Iterative optimization method for improving signal-to-noise ratio of ultra-shallow seismic data of sandstone type uranium mine

Abstract

The invention belongs to the field of sandstone type uranium mine resource geophysical exploration and development, and discloses an iterative optimization method for improving the signal-to-noise ratio of sandstone type uranium mine extremely-shallow layer seismic data, which firmly grasps the problem essence of the extremely-shallow layer seismic data of a sandstone type uranium mine reservoir, and adopts an iterative signal-to-noise ratio improving method to improve the signal-to-noise ratio of the extremely-shallow layer seismic data of the sandstone type uranium mine reservoir. On this basis, data regularization based on Fourier reconstruction is also performed for the problems of low coverage frequency and irregular data caused by the influence of earth surface factors and acquisition factors on the extremely shallow seismic data, and the method has very good theoretical technical progress and is suitable for popularization and application. The problems of low signal-to-noise ratio, low coverage frequency and irregular data of the ultra-shallow seismic data of the sandstone type uranium mine are fully solved.

Description

Technical Field

[0001] The invention belongs to the field of geophysical exploration and development of sandstone-type uranium ore resources, and relates to a targeted processing method for extremely shallow seismic data containing sandstone-type uranium ore, in particular to an iterative optimization method for improving the signal-to-noise ratio of extremely shallow seismic data of sandstone-type uranium ore. Background Art

[0002] Although similar to oil and gas resources, sandstone-type uranium deposits are sedimentary deposits in sandstones that are extremely shallow relative to oil and gas reservoirs (about 500ms in the southern Songliao Basin). The three-dimensional seismic data used for oil and gas exploration are greatly affected by surface factors in this layer, and the number of coverage is seriously insufficient, resulting in a serious lack of signal-to-noise ratio in the vicinity of the target layer in the seismic processing results, which restricts the ability to use three-dimensional seismic exploration data to find sandstone-type uranium deposits. The key to solving this problem lies in whether the noise in the extremely shallow seismic data can be scientifically and effectively suppressed by a relatively fidelity and amplitude preservation method, and the regularity of the seismic data of the extremely shallow target layer can be moderately improved by using the existing effective information. In shallow seismic data, surface waves, strong energy and shallow refraction interference (especially shallow refraction) greatly reduce the signal-to-noise ratio of the reflection signal, which has a serious negative effect on stacking and migration imaging. Therefore, noise should be suppressed or filtered out from the seismic data.

[0003] Data regularization has unique advantages in solving the regularization of shot and inspection point positions and improving the density of CMP facets. From the perspective of processing applications, regularization will not add additional geological information, but will only improve the spatial sampling properties of the observation system to a certain extent to better meet the needs of subsequent processing. Therefore, the denoising technology and data regularization technology in the processing of oil and gas exploration seismic data are specifically designed and optimized in the processing of extremely shallow seismic data in sandstone-type uranium mines. Only the surface waves, wild values ​​and shallow refractions in the extremely shallow seismic data are suppressed, and the processes and parameters used are also highly targeted. Data regularization fully considers the contribution of the near-shot offset to the extremely shallow seismic data, and uses highly targeted parameters to regularize the seismic traces within the 2500-meter shot offset. Summary of the invention

[0004] In order to make up for the shortcomings of the prior art, the present invention proposes an iteratively optimized and highly targeted data processing method for improving the signal-to-noise ratio and data regularization of extremely shallow seismic data for sandstone-type uranium deposits.

[0005] Aiming at the special needs of data characteristics and data processing of extremely shallow layers of sandstone-type uranium deposits, the present invention proposes an iterative optimization method for improving the signal-to-noise ratio of seismic data of extremely shallow layers of sandstone-type uranium deposits, so as to solve the problems of low signal-to-noise ratio, low coverage times and poor data regularity caused by surface factors and acquisition factors in seismic data of extremely shallow layers of sandstone-type uranium deposit reservoirs. Through the method of the present invention, the data signal-to-noise ratio is significantly improved, the coverage times and data regularity are better, which is beneficial to subsequent stacking and offset imaging.

[0006] The purpose of the present invention is achieved through the following technical solutions and steps:

[0007] a. Read in the 3D seismic data with the observation system defined in the early stage and after field static correction, extract the seismic traces with the shot offset less than or equal to 2500 meters, and cut the seismic trace length to 1500ms;

[0008] b. Analyze the distribution area and seismic characteristics of shallow refraction in the data, subdivide the seismic data into regions according to their distribution characteristics, select typical single shots in different regions, and perform denoising on the typical single shots;

[0009] c. Analyze the typical single-shot denoising effect, iteratively optimize and adjust the typical single-shot denoising scheme, set the threshold for manual judgment of denoising effect, and complete the optimization of sample denoising parameters when the expected result is achieved;

[0010] d. De-noise the whole area with a single shot, and select a single shot for effect quality control. Fine-tune the denoising parameters according to the quality control results, and finally achieve the suppression and removal of surface waves, outliers, and shallow refraction in the whole area;

[0011] e. Sorting the denoised seismic data into the CMP domain and performing dynamic correction processing;

[0012] f. Set the data regularization parameters and transform the seismic data into the Fourier domain using the anti-leakage two-dimensional Fourier transform method;

[0013] g. According to the data characteristics and actual data processing needs, the expected output parameters are set, and the data is transformed back to the time-space domain using inverse Fourier transform. After a forward-inverse transform, the seismic data is transformed from the original irregular input data to regular output data;

[0014] h. Perform reaction correction on the regularized seismic data, suppress random noise on the gathers, and finally output the CMP gather data with improved signal-to-noise ratio.

[0015] Compared with the prior art, the present invention has the following beneficial effects:

[0016] The present invention aims at the problems of low signal-to-noise ratio, low coverage times, irregular data, etc. in the extremely shallow seismic data of sandstone-type uranium reservoirs, and proposes a method for improving the signal-to-noise ratio and data regularity by progressive iterative optimization, thereby improving the quality of seismic data in this layer. The present invention firmly grasps the essence of the problems existing in the extremely shallow seismic data of sandstone-type uranium reservoirs, adopts an iterative method for improving the signal-to-noise ratio, and on this basis, aims at the problems of low coverage times and irregular data caused by the extremely shallow seismic data affected by surface factors and acquisition factors, and also performs data regularization based on Fourier reconstruction. The method has great theoretical and technical progress, and fully solves the problems of low signal-to-noise ratio, low coverage times, and irregular data in the extremely shallow seismic data of sandstone-type uranium reservoirs.

[0017] Compared with other existing seismic data processing methods for oil and gas exploration, the method of the present invention has the following advantages:

[0018] ① The method of the present invention sorts and intercepts the original data to a large extent according to the characteristics of sandstone-type uranium deposits being located in extremely shallow layers, and only retains the effective seismic data for sandstone-type uranium deposit prediction, which can not only ensure a stronger pertinence in the seismic data processing link, but also greatly reduce the amount of seismic data processing, thereby greatly reducing the data processing cost and shortening the seismic data processing cycle while meeting the processing target requirements;

[0019] ② Compared with the conventional denoising process, the method of the present invention fully considers the relatively low signal-to-noise ratio of extremely shallow seismic data facing sandstone-type uranium deposits in the iterative optimization denoising process, and adopts a fine and cautious denoising method of zoning parameter selection, quality control fine-tuning, and full-area secondary denoising, thereby ensuring that while removing noise as much as possible, the effective signal is basically not damaged;

[0020] ③ The regularization method for extremely shallow data is different from the regularization of conventional oil and gas exploration seismic data in that the data must be screened before regularization, especially in complex surface areas with serious changes in appearance. On the basis of targeted zoning denoising in the early stage, the anti-leakage two-dimensional Fourier transform method is used to perform forward and reverse transformation to obtain the regularized offset gather data;

[0021] ④ The targeted denoising method and data regularization method adopted by the method of the present invention have better applicability, so that the denoising effect after the partition iterative optimization is significantly better than the denoising effect of conventional seismic data;

[0022] ⑤ After the comparison of stacked sections before and after denoising and before and after regularization, the signal-to-noise ratio and event continuity of extremely shallow seismic data are greatly improved through the targeted processing of the present invention;

[0023] ⑥ The extremely shallow seismic data of sandstone-type uranium mines finally obtained by the present invention provide high signal-to-noise ratio gather data for subsequent resolution improvement processing and migration imaging, which has important practical significance and economic benefits for reducing the exploration risk of sandstone-type uranium mines and improving the development efficiency of sandstone-type uranium mines. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the present invention or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. The drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0025] Figure 1 It is a flow chart for realizing the method of the present invention;

[0026] Figure 2 It is a detailed diagram of the specific implementation process of the method of the present invention;

[0027] Figure 3 It is a seismic single shot profile before iterative optimization and denoising according to an embodiment of the present invention;

[0028] Figure 4 It is a seismic single shot profile after iterative optimization and denoising according to an embodiment of the present invention;

[0029] Figure 5 It is an embodiment of the present invention that iteratively optimizes the stacked profile before denoising;

[0030] Figure 6 It is a superimposed profile after iterative optimization and denoising according to an embodiment of the present invention;

[0031] Figure 7 This is one of the stacked sections before (left) and after (right) the regularization of shallow data in an embodiment of the present invention;

[0032] Figure 8 This is the second of the superimposed sections before (left) and after (right) the regularization of shallow data in an embodiment of the present invention.

[0033] Specific implementation methods

[0034] In order to make the technical means and objectives adopted by the present invention easy to understand, the present invention is further described below in conjunction with specific implementation methods.

[0035] Unless otherwise specified, the experimental methods used in the following examples are conventional methods. Example 1

[0036] The method of the present invention specifically comprises the following steps:

[0037] a. As attached Figure 2As shown in a, read in 3D seismic data. The 3D seismic data needs to be shot gather data that has been pre-processed, the observation system has been defined, the trace header has been set, and the field static correction has been completed. The processing method of the present invention is highly targeted and only processes the seismic data sample points that contribute to the imaging of sandstone-type uranium reservoirs. Therefore, the read-in shot gather data is screened to only retain seismic traces with a shot offset less than or equal to 2500 meters, and the seismic trace length is truncated to 1500ms;

[0038] b. As attached Figure 2 As shown in b, the noise distribution characteristics of the input 3D seismic data are analyzed in detail, and the data are partitioned according to the main noise types using elevation maps, surface lithology distribution maps, etc. In order to ensure the denoising effect, the partitioning should be as accurate as possible, so as to facilitate the setting of different characteristic denoising parameters. Select typical single shots from different partitions, and establish a characteristic parameter library based on the interference frequency, apparent velocity, amplitude and other characteristics such as surface waves, wild values, and shallow refraction. Then, use methods such as frequency-constrained apparent velocity filtering and frequency-constrained energy thresholds to denoise the typical single shots;

[0039] c. As attached Figure 2 As shown in Figure c, the typical single-shot denoising effect is analyzed, especially the denoising effect of shallow refraction interference is quality controlled shot by shot to ensure that while removing the interference, the shallow effective reflection signal is not damaged as much as possible; according to the quality control results, the denoising parameters of the area with unsatisfactory effect are adjusted, the typical single-shot denoising scheme is iteratively optimized, and the threshold for manual judgment of denoising effect is set. The optimization of sample denoising parameters is completed when the expected result is achieved;

[0040] d. As attached Figure 2 As shown in Figure d, all single shots in each partition are denoised using the denoising parameters after iterative optimization of the samples, and representative single shots are selected (the selection rules can be based on actual needs, such as selecting at fixed intervals) for quality control of the denoising effect. The quality control results are analyzed. If there are problems with the denoising effect of individual single shots, the denoising parameters are fine-tuned according to the actual situation and denoised again, ultimately achieving the suppression and removal of surface waves, outliers, and shallow refraction in the entire area;

[0041] e. As attached Figure 2 As shown in Figure e, in order to carry out subsequent data regularization processing, the denoised shot gather seismic data are sorted into the CMP domain, and the CMP gathers are subjected to dynamic correction processing; the dynamic correction velocity used here is the stacking velocity carefully picked on the denoised data;

[0042] f. As attached Figure 2f, the distribution characteristics of the bin density, shot point and receiver point positions, coverage times and azimuth angles of the CMP gathers after dynamic correction in the time-space domain are analyzed; then, the time window, space window and frequency-wavenumber domain parameters of Fourier transform are defined, and the anti-leakage two-dimensional Fourier transform method is used to transform the seismic data from the time-space domain to the Fourier domain;

[0043] g. As attached Figure 2 As shown in Figure g, according to the data characteristics and actual data processing needs, the data regularization parameters are set (the specific parameters need to be determined according to the seismic data characteristics and target processing requirements), and the data is transformed back to the time-space domain using the anti-leakage two-dimensional Fourier inverse transform. After a forward-inverse transformation, the seismic data is transformed from the original irregular input data to regular output data;

[0044] h. As attached Figure 2 As shown in h, the regularized seismic data is subjected to reaction correction. Since regularization will produce certain random noise, the noise is appropriately suppressed on the reaction-corrected data set, and the denoised and regularized seismic data for the extremely shallow layer without dynamic correction is obtained; finally, the CMP data set with improved signal-to-noise ratio and regularization is output.

[0045] Attached Figure 3 , Attachment Figure 4 Provided with Figure 2 The processing effect diagram after iterative optimization denoising and shallow data regularization is shown in the figure. Figure 3 To iteratively optimize the single shot data before denoising, Figure 4 Iterative optimization of denoised single shot data. Comparing the two, we can find that the signal-to-noise ratio of the denoised single shot data is greatly improved, and typical surface waves, outliers and shallow refraction are well suppressed.

[0046] Attached Figure 5 To iteratively optimize the stacked profile before denoising, Figure 6 The section is stacked after iterative optimization and denoising. Comparison between the two shows that after iterative optimization and denoising, the surface wave interference, outliers and shallow refraction on the section have been basically suppressed, and the signal-to-noise ratio of the stacked section has been greatly improved.

[0047] Attached Figure 7 and attached Figure 8 The following are comparison diagrams of the stacked sections before and after the regularization of shallow data. Analysis of the stacked sections shows that the signal-to-noise ratio of the stacked sections after denoising is significantly improved compared with that before denoising, the continuity of the shallow target layer events of the stacked sections after regularization is significantly enhanced, and many shallow data gaps are filled or significantly improved.

[0048] In general, the embodiments fully illustrate that the present invention has a highly targeted method, which has a significant effect on improving the signal-to-noise ratio of extremely shallow seismic data for sandstone-type uranium deposits, and has important practical significance for improving the interpretation results of seismic data for sandstone-type uranium deposits in the next step, thereby further improving the exploration effect of sandstone-type uranium deposits.

[0049] Although some embodiments of the present invention have been given herein, those skilled in the art should understand that the embodiments of the present invention may be modified without departing from the spirit of the present invention. The above embodiments are merely exemplary and should not be used as a limitation on the scope of the present invention.

Claims

1. An iterative optimization method for improving the signal-to-noise ratio of extremely shallow seismic data in sandstone-type uranium deposits, characterized in that: The following steps are involved: a. Read in the 3D seismic data with the observation system defined in the early stage and after field static correction; b. Analyze the distribution area and seismic characteristics of shallow refraction in the data, subdivide the seismic data into regions according to their distribution characteristics, select typical single shots in different regions, and perform denoising on the typical single shots; c. Analyze the typical single-shot denoising effect, iteratively optimize and adjust the typical single-shot denoising scheme, set the threshold for manual judgment of denoising effect, and complete the optimization of sample denoising parameters when the expected result is achieved; d. De-noise the whole area with a single shot, and select a single shot for effect quality control. Fine-tune the denoising parameters according to the quality control results, and finally achieve the suppression and removal of surface waves, outliers, and shallow refraction in the whole area; e. Sorting the denoised seismic data into the CMP domain and performing dynamic correction processing; f. Set the data regularization parameters and transform the seismic data into the Fourier domain using the anti-leakage two-dimensional Fourier transform method; g. According to the data characteristics and actual data processing needs, the expected output parameters are set, and the data is transformed back to the time-space domain using inverse Fourier transform. After a forward-inverse transform, the seismic data is transformed from the original irregular input data to regular output data; h. Perform back-movement correction on the regularized seismic data to obtain the denoised and regularized seismic data for the extremely shallow layer without back-movement correction, and finally output the CMP gather data with improved signal-to-noise ratio.

2. The iterative optimization method for improving the signal-to-noise ratio of extremely shallow seismic data in sandstone-type uranium deposits according to claim 1 is characterized in that: The operation of step a is specifically: extracting seismic traces with an offset less than or equal to 2500 meters, and truncating the length of the seismic traces to 1500 ms.

3. The iterative optimization method for improving the signal-to-noise ratio of extremely shallow seismic data in sandstone-type uranium deposits according to claim 1, characterized in that: The specific operation of step b is: select typical single shots from different partitions, establish a feature parameter library based on the characteristics of surface wave, wild value, interference frequency of shallow refraction, apparent velocity, and amplitude, and then use the frequency-constrained apparent velocity filtering and frequency-constrained energy threshold methods to denoise the typical single shots.

4. The iterative optimization method for improving the signal-to-noise ratio of extremely shallow seismic data in sandstone-type uranium deposits according to claim 1, characterized in that: The specific operations of step c are as follows: analyze the typical single-shot denoising effect, especially to quality control the denoising effect of shallow refraction interference shot by shot, to ensure that while removing the interference, the shallow effective reflection signal is not damaged as much as possible; according to the quality control results, adjust the denoising parameters of the area with unsatisfactory effect, iteratively optimize the typical single-shot denoising scheme, set the threshold for manual judgment of denoising effect, and complete the optimization of sample denoising parameters when the expected result is achieved.

5. The iterative optimization method for improving the signal-to-noise ratio of extremely shallow seismic data in sandstone-type uranium deposits according to claim 1, characterized in that: When selecting single shots in step d, select them at fixed intervals.

6. The iterative optimization method for improving the signal-to-noise ratio of extremely shallow seismic data in sandstone-type uranium deposits according to claim 1, characterized in that: The speed of the dynamic correction described in step e is the superposition speed of fine picking on the denoised data.

7. The iterative optimization method for improving the signal-to-noise ratio of extremely shallow seismic data in sandstone-type uranium deposits according to claim 1, characterized in that: The regularized parameters in step f include: a time window, a space window and frequency-wavenumber domain parameters of Fourier transform.

8. The iterative optimization method for improving the signal-to-noise ratio of extremely shallow seismic data in sandstone-type uranium deposits according to claim 1, characterized in that: The technical means used in step h to obtain the denoised and regularized seismic data for the extremely shallow layer without dynamic correction is to suppress random noise on the data set.

Images