An aftershock background noise random mixing superposition processing method based on delay cutting

Abstract

The application discloses a kind of aftershock background noise random mixed superposition processing methods based on time delay cutting, it includes the following steps: step 1, according to the time delay of a certain time period, the aftershock noise data of long recording test single shot is intercepted;Step 2, the aftershock noise data of different long recording test single shot is randomly combined according to the corresponding relationship of arrangement, and more aftershock noise single shot record is obtained;Step 3, the aftershock noise single shot record of different time period obtained is mixed and superimposed according to the actual production gun record length, sampling interval and arrangement corresponding relationship to normal production single shot data grouping;Step 4, by qualitative and quantitative comparative analysis, the appropriate firing time interval is optimized.The application provides reasonable reference for the determination of well gun firing time interval under different surface conditions in desert area, and has good popularization and application value and practical significance in other similar work area.

Description

Technical Field

[0001] This invention belongs to the field of seismic exploration technology, specifically relating to a method for random mixing and superposition of aftershock noise based on time delay cutting. Background Technology

[0002] With the increasing demand for energy driven by global economic and social development, oil and gas exploration and development are becoming increasingly sophisticated and challenging. For many years, exploration in the Tarim Basin has been plagued by desert aftershocks. Analysis of surface modeling, forward simulation, and experimental data has revealed that desert aftershocks are related to the relatively thick and loose desert surface, but effective suppression methods have yet to be found. Analysis shows that desert aftershocks are mainly composed of slow surface waves generated after the previous shot, dune rumbling that repeatedly oscillates on the surface, and interference waves formed by this generation. Their presence significantly impacts the quality of seismic data. The generation mechanism and mitigation methods of desert aftershocks have lacked systematic research, and their understanding and impact remain relatively vague. To ensure seismic data quality, long firing intervals are typically used for acquisition. However, the current inefficient long firing intervals lack theoretical and quantitative experimental support, and this measure severely affects the efficiency of seismic acquisition operations in desert areas. Since 2010, time intervals have been set for the firing of well guns in the Tarim Desert. The time intervals vary in different regions. According to statistics, the intervals are generally set at 30-35 seconds in the Tarim Basin edge Tahe area and at 45-55 seconds in the Shunbei and Yubei areas in the heart of the desert.

[0003] With the continuous increase in the number of acquisition channels and shots, high-density acquisition places increasingly higher demands on seismic acquisition efficiency, making the reduction of the excitation time interval a crucial challenge. With the improvement of coverage times and processing technology, there is still room for further reduction in the excitation time interval of well shots in desert areas. To further shorten the excitation time interval in desert areas, a scientific and reasonable method for determining the excitation time interval needs to be developed. In the Tarim Basin, the excitation time interval has traditionally been determined using the aftershock-background noise comparison method and the aftershock-target layer energy ratio method, but the results have been less than ideal.

[0004] Aftershock vs. Background Noise Comparison Method: In the past, long-record (56s) single shots were analyzed by selecting different time windows for root mean square amplitude energy analysis. When the aftershock energy of the selected time window was comparable to the background energy of the time window before the first arrival, it was determined as the excitation time interval. The advantage of this method is that the aftershock energy is basically dissipated to the level of environmental noise. However, in the desert area, the excitation time interval determined by this method is often too long (more than 50s), which does not meet the requirements of efficient acquisition.

[0005] Aftershock to Target Layer Energy Ratio Method: In the desert area of ​​the Tarim Basin, root mean square amplitude energy analysis is performed on single shots with long records (56s) at different time windows. When the ratio of the aftershock energy in the selected time window to the energy in the main target layer time window reaches a certain quantitative value (determined according to the target layer exploration needs of different work areas), the excitation time interval is determined. The advantage of this method is that it takes into account the exploration needs of different work areas. The depth of the target layer is different, and the energy difference is also very large, so it can be flexibly determined. The disadvantage is that it is difficult to determine a scientific and reasonable value for the ratio of aftershock energy to target layer energy. Different ratios will result in large differences in the results. Summary of the Invention

[0006] To address the problems existing in the background technology, the present invention provides a method for random mixing and superposition of aftershock noise based on time delay cutting. This method mainly utilizes long-record test single-shot data, extracts different aftershock noise recording segments, and superimposes them onto normal production single-shot data in various combinations to obtain aftershock single-shot records with approximately different excitation time intervals.

[0007] The technical solution adopted in this invention is: a method for random mixing and superposition of aftershock background noise based on time delay cutting, which includes the following steps:

[0008] Step 1: Use the time delay method to extract the aftershock noise data of the long-recorded single shot test from the aftershock noise data of the required time period according to a certain time delay.

[0009] Step 2: Using a random combination method, the aftershock noise data extracted from single shots of different record lengths are randomly combined according to the corresponding relationship of the arrangement to obtain more single-shot records of aftershock noise.

[0010] Step 3: Using the hybrid superposition method, the aftershock noise records of individual shots obtained at different time periods are grouped according to the recording length, sampling interval and arrangement correspondence of the actual production shots, and the aftershock noise is mixed and superimposed.

[0011] Step 4: Perform migration processing on the aftershock noise single-shot data with different excitation time intervals to obtain the aftershock influence migration profile with different excitation time intervals. Through qualitative and quantitative comparative analysis, select the appropriate excitation time interval.

[0012] Furthermore, the specific processing procedure of step 1 is as follows:

[0013] (1) First, the aftershock record is divided into 4 noise segments in 12s segments, representing the aftershock noise generated by the previous shot at the excitation time intervals of 8s, 20s, 32s, and 44s respectively.

[0014] (2) The aftershock noise of each segment was recorded using the time delay method, and it was repeatedly truncated at a time delay interval of 2ms or 4ms.

[0015] (3) According to the time delay interval of 2ms or 4ms, after removing the normal recording duration of 8s, each 12s aftershock noise segment is cut into 999 different aftershock noise segments, that is, each 12s aftershock noise segment of a long recording test single shot is cut into 999 aftershock noise segments.

[0016] Furthermore, the specific processing procedure for step 2 is as follows:

[0017] (1) In the three-dimensional seismic acquisition, each shot is received by multiple arrangements. According to the number of arrangements of the production shot, the long record test shot is also received by the same arrangement. Step 1 cuts out the aftershock noise of different time periods. The aftershock noise of each shot in the same time period corresponds to multiple arrangements. Therefore, the aftershock noise of different arrangements of different shots is randomly combined to finally obtain new aftershock noise shots. Without changing the arrangement order, the total number of new aftershock noise shots S = M × N is obtained, where M represents the total number of arrangements and N represents the number of test shots.

[0018] (2) A random combination method is used to obtain a large number of aftershock noise records from a small number of test single-shot data, which meets the requirement of random superposition of aftershock noise for production single-shot, and ensures that the aftershock noise of each shot is not repeated.

[0019] Furthermore, the processing procedure in step 3 is as follows: a portion of the aftershock noise cannon records obtained in step 2 are randomly selected and mixed and superimposed with the production single cannons with the same record length, sampling interval, and number of arrangements to generate aftershock noise cannons corresponding to different excitation time intervals.

[0020] Furthermore, the specific processing procedure for step 4 is as follows:

[0021] (1) According to steps 1 to 3, obtain the aftershock noise single shot data of four different excitation time intervals, and then use the same static correction, denoising, amplitude compensation, and deconvolution processing flow to perform offset processing on them to obtain the offset profiles of four different aftershock noises.

[0022] (2) Qualitative and quantitative comparative analysis of the above four different migration profiles was conducted. Through comparative analysis, the migration profile that meets the geological requirements was selected, and a reasonable excitation time interval was finally determined.

[0023] The beneficial effects of this invention are as follows: This method can generate tens of thousands of aftershock records by superimposing a small number of experimental single-shot records. Different aftershock noises are extracted and superimposed onto normal production records based on the desired excitation time interval. The obtained aftershock records from different time intervals are then processed using the same denoising, migration, and superposition procedures to ultimately obtain migration profiles with approximate excitation time intervals. By comparing and analyzing the differences in signal-to-noise ratio, resolution, spectral characteristics, and imaging effects of the main target layer in migration profiles with different excitation time intervals, a suitable migration profile can be selected based on geological requirements, thereby determining the optimal excitation time interval.

[0024] In summary, this patent solves the following technical problems:

[0025] 1. By using a small amount of experimental single-shot data and the aftershock noise combination superposition method, a large amount of single-shot data with approximate aftershock effects at different excitation time intervals can be obtained, thus avoiding the waste of exploration costs caused by conducting a large number of aftershock single-shot tests in actual production.

[0026] 2. Different excitation time intervals correspond to aftershock noise in different time periods of long records. According to the analysis requirements, the time delay method can be used to extract a variety of different aftershock noises. Each arrangement of aftershock noise corresponds one-to-one with the arrangement of single-shot records in normal production.

[0027] 3. The aftershock noise from different arrangements of the test shots can be obtained by random combination method to obtain more aftershock shot records. These desert aftershock shot records with different excitation time intervals can be processed through the same process to obtain multiple excitation interval migration profiles, which are convenient for comparative analysis. Attached Figure Description

[0028] Figure 1 This is a flowchart of the present invention.

[0029] Figure 2 This is a schematic diagram of a 56-second long recording and preliminary cutting; where (a) is the original single shot of the 56-second long recording, and (b) is the aftershock data after 8 seconds cut every 12 seconds.

[0030] Figure 3 Each 12-second aftershock noise segment is cut into 8-second noise segments with a 4ms delay.

[0031] Figure 4 This is a schematic diagram showing the overlay of aftershock background and production blast records.

[0032] Figure 5 It is a normal production offset profile and an aftershock noise superposition offset profile, where (a) is the normal production offset profile and (b) is the aftershock noise superposition offset profile.

[0033] Figure 6Spectral analysis diagram of offset profiles during normal production and offset profiles with mixed aftershock noise. Detailed Implementation

[0034] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0035] A method for combining and stacking single-shot aftershock records in desert earthquakes, the steps of which are as follows:

[0036] Step 1: Using the time delay method, extract the aftershock noise data of the long-recorded single shot test according to a certain time delay to extract the aftershock noise data of the required time period.

[0037] (1) Taking the 56s long record test single shot data in the desert work area of ​​Tarim Basin as an example, the normal production single shot record length is 8s, and the records after 8s are all aftershocks. First, the aftershock records from 8 to 56s are divided into 4 noise segments in 12s segments, namely 8-20s, 20-32s, 32-44s, and 44-56s aftershock noise, which represent the aftershock noise generated by the previous shot in the 8s, 20s, 32s, and 44s excitation time intervals, respectively.

[0038] (2) In order to obtain a large amount of different aftershock noise with a small amount of long-record test single shot data, the time delay method was used to record the 12s aftershock noise of each segment. The noise was truncated multiple times according to a certain time delay interval (4ms). The specific time delay interval was determined according to the actual sampling interval. In this long-record test single shot, 4ms sampling was used. If 2ms sampling is used, a 2ms time delay interval can also be used.

[0039] (3) Based on the 4ms time delay interval, after removing the 8s normal recording duration, each 12s aftershock noise segment can be divided into 999 different aftershock noise segments ((12-8)×1000 / 4-1=999), that is, each 12s aftershock noise segment of a long recording test single shot can be divided into 999 aftershock noise segments.

[0040] Step 2: Using a random combination method, the aftershock noise data extracted from single shots of different record lengths are randomly combined according to the corresponding arrangement to obtain more single-shot records of aftershock noise.

[0041] (1) In three-dimensional seismic acquisition, each shot is received by multiple arrangements. According to the number of arrangements of the production shot, the long record test shot is also received by the same arrangement. The aftershock noise of different time periods was cut out in the previous step. The aftershock noise of each shot in the same time period corresponds to multiple arrangements. Therefore, the aftershock noise of different arrangements of different shots can be randomly combined to obtain new aftershock noise shots. Without changing the arrangement order, the total number of new aftershock noise shots S = M × N (M - total number of arrangements, N - number of test shots).

[0042] (2) According to the formula above, taking a 36-line receiver arrangement as an example, with 10 test shots, a total of 360 aftershock noise records can be obtained. Multiplying these by the 999 aftershock noise segments cut using the time delay method, we can obtain 359,640 aftershock noise records. Therefore, the random combination method can obtain a large number of aftershock noise records with a small amount of test single-shot data, fully meeting the requirement of random superposition of aftershock noise for single-shot production, and ensuring that the aftershock noise of each shot is non-repetitive.

[0043] Step 3: Using the hybrid superposition method, the aftershock noise records of individual shots obtained at different time periods are grouped according to the recording length, sampling interval and arrangement correspondence of the actual production shots, and the aftershock noise is mixed and superimposed.

[0044] (1) A portion of the aftershock noise shot records obtained in the previous step are randomly selected and mixed and superimposed with the production single shot records with the same record length, sampling interval, and number of permutations to generate aftershock noise shots corresponding to different excitation time intervals. After the mixing and superimposition process, the difference in background energy before the first arrival of the aftershock noise shots corresponding to different excitation time intervals is displayed.

[0045] (2) Taking the 12,000 test beams actually produced in the desert work area of ​​the Tarim Basin as an example, according to the above method, 12,000 aftershock noise records were randomly selected from 359,640 aftershock noise records and mixed and superimposed with the 12,000 records of normal production to obtain the aftershock noise records corresponding to a certain excitation time interval. The production records used 2ms sampling and the test records used 4ms sampling. Therefore, it is necessary to resample the production records by 4ms and then mix and superimpose them to finally obtain 12,000 aftershock noise records.

[0046] Step 4: Using the same denoising and deconvolution processing procedures, the aftershock noise single-shot data with different excitation time intervals are migrated to obtain aftershock influence migration profiles with different excitation time intervals. Through qualitative and quantitative comparative analysis, a suitable excitation time interval is selected.

[0047] (1) Following the above three steps, aftershock noise single shot data with four different excitation time intervals (8s, 20s, 32s, 44s) can be obtained. Then, the same static correction, denoising, amplitude compensation, deconvolution and other processing procedures are used to perform migration processing on them to obtain migration profiles of four different aftershock noises.

[0048] (2) Qualitative and quantitative comparative analysis was conducted on the above four different migration profiles. The main comparative analysis included the signal-to-noise ratio, spectrum and imaging effect of the main target layer, etc. Through comparative analysis, the migration profile that meets the geological requirements was selected, and the reasonable excitation time interval was finally determined.

[0049] The method of this invention was first applied in the Shunbei work area of ​​the Tarim Basin. This work area is a desert surface with tall sand dunes, with a maximum elevation difference of over 100 meters. Well-fired firing was used for excitation, and the previous excitation interval was over 45 seconds. A 3D receiver array of 36 channels was used, with a total of 19,008 channels, a recording length of 8 seconds, and a sampling interval of 2 milliseconds. To further shorten the well-fired firing interval and provide a reasonable basis for selecting the excitation interval, an 8-shot, 50-second long recording test was conducted in the work area. The 8 shots were distributed in different parts of the sand dunes, and each shot used a 36-channel receiver array. Using the method of this invention, the aftershock noise of the 8 test shots was divided into 3 segments (8-20s, 20-32s, and 32-44s). Using a 4 millisecond delay method, 999 segments of aftershock noise were obtained for each shot per segment. Then, by combining 8 shots in 36 arrangements, 287,712 aftershock noise records can be obtained. 10,000 aftershock noise records are randomly selected from these and mixed and superimposed with 10,000 single-shot data from production shots, resulting in 12,000 shots of mixed aftershock noise data from each of the three different time periods. These three types of mixed aftershock noise single-shot data and normal acquisition single-shot data (45-second excitation interval) are then migrated and superimposed using the same processing procedure to obtain four different migration profiles. Qualitative and quantitative analyses are performed on the four profiles, and the migration profile that meets the geological requirements of the area is selected, along with a reasonable excitation interval.

[0050] Through qualitative and quantitative analysis of different aftershock noise aliasing migration profiles and normal production migration profiles, the 32-44s aftershock noise aliasing migration profile has the closest spectrum and imaging effect to the normal production profile. Therefore, we believe that the well shot excitation interval in this area can be further shortened to about 32s. Considering that the target layer in this area is ultra-deep Ordovician secondary fracture imaging, the excitation interval was finally determined to be 40s for seismic acquisition in actual production.

[0051] In summary, this invention addresses the challenge of aftershock impact in the Tarim Basin desert region by utilizing a small number of long-record test single shots and employing a random combination and stacking method to cut and combine a large amount of aftershock single-shot data. This facilitates the comparative analysis of migration profiles showing the impact of aftershocks under various excitation time intervals, providing reasonable reference for determining well shot excitation time intervals under different surface conditions in the Tarim Basin desert region. It also has good application value and practical significance for future similar work areas.

[0052] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the above embodiments do not limit the scope of protection of the present invention in any way, and all technical solutions obtained by equivalent substitution or other means fall within the scope of protection of the present invention. Parts not covered in this invention are the same as or can be implemented using existing technology.

Claims

1. A method for random mixing and superposition processing of aftershock background noise based on time delay cutting, characterized in that... Includes the following steps: Step 1: Using the time delay method, extract the aftershock noise data for the required time period from the long-recorded single-shot test data. The specific processing procedure is as follows: (1) First, the aftershock record is divided into 4 segments of noise, each segment is 12s, representing the aftershock noise generated by the previous shot at the excitation time intervals of 8s, 20s, 32s, and 44s respectively. (2) The aftershock noise of each segment for 12 seconds was recorded using the time delay method, and it was truncated multiple times at a time delay interval of 2ms or 4ms. (3) According to the time delay interval of 2ms or 4ms, after removing the normal recording duration of 8s, each 12s aftershock noise segment is cut into 999 different aftershock noise segments, that is, each 12s aftershock noise segment of a long recording test single shot is cut into 999 aftershock noise segments. Step 2: Using a random combination method, the aftershock noise data extracted from single shots of different record lengths are randomly combined according to the corresponding arrangement to obtain more single-shot records of aftershock noise; the specific processing procedure is as follows: (1) In the three-dimensional seismic acquisition, each shot is received by multiple arrangements. According to the number of arrangements of the production shot, the long record test shot is also received by the same arrangement. Step 1 cuts out the aftershock noise of different time periods. The aftershock noise of each shot in the same time period corresponds to multiple arrangements. Therefore, the aftershock noise of different arrangements of different shots is randomly combined to obtain a new aftershock noise shot. Without changing the arrangement order, the total number of new aftershock noise shots is S=M×N, where M represents the total number of arrangements and N represents the number of test shots. (2) A random combination method is used to obtain a large number of aftershock noise records with a small amount of test single shot data, which meets the requirement of random superposition of aftershock noise for production single shot and ensures that the aftershock noise of each shot is not repeated. Step 3: Using the hybrid superposition method, the aftershock noise records of individual shots obtained at different time periods are grouped according to the recording length, sampling interval and arrangement correspondence of the actual production shots, and the aftershock noise is mixed and superimposed. Step 4: Apply the same processing procedure to the aftershock noise single-shot data with different excitation time intervals to obtain aftershock influence migration profiles with different excitation time intervals. Through qualitative and quantitative comparative analysis, select the appropriate excitation time interval.

2. The method for random mixing and superposition of aftershock background noise based on time delay cutting according to claim 1, characterized in that, The processing procedure in step 3 is as follows: a large number of aftershock noise gun records obtained in step 2 are randomly selected, and a portion of the aftershock noise gun records are mixed and superimposed with the production single gun with the same record length, sampling interval and number of arrangements to generate aftershock noise guns corresponding to different excitation time intervals.

3. The method for random mixing and superposition processing of aftershock background noise based on time delay cutting according to claim 1, characterized in that, The specific processing procedure for step 4 is as follows: (1) According to steps 1 to 3, obtain the aftershock noise single shot data of four different excitation time intervals, and then use static correction, noise reduction, amplitude compensation and deconvolution processing to perform offset processing on them to obtain the offset profiles of four different aftershock noises. (2) Qualitative and quantitative comparative analysis of the above four different migration profiles was conducted. Through comparative analysis, the migration profile that meets the geological requirements was selected, and the reasonable excitation time interval was finally determined.

Images