Multiple adaptive subtraction method based on primary wave protection

Abstract

The invention belongs to the technical field of exploration data processing, and particularly discloses a multiple wave adaptive subtraction method based on primary wave protection, which is based on well data and comprises the following steps: firstly, identifying a primary wave region and a multiple wave region through data calibration to obtain primary wave prior information, and then carrying out mathematical processing on the prior information to obtain a multiple wave subtraction result; and finally, adding information subjected to mathematical processing into a target function of a self-adaptive subtraction algorithm, and solving and suppressing multiple waves to achieve the purpose of protecting primary waves. According to the method, prior information and a multiple model are uniformly incorporated into a least square method of an adaptive subtraction algorithm, and a least square adaptive subtraction method based on prior information such as well data is provided. According to the method, multiple waves can be effectively suppressed on the basis of protecting primary waves. The method is suitable for data processing involving primary waves and multiple waves.

Description

Technical Field

[0001] The invention belongs to the technical field of exploration data processing, and relates to a method for suppressing multiple waves, in particular to a method for adaptively reducing multiple waves based on primary wave protection. Background Art

[0002] Multiple waves are a common type of interference wave in seismic exploration. The existence of multiple waves will cause the amplitude, frequency and phase of the effective reflection wave to be distorted, thus having a serious adverse effect on the identification of the true attribute relationship of the effective wave. Therefore, when processing seismic exploration data, suppressing multiple waves becomes one of the key steps in data processing. At present, there are two main methods for suppressing multiple waves, namely, a mathematical transformation algorithm based on the time difference between the primary wave and the multiple waves, and a multiple wave suppression method based on an adaptive subtraction algorithm. However, in the actual data processing process, the first method often has a poor effect due to the small time difference between the primary wave and the multiple waves.

[0003] In the second method, although the specific types of adaptive subtraction algorithms are rich and varied, for example, pattern recognition adaptive subtraction, curvelet transform adaptive subtraction, etc. all have their specific application scenarios, the better application of adaptive subtraction algorithms is still based on the following two assumptions: the orthogonality assumption between the effective wave and the multiple wave and the wavelet consistency assumption. When the assumptions are not met, the adaptive subtraction algorithm will have problems of insufficient matching or excessive matching. These problems lead to residual multiple waves or damage to the primary wave in the adaptive subtraction process. In particular, when the multiple waves and the primary wave phase axis cross or almost overlap, it is difficult to protect the primary wave when using conventional adaptive subtraction algorithms to suppress the multiple waves. Summary of the invention

[0004] The purpose of the present invention is to provide a method for adaptively reducing multiple waves based on primary wave protection, by establishing a multiple wave model and primary wave prior information, performing mathematical processing, and utilizing the least squares method, so as to achieve the purpose of suppressing multiple waves by using the method for adaptively reducing multiple waves on the basis of protecting the primary wave.

[0005] To achieve the above object, the present invention provides a method for adaptively reducing multiple waves based on primary wave protection, which comprises the following steps performed in sequence:

[0006] S1. Generate work area masking operator

[0007] Based on the well data, using the information of seismic records, VSP corridor stacked sections and acoustic logging synthetic records, the data is calibrated in the stacked sections without multiple wavefronts to obtain the primary wave area and the multiple wave area;

[0008] The primary wave region and the multiple wave region both include coordinate information and time information;

[0009] If it is not marked as a multiple wave area, it is assigned a value of 0, and other areas are assigned a value of 1 to generate the work area masking operator f;

[0010] S2. Synthesizing prior information

[0011] Seismic data is calibrated through VSP corridor stacking profiles and acoustic logging synthetic record information, and the horizon is picked for the primary wave event axis to obtain the horizon information;

[0012] Using acoustic logging to synthesize record information and layer information, synthesize a priori information of the primary wave There are m1 layers in total, written in matrix form, and we get the primary wave prior information X1;

[0013] S3. Prediction of multiple wave models

[0014] The multiple wave model X2 is predicted by the multiple wave prediction algorithm, and the multiple wave model X'2 after dynamic correction is obtained by processing;

[0015] S4. Generate seismic sections with common offset and common azimuth

[0016] Perform dynamic correction processing and offset vector slice data domain sorting processing on the stacked section without multiple wavefronts to generate a seismic section Y with common offset and common azimuth;

[0017] S5. Establishing the objective function

[0018] For each offset vector slice data domain, in the sense of least squares, the objective function J(A) = argmin||Y-AX||+λA is established T A;

[0019] Among them, X = [X1, X'2], which represents the matrix composed of the primary wave prior information X1 and the multiple wave model X'2;

[0020] A=[A1,A2], A1 represents the adaptive matching operator of the primary wave prior information X1, A2 represents the adaptive matching operator of the multiple wave model X'2, and A is the total adaptive matching operator;

[0021] λA T A is the regularization term, λ is the regularization parameter;

[0022] S6. Solve the objective function and suppress the multiple waves to obtain the seismic data after suppressing the multiple waves.

[0023] As a limitation of the present invention, in step S3, the predicted multiple wave model is There are m2 layers in total, written in matrix form, we get the multiple wave model X2;

[0024] The post-NMO correction multiple wave model X'2 is obtained by performing NMO correction processing and offset vector slice data domain sorting processing on the multiple wave model X2 to generate a seismic profile with a common offset and a common azimuth.

[0025] Among them, in step S5, the mathematically processed prior information is incorporated into the adaptive subtraction algorithm to achieve the purpose of protecting the primary wave.

[0026] As another limitation of the present invention, in step S6, the objective function is solved to obtain the total adaptive matching operator A and the matched multiple waves A2X2.

[0027] As a third limitation of the present invention, in step S6, the suppression of multiple waves is performed using the formula Y-fA2X'2, and the seismic data after subtracting the multiple waves is subjected to a reaction correction process to obtain the seismic data after suppressing the multiple waves.

[0028] As a fourth limitation of the present invention, the VSP corridor stacked profile is obtained by processing VSP data.

[0029] As a fifth limitation of the present invention, the stacked section without multiple wavefronts is obtained by interpreting the primary wave velocity based on the acoustic logging velocity as a reference value and in combination with geological knowledge.

[0030] As a further limitation of the present invention, the geological knowledge includes sonic velocity, lithology and structural information.

[0031] Due to the adoption of the above scheme, the present invention has the following beneficial effects compared with the prior art:

[0032] The present invention uses prior information for adaptive subtraction to achieve effective protection of the primary wave, while the conventional method in the industry does not use the prior information of the primary wave, but directly uses the original data and the predicted multiple wave model for adaptive subtraction, which is easy to damage the effective primary wave.

[0033] The present invention establishes a priori information of a primary wave, performs mathematical processing, and utilizes an improved least square method to overcome the problems of insufficient matching and over-matching during adaptive matching, thereby making the process of adaptive subtraction more fidelity. It can effectively suppress multiple waves by using a multiple wave adaptive subtraction method while protecting the primary wave.

[0034] The invention is applicable to the multiple wave adaptive subtraction technology in exploration data processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0036] Figure 1 A seismic data diagram showing suppression of multiple wavefronts in an embodiment of the present invention;

[0037] Figure 2 This is a diagram of seismic data obtained after suppressing multiple waves in an embodiment of the present invention;

[0038] Figure 3 This is a diagram of seismic data after suppressing multiple waves obtained by conventional adaptive subtraction processing in an embodiment of the present invention.

[0039] Figure 4 This is a seismic data diagram of suppressing multiple wavefronts in Experiment I of the embodiment of the present invention;

[0040] Figure 5 This is a graph of the prior information of the primary wave obtained in the verification experiment I of the embodiment of the present invention;

[0041] Figure 6 This is a multiple wave model diagram obtained in the verification experiment I of the embodiment of the present invention;

[0042] Figure 7 This is a diagram of seismic data after suppressing multiple waves obtained in Experiment I for verifying the embodiment of the present invention;

[0043] Figure 8 This is a seismic data diagram after suppressing multiple waves obtained by conventional adaptive subtraction processing in Experiment I of the embodiment of the present invention;

[0044] Fig. 9 This is a seismic data diagram of suppressing multiple wavefronts in Experiment II for verifying the embodiment of the present invention;

[0045] Fig.10 This is a graph of the prior information of a primary wave in Experiment II for verifying the embodiment of the present invention;

[0046] Fig.11 This is a multiple wave model diagram in Experiment II for verifying the embodiment of the present invention;

[0047] Fig.12 This is a seismic data diagram after suppressing multiple waves obtained in Experiment II for verifying the embodiment of the present invention;

[0048] Fig.13 This is a diagram of seismic data after suppressing multiple waves obtained through conventional adaptive subtraction processing in Experiment II for verifying the embodiment of the present invention. DETAILED DESCRIPTION

[0049] The present invention is further described below in conjunction with embodiments, but those skilled in the art should understand that the present invention is not limited to the following embodiments, and any improvements and equivalent changes made based on the specific embodiments of the present invention are within the scope of protection of the claims of the present invention.

[0050] Embodiment Multiple wave adaptive reduction method based on primary wave protection

[0051] In a field seismic survey, seismic data are obtained. In the seismic profile, multiple waves are intermittent high-frequency phase axes. Before suppressing the multiple waves, such as Figure 1 ,Affected by the multiple waves, the primary wave is seriously disturbed by the multiple waves. If the multiple waves are suppressed by the conventional adaptive subtraction process, it is easy to damage the phase axis of the primary wave. Focus on the primary wave of the earthquake in the white box to observe whether the adaptive subtraction causes damage to the primary wave.

[0052] This embodiment is a method for adaptively reducing multiple waves based on primary wave protection. This method is used to effectively suppress high-frequency multiple waves while protecting the primary wave in seismic profile data.

[0053] (I) The specific method of this embodiment is as follows:

[0054] S1. Generate work area masking operator

[0055] Based on the well data, using the information of seismic records, VSP corridor stacked sections and acoustic logging synthetic records, the data is calibrated in the stacked sections without multiple wavefronts to obtain the primary wave area and the multiple wave area;

[0056] The primary wave area and the multiple wave area both include coordinate information and time information;

[0057] If it is not marked as a multiple wave area, it is assigned a value of 0, and other areas are assigned a value of 1 to generate the work area masking operator f;

[0058] S2. Synthesizing prior information

[0059] Seismic data is calibrated through VSP corridor stacking profiles and acoustic logging synthetic record information, and the horizon is picked for the primary wave event axis to obtain the horizon information;

[0060] Using acoustic logging to synthesize record information and layer information, synthesize a priori information of the primary wave There are m1 layers in total, written in matrix form, and we get the primary wave prior information X1;

[0061] S3. Prediction of multiple wave models

[0062] Through the multiple wave prediction algorithm, the multiple wave model is predicted as follows: There are m2 layers in total, written in matrix form, we get the multiple wave model X2;

[0063] Performing dynamic correction processing and offset vector slice data domain sorting processing on the multiple wave model X2, generating a seismic section with common offset and common azimuth to obtain a dynamic correction multiple wave model X'2;

[0064] S4. Generate seismic sections with common offset and common azimuth

[0065] Perform dynamic correction processing and offset vector slice data domain sorting processing on the stacked section without multiple wavefronts to generate a seismic section Y with common offset and common azimuth;

[0066] S5. Establishing the objective function

[0067] For each offset vector slice data domain, in the sense of least squares, the objective function J(A) = argmin||Y-AX||+λA is established T A;

[0068] Among them, X = [X1, X'2], which represents the matrix composed of the primary wave prior information X1 and the multiple wave model X'2;

[0069] A=[A1,A2], A1 represents the adaptive matching operator of the primary wave prior information X1, A2 represents the adaptive matching operator of the multiple wave model X'2, and A is the total adaptive matching operator;

[0070] λA T A is the regularization term, λ is the regularization parameter;

[0071] S6. Solve the objective function and suppress the multiple waves to obtain the seismic data after suppressing the multiple waves.

[0072] Solve the objective function and obtain the total adaptive matching operator A and the matched multiple waves A2X2;

[0073] Using the formula Y-fA2X'2, we can get the seismic data after subtracting the multiple waves;

[0074] Among them, the seismic data after suppressing multiple waves, such as Figure 2 shown.

[0075] (ii) Comparative Example:

[0076] This comparative embodiment uses conventional adaptive subtraction processing, that is, there is no prior information of the primary wave in the adaptive subtraction process, and directly uses the original data and the predicted multiple wave model for adaptive subtraction to obtain seismic data after suppressing the multiple waves, such as Figure 3 shown.

[0077] in, Figure 3 It is shown that conventional adaptive subtraction processing causes serious damage to the primary wave, for example Figure 1 The white box selects the area affected by multiple waves. Figure 3 The primary wave is seriously damaged in the corresponding area, and the method of the present invention obtains Figure 2 The results show that compared with conventional adaptive subtraction processing, the present invention has a good primary wave protection effect when suppressing high-frequency multiple waves in seismic profile data processing.

[0078] (III) Verification experiment:

[0079] This experiment generates model data with known relevant information including prior information through forward modeling, which is used to verify whether the objective function containing prior information in the method of the present invention is effective.

[0080] Verification Experiment I is a case where the phase axes of multiple waves and primary waves cross, and verifies whether the objective function containing prior information in the method of the present invention can effectively suppress multiple waves.

[0081] S11. Data pre-processing

[0082] Before suppressing multiple waves, the model data is generated by forward modeling. The resulting seismic data is as follows: Figure 4 shown.

[0083] S12. Generate work area masking operator

[0084] Since this is model data and the multiple wave model is known, the multiple wave area is assigned a value of 1 and the others are assigned a value of 0.

[0085] S13. Synthesizing prior information

[0086] Since it is model data, the prior information of the primary wave is known, such as Figure 5 .

[0087] S14. Prediction of multiple wave models

[0088] This step is basically the same as S3, except that the data substituted is generated by forward modeling or obtained from step S11 to step S13, and the multiple wave model diagram is obtained through the multiple wave prediction algorithm, such as Figure 6 shown.

[0089] S15. Generate seismic sections with common offset and common azimuth

[0090] Perform dynamic correction processing and offset vector slice data (OVT) domain sorting processing on the stacked section without multiple wavefronts to generate a seismic section Y with common offset and common azimuth;

[0091] S16. Establishing the objective function

[0092] For each offset vector slice data domain, in the sense of least squares, the objective function J(A) = argmin||Y-AX||+λA is established T A;

[0093] Wherein, X = [X1, X'2], which represents the matrix composed of the primary wave model X1 and the multiple wave model X'2;

[0094] A=[A1,A2], A1 represents the adaptive matching operator of the primary wave model X1, A2 represents the adaptive matching operator of the multiple wave model X'2, and A is the total adaptive matching operator;

[0095] λA T A is the regularization term, λ is the regularization parameter;

[0096] S17. Solving the objective function

[0097] Solve the objective function and obtain the total adaptive matching operator A and the matched multiple waves A2X2;

[0098] Using the formula Y-fA2X'2, we can get the seismic data after subtracting the multiple waves;

[0099] The seismic data after multiple waves are subtracted are processed by reaction correction to obtain the seismic data after multiple waves are suppressed, such as Figure 7 shown.

[0100] Comparative experiment:

[0101] This comparative experiment uses conventional adaptive subtraction processing, that is, there is no prior information of the primary wave in the adaptive subtraction process, and directly uses the original data and the predicted multiple wave model for adaptive subtraction to obtain the seismic data after suppressing the multiple waves, such as Figure 8 shown.

[0102] in, Figure 8 The circled area is where the multiple waves and the primary wave intersect. Conventional adaptive subtraction processing changes the amplitude characteristics of the primary wave, while the method used in this verification experiment Figure 7 In the effect shown, the amplitude characteristics of the primary wave are effectively maintained. The results show that compared with conventional adaptive subtraction processing, in the case of a crossover of the phase axis of the multiple waves and the primary wave, the objective function containing prior information in the method of the present invention can effectively suppress the multiple waves, and the present invention has a good primary wave protection effect.

[0103] Verification Experiment II is a case where the times of the events of the multiple waves and the primary wave are close to each other, and is used to verify whether the objective function containing prior information in the method of the present invention can effectively suppress the multiple waves.

[0104] The specific method is basically the same as that in the verification experiment I, the only difference is that the model data processed is different, so the data processing results are different.

[0105] In the corresponding step S11, before suppressing the multiple waves, the seismic data obtained in this verification experiment are as follows: Fig. 9 shown.

[0106] Corresponding to step S13, the primary wave prior information diagram obtained in this verification experiment is as follows: Fig.10 shown.

[0107] Corresponding to step S14, the multiple wave model diagram obtained in this verification experiment is as follows: Fig.11 shown.

[0108] In the corresponding step S17, the verification experiment performs a reaction correction process on the seismic data after subtracting the multiple waves, and obtains the seismic data after suppressing the multiple waves, such as Fig.12 shown.

[0109] According to the comparative experiment in verification experiment Ⅰ, conventional adaptive subtraction processing is performed to obtain the seismic data after suppressing multiple waves, as shown in Fig.13 shown.

[0110] in, Fig.13 It shows that conventional adaptive subtraction processing causes serious damage to the primary wave, while the method of the present invention obtains Fig.12 The results show that, compared with the conventional adaptive subtraction processing, the present invention has a good primary wave protection effect when the multiple waves and the primary wave are close in time.

[0111] It should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention is described in detail with reference to the above embodiments, those skilled in the art can still modify the technical solutions described in the above embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the scope of protection of the claims of the present invention.

Claims

1. A method for adaptively reducing multiple waves based on primary wave protection, characterized in that: The method comprises the following steps performed in sequence: S1. Generate work area masking operator Using seismic records, VSP corridor stacked sections and acoustic logging synthetic record information, the data is calibrated in the stacked sections without multiple wavefronts to obtain the primary wave area and the multiple wave area. The primary wave region and the multiple wave region both include coordinate information and time information; If it is not marked as a multiple wave area, it is assigned a value of 0, and other areas are assigned a value of 1 to generate the work area masking operator f; S2. Synthesizing prior information Seismic data is calibrated through VSP corridor stacking profiles and acoustic logging synthetic record information, and the horizon is picked for the primary wave event axis to obtain the horizon information; Using acoustic logging to synthesize record information and layer information, synthesize a priori information of the primary wave There are m1 layers in total, written in matrix form, and we get the primary wave prior information X1; S3. Prediction of multiple wave models The multiple wave model X2 is predicted by the multiple wave prediction algorithm, and the multiple wave model X'2 after dynamic correction is obtained by processing; S4. Generate seismic sections with common offset and common azimuth Perform dynamic correction processing and offset vector slice data domain sorting processing on the stacked section without multiple wavefronts to generate a seismic section Y with common offset and common azimuth; S5. Establishing the objective function For each offset vector slice data domain, in the sense of least squares, the objective function J(A) = argmin||Y-AX||+λA is established T A; Among them, X = [X1, X'2], which represents the matrix composed of the primary wave prior information X1 and the multiple wave model X'2; A=[A1,A2], A1 represents the adaptive matching operator of the primary wave prior information X1, A2 represents the adaptive matching operator of the multiple wave model X'2, and A is the total adaptive matching operator; λA T A is the regularization term, λ is the regularization parameter; S6. Solve the objective function and suppress the multiple waves to obtain the seismic data after suppressing the multiple waves.

2. The method for adaptively reducing multiple waves based on primary wave protection according to claim 1, characterized in that: In step S3, the predicted multiple wave model is X 21 ,X 22 ,…,X 2m2 , a total of m2 layers, written in matrix form, we get the multiple wave model X2; The post-NMO correction multiple wave model X'2 is obtained by performing NMO correction processing and offset vector slice data domain sorting processing on the multiple wave model X2 to generate a seismic profile with a common offset and a common azimuth.

3. The method for adaptively reducing multiple waves based on primary wave protection according to claim 1, characterized in that: In step S6, the objective function is solved to obtain the total adaptive matching operator A and the matched multiple waves A2X2.

4. The method for adaptively reducing multiple waves based on primary wave protection according to claim 1, characterized in that: In step S6, the multiple waves are suppressed by using the formula Y-fA2X'2, and the seismic data after subtracting the multiple waves are subjected to reaction correction processing to obtain the seismic data after suppressing the multiple waves.

5. The method for adaptively reducing multiple waves based on primary wave protection according to claim 1, characterized in that: The VSP corridor stacked profile is obtained by processing VSP data.

6. The method for adaptively reducing multiple waves based on primary wave protection according to any one of claims 1 to 5, characterized in that: The stacked section without multiple wavefronts is obtained by interpreting the primary wave velocity with the acoustic logging velocity as a reference value in combination with geological knowledge.

7. The method for adaptively reducing multiple waves based on primary wave protection according to claim 6, characterized in that: The geological knowledge includes acoustic velocity, lithology and structural information.

Images