Prestack seismic 3D multi-channel inversion method based on spatial structural constraints

By adopting a three-dimensional multi-channel inversion method based on spatial structural constraints, the problem of low efficiency in three-dimensional pre-stack seismic inversion is solved, and efficient three-dimensional seismic data inversion is achieved, which improves the stability and fidelity of reservoir quantitative prediction.

CN114755724BActive Publication Date: 2024-11-22CHINA UNIV OF PETROLEUM (EAST CHINA)
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210345379.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-03-31
Publication Date
2024-11-22
Estimated Expiration
2042-03-31

AI Technical Summary

Technical Problem

Existing structurally constrained inversion methods are difficult to apply to 3D seismic multichannel inversion, especially 3D pre-stack seismic inversion, resulting in low inversion efficiency and affecting the stability and fidelity of quantitative reservoir prediction.

Method used

A three-dimensional multi-channel inversion method based on spatial tectonic constraints is adopted. A spatial tectonic operator is constructed by estimating the dip angle of the strata, a three-dimensional matrix operation rule is defined, and the objective functional is introduced by using the Hadamard product operator to improve the strategy of introducing the tectonic operator and improve the inversion efficiency.

Benefits of technology

It improves the stability and fidelity of 3D pre-stack seismic inversion, enhances inversion efficiency, and enables simultaneous multi-channel inversion of large-scale 3D seismic data.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114755724B_ABST
    Figure CN114755724B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of oil and gas exploration and development, and specifically discloses a pre-stack seismic three-dimensional multi-channel inversion method based on spatial structural constraints. The method comprises: step S1, estimating the formation dip based on three-dimensional seismic data, and constructing a spatial structural operator; step S2, defining a three-dimensional matrix operation rule; step S3, constructing a pre-stack seismic multi-channel inversion target functional with spatial structural constraints. The pre-stack seismic multi-channel inversion method based on spatial structural constraints proposed by the present invention develops a construction method of a spatial structural operator, improves a strategy for introducing a target functional into a structural operator, and greatly improves the inversion efficiency of three-dimensional pre-stack seismic.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of oil and gas exploration and development, and in particular to the field of pre-stack seismic multi-channel inversion technology. Background Art

[0002] Prestack seismic inversion is an effective way to obtain the elasticity, physical properties, rock properties and fluid properties of underground media, and is a key technology for achieving quantitative reservoir prediction. The stability, reliability and fidelity of prestack seismic inversion of seismic amplitude are key factors affecting the quantitative interpretation of reservoirs. Multi-channel seismic inversion considers the intrinsic connection between adjacent seismic traces, which can preserve the structural characteristics of seismic inversion and improve the stability and fidelity of inversion. Therefore, conducting prestack seismic multi-channel inversion research is of great significance for quantitative reservoir prediction.

[0003] There are many prestack seismic multi-channel inversion methods now. Some of them are based on the horizontal layered assumption of the medium to achieve total variation regularized multi-channel inversion, etc. Some use seismic reflection characteristics as lateral constraints to achieve multi-channel inversion, and some consider the formation dip characteristics and achieve prestack seismic inversion based on structural feature constraints. Among them, the structural constraint multi-channel inversion method has been widely studied in recent years because of its good application effect in complex structural areas such as high dip and faults. There are two key steps in structural constraint inversion: one is the estimation of formation dip; the other is the construction of structural constraint operators. However, the existing structural constraint inversion methods are limited by the inversion efficiency and are difficult to apply to 3D seismic multi-channel inversion, especially 3D prestack seismic inversion. Summary of the invention

[0004] The present invention improves the method of introducing structural constraint operators for three-dimensional seismic data, derives a fast optimization algorithm for three-dimensional multi-channel inversion, and develops a pre-stack seismic three-dimensional multi-channel inversion method based on spatial structural constraints.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: a pre-stack seismic three-dimensional multi-channel inversion method based on spatial structural constraints, comprising:

[0006] Step S1, estimating the formation dip based on three-dimensional seismic data and constructing a spatial structural operator;

[0007] Step S2, defining three-dimensional matrix operation rules;

[0008] Step S3, constructing a pre-stack seismic multi-channel inversion target functional constrained by spatial structure.

[0009] Preferably, the step S1 comprises:

[0010] For any point O on a stratum, three-dimensional coordinate axes x, y and z are established, and the stratum dip angles of the xoz plane and the yoz plane are estimated to be α and β respectively. The rotation coordinate axis x is constructed according to α and β. r ,yr and z r ;

[0011] The spatial structural operator R parallel to the formation dip parl It is expressed as:

[0012] R parl =cosαR x +cosβR y +(sinα+sinβ)R z (1),

[0013] In formula (1), R x , R y and R z are the first-order difference operators along the x, y and z directions respectively;

[0014] Construct a spatial structural operator R perpendicular to the formation dip perp :

[0015] Let points A, B and C be the rotation axes x respectively. r ,y r and z r A point on the circle, and the coordinates of A and B are known to be A(1,0,tanα), B(0,1,tanβ), OA⊥OC and OB⊥OC, and the coordinates of point C are Then we have:

[0016]

[0017] The coordinates of point C are expressed as According to the triangular cosine formula, the original coordinate axes (x, y, z) and z are calculated respectively. r The cosine of the angle:

[0018]

[0019] The spatial structural operator R perpendicular to the formation dip perp Denoted as:

[0020]

[0021] For 3D seismic data, R parl and R perp Respectively expressed as:

[0022]

[0023]

[0024] In formula (5) and formula (6), the coefficient matrix Q1 is Q2, Q3, and Q are:

[0025]

[0026] Preferably, step S2 comprises:

[0027] Definition of operation rule 1: Three-dimensional matrix A (A∈R m×n×h ) and the two-dimensional matrix B(B∈R s×t ) along the x direction is expressed as:

[0028]

[0029] In formula (7), [] ijk represents the element at position (i, j, k) in the three-dimensional matrix, p is the subscript variable in matrix multiplication, m, n, h represent the number of elements in the row, column, and height of the three-dimensional matrix A, and s and t correspond to the rows and columns of the two-dimensional matrix B;

[0030] Definition of operation rule 2: Three-dimensional matrix A (A∈R m×n×h ) and the two-dimensional matrix B(B∈R s×t ) along the y direction is expressed as:

[0031]

[0032] In formula (7) and formula (8), the operator “*”, Respectively represent the multiplication of a two-dimensional matrix and a three-dimensional matrix along the x direction and the y direction.

[0033] Preferably, step S3 comprises:

[0034] Assume that the forward model of prestack seismic multi-channel inversion is:

[0035]

[0036] In formula (9), θ i is the incident angle, i=1,…,l, The angle is θ i , S represents the three-dimensional seismic data after the three-dimensional seismic data of l angles are arranged, G represents the forward operator, and m represents the three-dimensional parameters to be inverted;

[0037] According to the spatial construction operator derived in step S1 and the three-dimensional matrix operation rule defined in step S2, the spatial construction operator R parallel to the formation dip is converted into parl and the spatial structural operator R perpendicular to the formation dip perp Introducing the target functional of multi-channel inversion, the target functional based on spatial structural constraints is expressed as:

[0038]

[0039] In equation (10), is the Hadamard product operator, which means the multiplication of corresponding elements of two matrices of the same size. is the F norm, which represents the sum of squares of all elements in the matrix;

[0040] Solving equation (10) completes the prestack seismic 3D multi-channel inversion with spatial structural constraints.

[0041] Compared with the prior art, the prestack seismic three-dimensional multi-channel inversion method based on spatial structural constraints proposed in the present invention takes into account the spatial structural constraints, develops a construction method of spatial structural operators based on three-dimensional seismic data volumes, and improves stability and fidelity; improves the structural operator, and adopts the Hadamard product operator instead of the traditional dimensionality reduction strategy to introduce the structural operator into the target functional, which greatly improves the inversion efficiency of three-dimensional prestack seismic. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 A flow chart of the pre-stack seismic three-dimensional multi-channel inversion method based on spatial structural constraints provided by the present invention.

[0043] Figure 2 A schematic diagram of constructing a spatial structural operator in an embodiment of a pre-stack seismic three-dimensional multi-channel inversion method based on spatial structural constraints provided by the present invention.

[0044] Figure 3a A schematic diagram of synthesizing three-dimensional seismic data by multiplying a wavelet matrix and a three-dimensional reflection coefficient along the x direction in an embodiment of a pre-stack seismic three-dimensional multi-channel inversion method based on spatial structural constraints provided by the present invention.

[0045] Figure 3b A schematic diagram of synthesizing three-dimensional seismic data by multiplying a wavelet matrix and a three-dimensional reflection coefficient along the y direction in an embodiment of a pre-stack seismic three-dimensional multi-channel inversion method based on spatial structural constraints provided by the present invention.

[0046] Figure 4a The figure is a schematic diagram of longitudinal wave impedance obtained by inverting the three-dimensional actual seismic data of a certain work area using the method of the present invention.

[0047] Figure 4b The figure is a schematic diagram of shear wave impedance obtained by inverting the three-dimensional actual seismic data of a certain work area using the method of the present invention. DETAILED DESCRIPTION

[0048] In order to facilitate a further understanding of the method of the present invention and the effects achieved, the present invention is described in detail below with reference to the accompanying drawings. However, the accompanying drawings are only provided for reference and illustration and are not intended to limit the present invention.

[0049] The forward model of pre-stack seismic single-channel inversion is (taking the inversion of P-wave impedance, S-wave impedance and density as an example):

[0050]

[0051] In formula (A1), c1, c2 and c3 represent the three coefficients related to the incident angle in the reflection coefficient equation, S i represents the seismic data of the ith channel, θ j (j=1,…,h) represents the incident angle of prestack seismic data, W is the angle-dependent seismic wavelet kernel matrix, and L is the first-order difference matrix. G is the forward operator, m i The longitudinal wave impedance I is given by the logarithmic form P , Shear wave impedance I S and density I D The parameters to be inverted.

[0052] The existing pre-stack seismic multi-channel inversion method usually derives the forward model of multi-channel inversion according to formula (A1), which can usually be expressed as:

[0053]

[0054] In formula (A2), represents a column vector composed of multiple seismic traces arranged one by one, is the convolution operator for multi-channel seismic inversion, are the parameters to be inverted for multiple seismic traces. It can be seen that It is a large matrix that grows geometrically with the increase in the number of inverted seismic traces, which seriously affects the efficiency of prestack seismic inversion and also restricts the application of existing technologies in three-dimensional prestack seismic inversion.

[0055] like Figure 1 As shown, in one embodiment of the pre-stack seismic three-dimensional multi-channel inversion method based on spatial structural constraints provided by the present invention, the method may include the following steps:

[0056] S1, estimate the formation dip based on 3D seismic data and construct a spatial structural operator;

[0057] S2, define the three-dimensional matrix operation rules;

[0058] S3, constructing the target functional of prestack seismic multi-channel inversion with spatial structural constraints.

[0059] S1, estimate the formation dip based on 3D seismic data and construct a spatial structural operator:

[0060] Establish three-dimensional coordinate axes (x, y, z) for any point O on a certain stratum (such as Figure 2According to the method of estimating the formation dip angle of the two-dimensional profile, the formation dip angles α and β of the xoz plane and yoz plane are estimated respectively, and then the rotation coordinate axis (x r ,y r and z r ).

[0061] As with the existing structural constraint inversion method, the spatial structural operators parallel and perpendicular to the formation dip are estimated. The spatial structural operator parallel to the formation dip is R parl It can be simply expressed as:

[0062] R parl =cosαR x +cosβR y +(sinα+sinβ)R z (1),

[0063] In formula (1), R x , R y and R z are the first-order difference operators along the x, y and z directions respectively.

[0064] Construct a spatial structural operator R perpendicular to the formation dip perp It can be obtained by deduction. Assume that points A, B and C are the rotation axis x r ,y r and z r A point on the circle, and the coordinates of A and B are known to be A(1,0,tanα) and B(0,1,tanβ). Since points A, B, and C are on three coordinate axes, OA⊥OC and OB⊥OC, and let the coordinates of point C be Then we have:

[0065]

[0066] The coordinates of point C can be expressed as According to the triangular cosine formula, the original coordinate axes (x, y, z) and z can be calculated respectively. r The cosines of the angles are:

[0067]

[0068] Therefore, the spatial structural operator perpendicular to the formation dip can be written as:

[0069]

[0070] For 3D seismic data, the spatial construction operator R parl and R perp Can be expressed as follows:

[0071]

[0072]

[0073] In formula (5) and formula (6), the coefficient matrix Q1 is Q2, Q3, and Q are:

[0074]

[0075] S2, define the three-dimensional matrix operation rules:

[0076] In order to facilitate the construction of the target functional and the derivation of the optimization algorithm, two operation rules directly oriented to three-dimensional matrix operations are firstly defined.

[0077] Definition 1: A three-dimensional matrix A(A∈R m×n×h ) and the two-dimensional matrix B(B∈R s×t ) along the x direction can be expressed as (m, n, h represent the number of elements in the rows, columns, and height of the three-dimensional matrix A, and s and t correspond to the rows and columns of the two-dimensional matrix B):

[0078]

[0079] In formula (7), [] ijk Represents the element at position (i,j,k) in the three-dimensional matrix, and p is the subscript variable in matrix multiplication. The multiplication of matrices A and B is divided into left multiplication and right multiplication.

[0080] Definition 2: Three-dimensional matrix A(A∈R m×n×h ) and the two-dimensional matrix B(B∈R s×t ) along the y direction can be expressed as:

[0081]

[0082] In formula (7) and formula (8), the operator “*”, Respectively represent the multiplication of a two-dimensional matrix and a three-dimensional matrix along the x direction and the y direction. The specific operation rules are shown in the schematic diagram Figure 3a and Figure 3b According to the definitions of the two matrix operators, we can notice that Right now Figure 3a and Figure 3b The two methods of synthesizing seismic records shown are equivalent, which is consistent with the actual situation of synthesizing three-dimensional seismic data, that is, whether synthesizing three-dimensional seismic data profile by profile along the x or y direction, the results obtained are the same, which proves the rationality of the matrix operation rules defined in the present invention.

[0083] S3, constructing the target functional of pre-stack seismic multi-channel inversion with spatial structural constraints:

[0084] The multi-channel forward model adopted by the existing method is easy to generate a large matrix, which is not conducive to computer storage and calculation. In order to avoid this problem, the present invention adopts a new forward model based on formula (A), and its expression is:

[0085]

[0086] In formula (9), θ i (i=1,…,l) is the incident angle, The angle is θ i , S represents the seismic data after the seismic data of l angles are arranged, G represents the forward operator, and m represents the three-dimensional parameters to be inverted.

[0087] According to the spatial construction operator derived in step S1 and the three-dimensional matrix operation rule defined in step S2, the present invention uses the Hadamard product operator to introduce the spatial construction constraint operator into the target functional of multi-channel inversion. The target functional based on the spatial construction constraint can be expressed as:

[0088]

[0089] In equation (10), is the Hadamard product operator, which means the multiplication of corresponding elements of two matrices of the same size. is the F norm, which represents the sum of the squares of all elements in the matrix. Solving equation (10) can complete the prestack seismic 3D multi-channel inversion with spatial structural constraints.

[0090] Figure 4a , Figure 4b They are the P-wave impedance and S-wave impedance inverted by the method of the present invention based on the actual 3D seismic data of a certain work area. The pre-stack multi-channel simultaneous inversion of large-scale 3D seismic data is realized, proving the rationality and effectiveness of the method, and providing a technical reference for the research in the fields of 3D pre-stack multi-channel inversion and 3D multi-channel deconvolution.

[0091] The prestack seismic multi-channel inversion method based on spatial structural constraints proposed in the present invention develops a construction method of spatial structural operators, improves the strategy of introducing target functionals into structural operators, and greatly improves the inversion efficiency of three-dimensional prestack seismic.

[0092] The above embodiments are only preferred embodiments for fully illustrating the present invention, and the protection scope of the present invention is not limited thereto. Any equivalent substitution or change made by a person skilled in the art based on the present invention is within the protection scope of the present invention.

Claims

1. A pre-stack seismic 3D multi-channel inversion method based on spatial structural constraints, characterized in that: include: Step S1, estimating the formation dip based on three-dimensional seismic data and constructing a spatial structural operator; Step S2, defining three-dimensional matrix operation rules; Step S3, constructing a pre-stack seismic multi-channel inversion target functional constrained by spatial structure; The step S1 comprises: For any point O on a stratum, three-dimensional coordinate axes x, y and z are established, and the stratum dip angles of the xoz plane and the yoz plane are estimated to be α and β respectively. The rotation coordinate axis x is constructed according to α and β. r ,y r and z r ; The spatial structural operator R parallel to the formation dip parl It is expressed as: R parl =cosαR x +cosβR y +(sinα+sinβ)R z (1), In formula (1), R x , R y and R z are the first-order difference operators along the x, y and z directions respectively; Construct a spatial structural operator R perpendicular to the formation dip perp : Let points A, B and C be the rotation axes x respectively. r ,y r and z r A point on the circle, and the coordinates of A and B are known to be A(1,0,tanα), B(0,1,tanβ), OA⊥OC and OB⊥OC, and the coordinates of point C are Then we have: The coordinates of point C are expressed as According to the triangular cosine formula, the original coordinate axes (x, y, z) and z are calculated respectively. r The cosine of the angle: The spatial structural operator R perpendicular to the formation dip perp Denoted as: For 3D seismic data, R parl and R perp Respectively expressed as: In formula (5) and formula (6), the coefficient matrix Q1 is Q2, Q3, and Q are: The step S2 comprises: Definition of operation rule 1: Three-dimensional matrix A (A∈R m×n×h ) and the two-dimensional matrix B(B∈R s×t ) along the x direction is expressed as: In formula (7), [] ijk represents the element at position (i, j, k) in the three-dimensional matrix, p is the subscript variable in matrix multiplication, m, n, h represent the number of elements in the row, column, and height of the three-dimensional matrix A, and s and t correspond to the rows and columns of the two-dimensional matrix B; Definition of operation rule 2: Three-dimensional matrix A(A∈R m×n×h ) and the two-dimensional matrix B(B∈R s×t ) along the y direction is expressed as: In equations (7) and (8), operators "*" and "*" represent the multiplication of a two-dimensional matrix and a three-dimensional matrix along the x direction and the y direction respectively; The step S3 comprises: Assume that the forward model of prestack seismic multi-channel inversion is: In formula (9), θ i is the incident angle, i=1,…,l, The angle is θ i , S represents the three-dimensional seismic data after the three-dimensional seismic data of l angles are arranged, G represents the forward operator, and m represents the three-dimensional parameters to be inverted; According to the spatial construction operator derived in step S1 and the three-dimensional matrix operation rule defined in step S2, the spatial construction operator R parallel to the formation dip is converted into parl and the spatial structural operator R perpendicular to the formation dip perp Introducing the target functional of multi-channel inversion, the target functional based on spatial structural constraints is expressed as: In equation (10), is the Hadamard product operator, which means the multiplication of corresponding elements of two matrices of the same size. is the F norm, which represents the sum of the squares of all elements in the matrix; Solving equation (10) completes the prestack seismic 3D multi-channel inversion with spatial structural constraints.