Irregular multiple wave separation method and system

Abstract

The present invention relates to a method and system for separating irregular multiple waves. The separation method and system break through the assumption of regular interference in traditional methods. Based on focusing transformation, by analyzing the energy distribution of different types of reflected waves in the focusing domain, high-resolution τ‑p transformation and regularized non-steady-state regression adaptive matching filtering are used to simultaneously separate primary waves, free surface multiple waves and interlayer multiple waves. The method specifically includes a data acquisition step, a focusing operator construction step, a focusing transformation step, a τ‑p transformation step, a primary wave separation step, a free surface multiple wave separation step, and an interlayer multiple wave separation step. The free surface multiple waves and interlayer multiple waves accurately separated by the present invention can lay a good data foundation for subsequent multiple wave suppression and imaging applications.

Description

Technical Field

[0001] The present invention belongs to the field of geological exploration technology, and relates to seismic exploration technology, and in particular, to an irregular multiple wave separation method and system. Background Art

[0002] In seismic exploration, especially in marine seismic exploration, multiple waves in seismic data can seriously affect the imaging quality of underground structures. Therefore, geophysical experts have conducted a lot of research on the separation of multiple waves. Currently, the commonly used multiple wave separation methods can be divided into two categories: one is based on filtering theory, such as Radon transform, predictive deconvolution, etc. The other is based on the prediction subtraction method of wave theory, such as feedback iteration method, inverse scattering series, etc. The latter is usually data-driven. Due to its higher accuracy in multiple wave prediction and stronger adaptability to complex situations, it has received more and more attention in recent decades. However, most separation methods based on wave theory have a priori assumption that multiple waves are regular interference. This requires that in the separation process, irregular interference must be eliminated first. In other words, most separation methods for free surface multiple waves require data without interlayer multiple waves as a driver, while separation methods for interlayer multiple waves also require the removal of free surface multiple waves. If no prior processing is performed, accurate prediction of the corresponding multiple waves cannot be achieved.

[0003] However, the increasingly complex underground stratigraphic structures in seismic exploration often develop complex types of multiple waves. Under complex geological conditions, free surface multiple waves and interlayer multiple waves often appear simultaneously in seismic recording data. In this case, traditional methods will no longer meet the assumption that multiple waves are regular interference and are no longer applicable to the prediction and suppression of multiple waves in this type of situation. In addition, in order to meet the high-precision requirements of seismic exploration, especially for complex structures, the use of free surface multiple waves and interlayer multiple waves is gradually regarded as an effective method to obtain more underground stratigraphic information. However, for seismic data that contain both free surface multiple waves and interlayer multiple waves, research on methods for simultaneously predicting free surface multiple waves and interlayer multiple waves is relatively rare. How to separate the two at the same time without affecting each other is a difficult problem that needs to be solved urgently. Summary of the invention

[0004] In view of the above problems existing in the prior art, the present invention provides a method and system for separating irregular multiple waves, which can simultaneously separate primary waves, free surface multiple waves and interlayer multiple waves.

[0005] In order to achieve the above object, the present invention provides an irregular multiple wave separation method, comprising the following steps:

[0006] Data acquisition steps: obtaining seismic record data;

[0007] Focusing operator construction steps: The positive focusing transformation and the inverse focusing transformation are defined as:

[0008] Q=FP (1)

[0009] P=GQ (2)

[0010] Where Q is the total response of the seismic record data in the focus domain, F is the positive focusing operator, and G is the anti-focusing operator;

[0011] Velocity analysis is performed on seismic data to obtain the velocity field, and the estimated primary wave of the seismic data is obtained based on the velocity field forward modeling. To estimate a wave is the anti-focusing operator G, is the positive focusing operator;

[0012] Focusing transformation step: according to the relationship between the reflected waves of the seismic recording data, the seismic recording data is subjected to positive focusing transformation to obtain the focusing domain data;

[0013] τ-p transformation step: Use high-resolution τ-p transformation to transfer the focus domain data to the τ-p domain, where the primary wave energy is converted into a straight line at time zero;

[0014] The primary wave separation step is: removing the energy of the straight line in the τ-p domain, transforming the removed straight line energy back to the focus domain with high resolution τ-p, and then focusing and transforming it back to the original time-space domain of the seismic record data to obtain the separated primary wave;

[0015] Free surface multiple wave separation step: after the linear energy is cut off, the high-resolution τ-p of other energies in the τ-p domain is inversely transformed back to the focusing domain to obtain the multiple wave response in the focusing domain. According to the similarity between the free surface multiple waves in the focusing domain and the seismic recording data in the original time-space domain, the regularized non-steady-state regression adaptive matched filtering method is used to extract the free surface multiple waves in the focusing domain, and the inverse focusing transformation is performed back to the original time-space domain of the seismic recording data to obtain the separated free surface multiple waves.

[0016] Interlayer multiple wave separation step: subtract the separated primary wave and the separated free surface multiple waves from the seismic record data to obtain the interlayer multiple waves.

[0017] Preferably, in the focusing transformation step, the relationship between the reflected waves of the seismic recording data is the relationship between the primary wave and the multiple waves of the seismic recording data in the frequency domain, which is expressed as:

[0018] P=ΔP+M S +M I =ΔP+ΔPAP+M I (3)

[0019] Where P is the seismic record data matrix containing the primary wave, free surface multiple waves and interlayer multiple waves, ΔP is the primary wave, M S is the free surface multiple wave, M I is the interlayer multiple wave, and A is the weighting factor related to the surface reflection;

[0020] Combining formula (1) and formula (3), we have:

[0021]

[0022] Where I is the focal point of the primary wave focused around the source point.

[0023] Preferably, in the focusing operator construction step: based on the velocity field, the Born approximation forward modeling method is used to forward model the estimated primary wave of the seismic recording data.

[0024] In order to achieve the above object, the present invention also provides an irregular multiple wave separation system, comprising:

[0025] Data acquisition steps: obtaining seismic record data;

[0026] A focusing operator construction module, used to construct positive focusing operators and anti-focusing operators;

[0027] A focusing transformation module performs positive focusing transformation on the seismic recording data according to the relationship between the reflected waves of the seismic recording data to obtain the focusing domain data;

[0028] A τ-p transformation module, used to transfer the focus domain data to the τ-p domain using a high-resolution τ-p transformation, in which the primary wave energy is converted into a straight line at time zero;

[0029] The primary wave separation module is used to remove the energy of the straight line in the τ-p domain, transform the removed straight line energy back to the focus domain with high resolution τ-p, and then focus and transform it back to the original time-space domain of the seismic record data to obtain the separated primary wave;

[0030] The free surface multiple wave separation module is used to perform a high-resolution τ-p inverse transformation of other energies in the τ-p domain after the linear energy is cut off back to the focusing domain to obtain the multiple wave response in the focusing domain. According to the similarity between the free surface multiple waves in the focusing domain and the seismic recording data in the original time-space domain, the regularized non-steady-state regression adaptive matched filtering method is used to extract the free surface multiple waves in the focusing domain, and the inverse focusing transformation is performed back to the original time-space domain of the seismic recording data to obtain the separated free surface multiple waves.

[0031] The interlayer multiple wave separation module is used to subtract the separated primary wave and the separated free surface multiple waves from the seismic record data to obtain the interlayer multiple waves.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] The irregular multiple wave separation method and system of the present invention breaks through the assumption that multiple waves are regular interference in the traditional multiple wave separation method. Based on the focus transformation, by analyzing the energy distribution of different types of reflected waves in the focus domain, the high-resolution τ-p transformation and regularized non-steady-state regression adaptive matched filtering are used to simultaneously separate the primary wave, free surface multiple waves and interlayer multiple waves. The free surface multiple waves and interlayer multiple waves accurately separated by the present invention can lay a good data foundation for subsequent multiple wave suppression and imaging applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a flow chart of the irregular multiple wave separation method according to an embodiment of the present invention;

[0035] Figure 2a The velocity field and reflection diagrams with different propagation paths described in the embodiment of the present invention;

[0036] Figure 2b The simulated seismic cannon recording diagram according to the embodiment of the present invention;

[0037] Figure 2c This is a focus area result diagram according to an embodiment of the present invention;

[0038] Figure 2d for Figure 2c τ-p domain result diagram;

[0039] Figure 3a A schematic diagram of a primary wave separated by the irregular multiple wave separation method and system according to an embodiment of the present invention;

[0040] Figure 3b A schematic diagram of free surface multiple waves separated by the irregular multiple wave separation method and system according to an embodiment of the present invention;

[0041] Figure 3c A schematic diagram of inter-layer multiple waves separated by the irregular multiple wave separation method and system according to an embodiment of the present invention.

[0042] In the figure, 1 and 2 are primary waves, 4 and 7 are interlayer multiple waves, and 3, 5, 6, 8-11 are free surface multiple waves. DETAILED DESCRIPTION

[0043] The present invention is described in detail below by way of exemplary embodiments. However, it should be understood that elements, structures, and features in one embodiment may also be beneficially combined in other embodiments without further description.

[0044] During seismic exploration, under complex geological conditions, free surface multiple waves and interlayer multiple waves often appear simultaneously in seismic recording data. In the present invention, the free surface multiple waves and interlayer multiple waves that exist simultaneously are referred to as irregular multiple waves. The focusing transformation developed by SRME can be regarded as a weighted cross-correlation process, which can reduce and improve the order of free surface multiple waves. Therefore, the present invention provides a method and system for separating irregular multiple waves, which is based on focusing transformation, analyzes the energy distribution of different types of reflected waves in the focusing domain, and uses high-resolution τ-p transformation and regularized non-steady-state regression adaptive matching filtering to simultaneously achieve the separation of primary waves, free surface multiple waves and interlayer multiple waves. The above method is described in detail below through specific embodiments.

[0045] See also Figure 1 The embodiment of the present invention provides an irregular multiple wave separation method, the steps of which are:

[0046] S1. Data acquisition step: acquiring earthquake record data.

[0047] S2. Focusing operator construction steps: The positive focusing transformation and the negative focusing transformation are defined as:

[0048] Q=FP (1)

[0049] P=GQ (2)

[0050] Where Q is the total response of the seismic record data in the focus domain, F is the positive focusing operator, and G is the anti-focusing operator;

[0051] Velocity analysis is performed on seismic data to obtain the velocity field, and the estimated primary wave of the seismic data is obtained based on the velocity field forward modeling. To estimate a wave is the anti-focusing operator G, is the positive focusing operator.

[0052] Specifically, the Born approximation forward modeling method is used to forward model the estimated primary wave of the seismic data based on the velocity field.

[0053] S3, focusing transformation step: according to the relationship between the reflected waves of the seismic recording data, a positive focusing transformation is performed on the seismic recording data to obtain the focusing domain data.

[0054] Specifically, the relationship between the reflected waves of the seismic recording data is the relationship between the primary wave and the multiple waves of the seismic recording data in the frequency domain, which can be expressed as:

[0055] P=ΔP+M S +M I =ΔP+ΔPAP+M I (3)

[0056] Where P is the seismic record data matrix containing the primary wave, free surface multiple waves and interlayer multiple waves, ΔP is the primary wave, M S is the free surface multiple wave (according to the reciprocity principle between the source point and the receiver point, M S including various free surface multiple waves with different propagation paths), M I is the interlayer multiple wave, and A is the weighting factor related to the surface reflection;

[0057] Combining formula (1) and formula (3), we have:

[0058]

[0059] Where I is the focal point of the primary wave focused around the source point (i.e., zero position).

[0060] It should be noted that in formula (4), it is assumed that P' = AP represents the total response recovered from the free surface multiple waves. Since A is a weighting factor related to the surface reflection, the free surface multiple waves are converted into responses similar to the original seismic record data matrix P, that is, the free surface multiple waves achieve an order reduction process. The last term is the response of the interlayer multiple waves, which will be distributed at other locations in the focal domain. That is, in the focal domain, the primary waves are distributed at the focus point at the zero position, the free surface multiple waves show the form of the original seismic record, and the interlayer multiple waves are distributed at other locations.

[0061] It should also be noted that in traditional SRME, according to the feedback model, the relationship between the primary wave and the multiple waves can be expressed in the frequency domain using a data matrix as follows:

[0062] P=ΔP+M=ΔP+ΔPAP

[0063] In conventional SRME, interlayer multiple waves are regarded as "primary waves", and the irregular multiple waves described in the present invention are not applicable. In this embodiment, the relationship between the primary waves and the multiple waves in conventional SRME is corrected, and the mutual influence between the free surface multiple waves and the interlayer multiple waves is considered, so that the irregular multiple waves described in the present invention can be applied.

[0064] S4, τ-p transformation step: Use high-resolution τ-p transformation to transfer the focus domain data to the τ-p domain. In the τ-p domain, the primary wave energy is converted into a straight line at time zero.

[0065] It should be noted that since the primary wave appears as scissor-divergent energy focused around the zero point in the focusing domain, it is converted into a straight line at time zero through high-resolution τ-p forward transformation. Therefore, the subsequent primary wave separation can be simpler and more accurate.

[0066] S5. Primary wave separation step: cut off the energy of the straight line in the τ-p domain, transform the cut straight line energy back to the focusing domain with high resolution τ-p, and then focus and transform it back to the original time-space domain of the seismic record data to obtain the separated primary wave.

[0067] S6. Free surface multiple wave separation step: After cutting off the linear energy, perform a high-resolution τ-p inverse transformation of other energies in the τ-p domain back to the focusing domain to obtain the multiple wave responses in the focusing domain. According to the similarity between the free surface multiple waves in the focusing domain and the seismic recording data in the original time-space domain, a regularized non-steady-state regression adaptive matched filtering method is used to extract the free surface multiple waves in the focusing domain, and then perform an inverse focusing transformation back to the original time-space domain of the seismic recording data to obtain the separated free surface multiple waves.

[0068] S7. Interlayer multiple wave separation step: subtract the separated primary wave and the separated free surface multiple waves from the seismic record data to obtain the interlayer multiple waves.

[0069] The above-mentioned irregular multiple wave separation method of this embodiment is based on focusing transformation, which breaks through the assumption that multiple waves are regular interference in traditional methods. By analyzing the energy distribution of different types of reflected waves in the focusing domain, high-resolution τ-p transformation and regularized non-steady-state regression adaptive matched filtering are used to simultaneously realize the separation of primary waves, free surface multiple waves and interlayer multiple waves. The accurately separated free surface multiple waves and interlayer multiple waves can lay a good data foundation for subsequent multiple wave suppression and imaging applications.

[0070] The embodiment of the present invention further provides an irregular multiple wave separation system, comprising:

[0071] A data acquisition module, used for acquiring earthquake record data;

[0072] A focusing operator construction module, used to construct positive focusing operators and anti-focusing operators;

[0073] A focusing transformation module performs positive focusing transformation on the seismic recording data according to the relationship between the reflected waves of the seismic recording data to obtain the focusing domain data;

[0074] A τ-p transformation module, for transferring the focus domain data to the τ-p domain using a high-resolution τ-p transformation, where the primary wave energy is converted into a straight line at time zero;

[0075] The primary wave separation module is used to remove the energy of the straight line in the τ-p domain, transform the removed straight line energy back to the focus domain with high resolution τ-p, and then focus and transform it back to the original time-space domain of the seismic recording data to obtain the separated primary wave;

[0076] The free surface multiple wave separation module is used to perform a high-resolution τ-p inverse transformation of other energies in the τ-p domain after the linear energy is cut off back to the focusing domain to obtain the multiple wave response in the focusing domain. According to the similarity between the free surface multiple waves in the focusing domain and the seismic recording data in the original time-space domain, the regularized non-steady-state regression adaptive matched filtering method is used to extract the free surface multiple waves in the focusing domain, and the inverse focusing transformation is performed back to the original time-space domain of the seismic recording data to obtain the separated free surface multiple waves.

[0077] The interlayer multiple wave separation module is used to subtract the separated primary wave and the separated free surface multiple waves from the seismic record data to obtain the interlayer multiple waves.

[0078] It should be noted that since the primary wave in the focal domain appears as energy focused around the zero point with a scissors-divergent shape, the τ-p transformation module uses a high-resolution τ-p transformation to transfer the focal domain data to the τ-p domain. After the high-resolution τ-p forward transformation, it is converted into a straight line at the zero moment. Therefore, the subsequent primary wave separation can be simpler and more accurate.

[0079] Specifically, in the focusing operator construction module, the specific steps of constructing the positive focusing operator and the anti-focusing operator are:

[0080] The positive focusing transformation and the negative focusing transformation are defined as:

[0081] Q=FP (1)

[0082] P=GQ (2)

[0083] Where Q is the total response of the seismic record data in the focus domain, F is the positive focusing operator, and G is the anti-focusing operator;

[0084] Velocity analysis is performed on seismic data to obtain the velocity field, and the estimated primary wave of the seismic data is obtained based on the velocity field forward modeling. To estimate a wave is the anti-focusing operator G, Specifically, the estimated primary wave of the seismic data is obtained by using the Born approximation forward modeling method based on the velocity field.

[0085] Specifically, in the focus transformation module, the relationship between the reflected waves of the seismic recording data is the relationship between the primary wave and the multiple waves of the seismic recording data in the frequency domain, which is expressed as:

[0086] P=ΔP+M S +M I =ΔP+ΔPAP+M I (3)

[0087] Where P is the seismic record data matrix containing the primary wave, free surface multiple waves and interlayer multiple waves, ΔP is the primary wave, M S is the free surface multiple wave, M I is the interlayer multiple wave, and A is the weighting factor related to the surface reflection;

[0088] Combining formula (1) and formula (3), we have:

[0089]

[0090] Where I is the focal point of the primary wave focused around the source point.

[0091] It should be noted that in formula (4), it is assumed that P' = AP represents the total response recovered from the free surface multiple waves. Since A is a weighting factor related to the surface reflection, the free surface multiple waves are converted into responses similar to the original seismic record data matrix P, that is, the free surface multiple waves achieve an order reduction process. The last term is the response of the interlayer multiple waves, which will be distributed at other locations in the focal domain. That is, in the focal domain, the primary waves are distributed at the focus point at the zero position, the free surface multiple waves show the form of the original seismic record, and the interlayer multiple waves are distributed at other locations.

[0092] It should also be noted that in traditional SRME, according to the feedback model, the relationship between the primary wave and the multiple waves can be expressed in the frequency domain using a data matrix as follows:

[0093] P=ΔP+M=ΔP+ΔPAP

[0094] In conventional SRME, interlayer multiple waves are regarded as "primary waves", and the irregular multiple waves described in the present invention are not applicable. In this embodiment, the relationship between the primary waves and the multiple waves in conventional SRME is corrected, and the mutual influence between the free surface multiple waves and the interlayer multiple waves is considered, so that the irregular multiple waves described in the present invention can be applied.

[0095] The above-mentioned irregular multiple wave separation system in this embodiment is based on focusing transformation, which breaks through the assumption that multiple waves are regular interference in traditional methods. By analyzing the energy distribution of different types of reflected waves in the focusing domain, high-resolution τ-p transformation and regularized non-steady-state regression adaptive matching filtering are used to simultaneously realize the separation of primary waves, free surface multiple waves and interlayer multiple waves. The accurately separated free surface multiple waves and interlayer multiple waves can lay a good data foundation for subsequent multiple wave suppression and imaging applications.

[0096] The irregular multiple wave separation method and system of the present invention are described in detail below with reference to a specific embodiment.

[0097] See also Figure 2aThe velocity field is calculated. The velocity field has 101 grid points horizontally and 101 grid points vertically, with a grid spacing of 10m. The Ricker wavelet with a main frequency of 30Hz is used for acoustic wave finite difference forward modeling. The forward modeling records a total of 51 shots with a shot spacing of 10m; 51 channels are received for each shot with a channel spacing of 20m; the source is located 10m below the surface, the total recording time is 0.8s, and the time sampling rate is 4ms. The figure shows the propagation paths of different types of reflection waves, among which 1 and 2 are primary waves, 4 and 7 are interlayer multiple waves, and the rest are free surface multiple waves.

[0098] Attached Figure 2b A simulated seismic cannon recording is shown. Figure 2b In the figure, we can see irregular multiple waves that contain both complex free surface multiple waves and interlayer multiple waves. In this embodiment, the estimated primary wave obtained by the forward modeling of the Born approximation method is used to construct the focusing operator. When performing irregular multiple wave separation, the shot record is firstly subjected to positive focusing transformation. The response in the focusing domain is shown in the attached figure. Figure 2c As shown by Figure 2c It can be seen that the primary wave is focused around the zero-position focus point, the free surface multiple waves in the focus domain are similar to the original seismic record data, and the interlayer multiple waves are distributed in other positions. In order to accurately separate the primary wave, the data in the focus domain (see Appendix Figure 2c ) to perform τ-p transformation, the result is as shown in the attached figure. Figure 2d As shown in the figure, it can be seen that the response distribution of the primary wave energy is on a straight line near τ = 0. Then, the above-mentioned irregular multiple wave separation method and system of the embodiment of the present invention can be used to achieve the separation of the primary wave, free surface multiple waves and interlayer multiple waves. The separation results of different types of reflected waves are shown in Figure 3a-3c As shown in the figure, the free surface multiple waves and interlayer multiple waves in the irregular multiple waves are effectively separated, and the primary waves are also effectively preserved.

[0099] The above embodiments are used to explain the present invention rather than to limit the present invention. Any modification and change made to the present invention within the spirit of the present invention and the protection scope of the claims shall fall within the protection scope of the present invention.

Claims

1. A method for separating irregular multiple waves, characterized in that: The following steps are involved: Data acquisition steps: obtaining seismic record data; Focusing operator construction steps: The positive focusing transformation and the inverse focusing transformation are defined as: Q=FP (1) P=GQ (2) Where Q is the total response of the seismic record data in the focus domain, F is the positive focusing operator, and G is the anti-focusing operator; Velocity analysis is performed on seismic data to obtain the velocity field, and the estimated primary wave of the seismic data is obtained based on the velocity field forward modeling. To estimate a wave is the anti-focusing operator G, is the positive focusing operator; Focusing transformation step: according to the relationship between the reflected waves of the seismic recording data, the seismic recording data is subjected to positive focusing transformation to obtain the focusing domain data; τ-p transformation step: Use high-resolution τ-p transformation to transfer the focus domain data to the τ-p domain, where the primary wave energy is converted into a straight line at time zero; The primary wave separation step is: removing the energy of the straight line in the τ-p domain, transforming the removed straight line energy back to the focus domain with high resolution τ-p, and then focusing and transforming it back to the original time-space domain of the seismic record data to obtain the separated primary wave; Free surface multiple wave separation step: after the linear energy is cut off, the high-resolution τ-p of other energies in the τ-p domain is inversely transformed back to the focusing domain to obtain the multiple wave response in the focusing domain. According to the similarity between the free surface multiple waves in the focusing domain and the seismic recording data in the original time-space domain, the regularized non-steady-state regression adaptive matched filtering method is used to extract the free surface multiple waves in the focusing domain, and the inverse focusing transformation is performed back to the original time-space domain of the seismic recording data to obtain the separated free surface multiple waves. Interlayer multiple wave separation step: subtract the separated primary wave and the separated free surface multiple waves from the seismic record data to obtain the interlayer multiple waves.

2. The irregular multiple wave separation method according to claim 1, characterized in that: In the focusing transformation step, the relationship between the reflected waves of the seismic recording data is the relationship between the primary wave and the multiple waves of the seismic recording data in the frequency domain, which can be expressed as: P=ΔP+M S +M I =ΔP+ΔPAP+M I (3) Where P is the seismic record data matrix containing the primary wave, free surface multiple waves and interlayer multiple waves, ΔP is the primary wave, M S is the free surface multiple wave, M I is the interlayer multiple wave, and A is the weighting factor related to the surface reflection; Combining formula (1) and formula (3), we have: Where I is the focal point of the primary wave focused around the source point.

3. The irregular multiple wave separation method according to claim 1 or 2, characterized in that: In the focusing operator construction step: the estimated primary wave of the seismic data is obtained by using the Born approximation forward modeling method based on the velocity field.

4. An irregular multiple wave separation system, characterized in that: include: Data acquisition steps: obtaining seismic record data; A focusing operator construction module, used to construct positive focusing operators and anti-focusing operators; A focusing transformation module performs positive focusing transformation on the seismic recording data according to the relationship between the reflected waves of the seismic recording data to obtain the focusing domain data; A τ-p transformation module, used to transfer the focus domain data to the τ-p domain using a high-resolution τ-p transformation, in which the primary wave energy is converted into a straight line at time zero; The primary wave separation module is used to remove the energy of the straight line in the τ-p domain, transform the removed straight line energy back to the focus domain with high resolution τ-p, and then focus and transform it back to the original time-space domain of the seismic recording data to obtain the separated primary wave; The free surface multiple wave separation module is used to perform a high-resolution τ-p inverse transformation of other energies in the τ-p domain after the linear energy is cut off back to the focusing domain to obtain the multiple wave response in the focusing domain. According to the similarity between the free surface multiple waves in the focusing domain and the seismic recording data in the original time-space domain, the regularized non-steady-state regression adaptive matched filtering method is used to extract the free surface multiple waves in the focusing domain, and the inverse focusing transformation is performed back to the original time-space domain of the seismic recording data to obtain the separated free surface multiple waves. The interlayer multiple wave separation module is used to subtract the separated primary wave and the separated free surface multiple waves from the seismic record data to obtain the interlayer multiple waves.

5. The irregular multiple wave separation system according to claim 4, characterized in that: In the focusing operator construction module, the specific steps for constructing the positive focusing operator and the anti-focusing operator are: The positive focusing transformation and the negative focusing transformation are defined as: Q=FP (1) P=GQ (2) Where Q is the total response of the seismic record data in the focus domain, F is the positive focusing operator, and G is the anti-focusing operator; Velocity analysis is performed on seismic data to obtain the velocity field, and the estimated primary wave of the seismic data is obtained based on the velocity field forward modeling. To estimate a wave is the anti-focusing operator G, is the positive focusing operator.

6. The irregular multiple wave separation system according to claim 5, characterized in that: The estimated primary wave of seismic data is obtained by using the Born approximation forward modeling method based on the velocity field.

7. The irregular multiple wave separation system according to claim 5, characterized in that: In the focus transformation module, the relationship between the reflected waves of the seismic recording data is the relationship between the primary wave and the multiple waves of the seismic recording data in the frequency domain, which is expressed as: P=ΔP+M S +M I =ΔP+ΔPAP+M I (3) Where P is the seismic record data matrix containing the primary wave, free surface multiple waves and interlayer multiple waves, ΔP is the primary wave, M S is the free surface multiple wave, M I is the interlayer multiple wave, and A is the weighting factor related to the surface reflection; Combining formula (1) and formula (3), we have: Where I is the focal point of the primary wave focused around the source point.

Images