is to quickly and effectively extract the macroscopic seismic horizon information of the work area on the basis of retaining the accuracy of the underground structural framework, so as to significantly improve the efficiency of subsequent detailed interpretation of the horizon. Summary of the invention

[0004] The present invention proposes a method for quickly determining seismic horizons using depth domain migration velocity. The method uses structural feature information in depth domain migration velocity data to quickly extract macroscopic seismic horizon information, provide reliable seismic horizon basic data for structurally complex areas, significantly improve the efficiency of horizon identification, and greatly reduce the workload of interpreters. The method of the present invention includes the following main steps:

[0005] (1) After preprocessing the input initial depth domain migration velocity data, the processed three-dimensional depth domain migration velocity data V is obtained. p (i,j,k), where i represents the depth index value, j represents the xline index value, and k represents the inline index value. The depth range is H = [H(1),…,H(N)], and N is the number of depth sampling points;

[0006] (2) Estimated depth value of a given target layer and estimated speed Set the depth position weight w h and the layer velocity weight wV , so that the sum of the two is 1;

[0007] (3) The processed 3D depth domain migration velocity data V p (i, j, k) extracts its maximum value V(m, j, k) and the depth position h(m, j, k) corresponding to the maximum value one by one, where m represents the maximum value index value obtained one by one;

[0008] (4) Based on the results of the previous step, the layer depth h1(j,k) of the j-th xline and the k-th inline is calculated using the following formula:

[0009]

[0010] (5) For the layer depth h1(j,k), the following steps are used for iteration:

[0011] ① Update the depth position weight w h and the layer velocity weight w V ;

[0012] ② For the layer depth h1(j,k) in the previous step, remove the discontinuity points, abnormal points and jump points in the data, mark their depth positions as 0, obtain the effective layer depth h1′, and calculate the average layer velocity at the effective depth position

[0013] ③ For the jth xline and kth inline with a marked layer depth position of 0, the layer depth of the adjacent trace is obtained by the forward backtracking method to obtain the initial depth h0(j,k), and the formula is as follows:

[0014]

[0015] Among them, m1 and m2 are the forward traceback distances of xline and inline respectively;

[0016] ④ Based on the results of the previous step, use the depth domain to migrate the velocity V p , calculate the layer depth data h2(j,k) of the jth xline and the kth inline, the formula is as follows:

[0017]

[0018] ⑤ Repeat ① to ④ until the extracted layer depth data meets the requirements;

[0019] (6) Perform two-dimensional smoothing on the layer depth data h2(j,k) to obtain the final seismic layer data h(j,k). BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1This is the process of picking up seismic horizons by migrating seismic data in the depth domain in the embodiment of the present invention. The XLine line number range of this work area is 1-391, with a total of 391 lines, and the InLine line number range is 1-661, with a total of 661 lines. Figure 1 (a) is a top view of the seismic horizon picked in the first iteration. Figure 1 (b) is a top view of the seismic horizon picked in the second iteration. Figure 1 (c) is a top view of the seismic horizon picked in the third iteration. Figure 1 (d) is a top view of the final seismic horizon after smoothing. Figure 1 The ordinate of (a)-(d) is the InLine channel number, and the abscissa is the XLine channel number. Figure 1 (a)-(d) The colors of the color maps indicate the depth of the picked layers, sharing a color scale ranging from 3.5km to 5.5km, in kilometers.

[0021] Figure 2 This is the process of picking up seismic horizons through the depth domain migration velocity of the 461st InLine profile of the embodiment of the present invention. The XLine line number range of this work area is 1-391, with a total of 391 lines, a depth range of 0km-6km, a depth sampling interval of 5m, and a number of sampling points of 1201. Figure 2 (a) is the cross-section of the seismic layers picked in the first iteration. Figure 2 (b) is the cross-section of the seismic layers picked in the second iteration. Figure 2 (c) is the cross-section of the seismic layers picked in the third iteration. Figure 2 (d) is the cross-section of the smoothed seismic horizon. Figure 2 The ordinate of (a)-(d) is the depth, ranging from 0km to 6km, in kilometers, and the abscissa is the XLine number. Figure 2 The black lines in (a)-(d) represent the corresponding extracted layers. Figure 2 (a)-(d) The colors of the color maps represent the migration velocity of the 461st InLine profile, sharing a color scale ranging from 2.5km / s to 4.0km / s, in kilometers per second.

[0022] Figure 3This is a display diagram of the seismic layer picked up by the depth domain offset seismic data of the 461st InLine section of the embodiment of the present invention on the seismic section. The XLine line number range of this work area is 1-391, with a total of 391 lines, the display depth range is 3.8km-5.6km, the depth sampling spacing is 5m, and the number of sampling points is 361. The color background of the figure is the depth of the depth domain offset, and its velocity range is 2.5km / s-4.0km / s, in kilometers per second. The seismic records in this section are superimposed on the offset velocity map in a black waveform variable area display mode, and the white curve in the figure shows the final seismic layer after smoothing. DETAILED DESCRIPTION

[0023] Figure 1-3 A method for constructing seismic horizons using the present invention is shown, and its corresponding specific implementation method includes the following steps:

[0024] (1) Preprocess the depth domain migration velocity data to remove bad values ​​and obtain the processed three-dimensional depth domain migration velocity data V p (i,j,k), where i represents the depth index value, with a total of 1021 sampling points, j represents the xline index value, with a total of 391 xline sampling points, and k represents the inline index value, with a total of 661 inline sampling points. The depth range is H = [H(1),…,H(N)], and N is the number of depth sampling points, which is 1021.

[0025] (2) Estimated depth value of a given target layer and estimated speed Set the depth position weight w h and the layer velocity weight w V , the sum of the two is 1;

[0026] (3) The maximum value V(m,j,k) and the depth position h(m,j,k) corresponding to the maximum value are extracted for each channel of the 3D depth domain migration velocity, where m represents the maximum value index value obtained for each channel. The method is as follows:

[0027] ① For the depth domain migration velocity of the jth xline and kth inline, calculate the differential data ΔV p (i,j,k), the formula is as follows:

[0028] ΔV p (i,j,k)=V p (i+1,j,k)-V p (i,j,k),

[0029] ② For the jth xline, the velocity difference data ΔV of the kth inline track p(i,j,k), for each sampling point i from 1 to N-1, filter out those that satisfy ΔV p (i,j,k)<0 and ΔV p The sampling point index value i of the point where (i+1,j,k)>0;

[0030] ③ Check the sampling point i in the previous step to determine whether it satisfies V p (i-1,j,k)<V p (i,j,k)<V p (i+1,j,k), if yes, calculate the depth and speed value of its sampling point, the formula is as follows:

[0031]

[0032] (4) Based on the results of the previous step, calculate the layer depth h1(j,k) of the j-th xline and the k-th inline. The formula is as follows:

[0033]

[0034] (5) According to the depth h1(j, k) of the initial layer obtained in the previous step, the following steps are used for iteration:

[0035] ① Update the depth position weight w h and the layer velocity weight w V ;

[0036] ②Preprocess the layer depth result h1(j,k) of the previous step, remove the obvious discontinuous points, abnormal points and jump points, mark the depth of the removed positions as 0, obtain the effective layer depth h1′(j,k), and calculate the average layer velocity at the effective depth position

[0037] ③ When the layer depth of the j-th xline and the k-th inline is 0, the layer depth of the adjacent trace is obtained by backtracking to obtain the initial depth h0(j,k). The formula is as follows:

[0038]

[0039] Among them, m1 and m2 are the distances traced back in the xline and inline directions respectively;

[0040] ④ According to the results of the previous step, use the depth domain migration velocity Vp to calculate the layer depth h2(j,k) of the j-th xline and the k-th inline. The formula is as follows:

[0041]

[0042] ⑤ Repeat ① to ④ until the extracted layer data meets the requirements;

[0043] (6) Perform two-dimensional smoothing on the stratigraphic data h2(j,k) obtained in the previous step to obtain the final seismic stratigraphic data h(j,k).

[0044] Figure 1 This is the process of picking up seismic horizons through depth-domain migration seismic data in the embodiment of the present invention. First, in the depth-domain migration velocity data, the maximum value is identified for the area near the specific depth position, thereby obtaining Figure 1 (a) shows the preliminary results of seismic horizon picking. Figure 1 (a) The significant discontinuity points, outliers and jump points are removed, and the depth of the removed points is replaced by the depth of the adjacent layer point using the forward tracing method to ensure the continuity of the layer in the XLine and InLine directions. The new layer position is optimized using the set depth position weight and layer velocity weight, and then the Figure 1 (b) The optimization result of seismic horizon picking. The method of the present invention gradually increases the depth position weight during the iteration process to enhance the continuity of the seismic horizon in the XLine and InLine directions, and finally obtains Figure 1 (a) and Figure 1 (b) More stable seismic horizon results, such as Figure 1 (c) After three iterations, two-dimensional smoothing technology is used to finally obtain Figure 1 (d) Seismic horizon.

[0045] Figure 2 The process of picking seismic layers on the 461st InLine depth domain migration velocity profile in an embodiment of the present invention is shown. Figure 2 In (a), the black lines represent the seismic horizons picked up in the first iteration. In the initial iteration stage, because the initial depth was set to a fixed value and the horizon velocity weight was high, the horizons jumped in areas with significant structural changes. In the second iteration, in the preprocessing of the initial depth, these abnormal horizons were removed, and the depth of the adjacent horizon position was used as the position depth to exclude the abnormal horizons. A new round of iteration was then performed to generate Figure 2 (b) is a more reasonable seismic horizon. After three iterations, Figure 2 The horizon shown in (c) gradually stabilizes and meets the requirements of the interpretation work. Finally, the iterative horizon results are smoothed to ensure the continuity of the horizon in the XLine and InLine directions. Figure 2 (d) Seismic horizon.

[0046] Figure 3The smoothed seismic horizon on the 461st InLine seismic data profile in the embodiment of the present invention is shown. The white curve in the figure represents the smoothed seismic horizon, and its trend is consistent with the seismic reflection phase axis of the seismic profile, accurately depicting the changing trend of the underground structure. The seismic horizon line can be used as a reference horizon for interpretation, laying the foundation for subsequent seismic fine interpretation work. In addition, it can be seen from the figure that high-quality horizon information is obtained by only three iterations using the method of the present invention, which fully proves the effectiveness and reliability of the method of the present invention.

[0047] The advantages of the method of the present invention are: (1) based on the structural features in the depth domain offset velocity, the macroscopic seismic layer information of the work area is quickly extracted, which can accurately reflect the depth change trend of the underground structure; (2) the method fully considers the position change of the seismic layer in depth, ensures the continuity in the XLine and InLine directions, and realizes the adaptive update of the seismic layer picking process by applying different weight coefficients; (3) in practical applications, the present invention shows high efficiency and extremely high stability.

[0048] The above embodiments are only used to illustrate the present invention, wherein the various implementation steps of the method can be changed, and any equivalent transformations and improvements based on the technical solution of the present invention should not be excluded from the protection scope of the present invention.

Claims

1. A method for quickly determining seismic horizons using depth domain migration velocity, comprising the following main steps: (1) After preprocessing the input initial depth domain migration velocity data, the processed three-dimensional depth domain migration velocity data V is obtained. p (i,j,k), where i represents the depth index value, j represents the xline index value, and k represents the inline index value. The depth range is H = [H(1),…,H(N)], and N is the number of depth sampling points; (2) Estimated depth value of a given target layer and estimated speed Set the depth position weight w h and the layer velocity weight w V , so that the sum of the two is 1; (3) The processed 3D depth domain migration velocity data V p (i, j, k) extracts its maximum value V(m, j, k) and the depth position h(m, j, k) corresponding to the maximum value one by one, where m represents the maximum value index value obtained one by one; (4) Based on the results of the previous step, the layer depth h1(j,k) of the j-th xline and the k-th inline is calculated using the following formula: (5) For the layer depth h1(j,k), the following steps are used for iteration: ① Update the depth position weight w h and the layer velocity weight w V ; ② For the layer depth h1(j,k) of the previous step, remove the discontinuity points, abnormal points and jump points in the data, mark their depth positions as 0, obtain the effective layer depth h1′, and calculate the average layer velocity at the effective depth position ③ For the jth xline and kth inline with a marked layer depth position of 0, the layer depth of the adjacent trace is obtained by the forward backtracking method to obtain the initial depth h0(j,k), and the formula is as follows: in, m1 and m2 are the forward traceback distances of xline and inline respectively; ④ Based on the results of the previous step, use the depth domain to migrate the velocity V p , calculate the layer depth data h2(j,k) of the jth xline and the kth inline, the formula is as follows: ⑤ Repeat ① to ④ until the extracted layer depth data meets the requirements; (6) Perform two-dimensional smoothing on the layer depth data h2(j,k) to obtain the final seismic layer data h(j,k).

Images