Fractional order total variation-based lacustrine facies carbonate rock pre-stack fracture prediction method

By constructing a pre-stack elastic parameter inversion model based on fractional order full variation, the problem of low Young's modulus and Poisson's ratio accuracy in the prior art is solved, and high-precision crack prediction is achieved.

CN118938312AActive Publication Date: 2024-11-12CHINA PETROLEUM & CHEMICAL CORP +1
View PDF 9 Cites 0 Cited by

Patent Information

Application Number
CN202310531952.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-05-09
Publication Date
2024-11-12
Estimated Expiration
2043-05-09

AI Technical Summary

Technical Problem

The accuracy of Young's modulus and Poisson's ratio in the existing pre-stack elastic parameter inversion method is low, resulting in poor crack prediction effect.

Method used

Using the method based on fractional order full variation, the forward model and objective function of pre-stack elastic parameter are constructed, and the parameter matrix is ​​updated through alternating direction multiplication method, gradient zero-take extreme value, and eigenvalue decomposition, to improve the accuracy of the inversion result and stability.

Benefits of technology

The accuracy of the inversion result of Young's modulus and Poisson's ratio is improved, and the accuracy and stability of crack prediction are improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118938312A_ABST
    Figure CN118938312A_ABST
Patent Text Reader

Abstract

The invention relates to the field of geophysical inversion and oil and gas reservoir prediction, in particular to a lacustrine facies carbonate rock prestack fracture prediction method based on fractional order total variation. The method comprises the following steps: preprocessing seismic data to obtain an initial model of a to-be-inverted parameter; constructing a pre-stack elastic parameter forward modeling model and an objective function based on fractional order total variation; carrying out inversion updating on the logarithmic matrix by combining an alternating direction multiplier method, taking an extreme value when the gradient is zero and carrying out eigenvalue decomposition; the Young modulus, the Poisson's ratio and the density are obtained according to the relation between the Young modulus, the Poisson's ratio and the density and the elastic three-parameter logarithmic matrix Li + 1; and slicing the inversion results of the Young modulus and the Poisson's ratio according to horizon to obtain a crack prediction result. According to the method, a sparse regularization item is established by adopting fractional order total variation, and an objective function of pre-stack elastic three-parameter seismic inversion is constructed, so that the precision and the stability of an inversion result are improved, and the crack prediction precision is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the field of geophysical inversion and oil and gas reservoir prediction, and in particular to a lacustrine carbonate prestack fracture prediction method based on fractional-order total variation. Background Art

[0002] Seismic inversion is an important means of oil and gas reservoir prediction. It is a process of establishing a forward model based on the mathematical relationship between the detected seismic records and the physical quantities to be inverted, and using optimization methods to solve the forward model for the best estimate. Seismic inversion technology based on sparse regularization is an important method of seismic inversion. It introduces sparse regularization constraints into the forward model and successfully improves the resolution and robustness of the inversion results by using sparse information.

[0003] As the difficulty of obtaining conventional oil and gas resources increases, global oil and gas exploration has entered an era of equal development of conventional and unconventional oil and gas. Rock elastic parameters such as Young's modulus and Poisson's ratio are proportional to the porosity of Hunan carbonate reservoirs, so prestack elastic parameter inversion is an effective way to predict fractures from seismic data. For example, in the Bayesian inversion method, Alemie and Sacchi introduced multivariate Cauchy distribution into Bayesian inversion. This prior distribution can merge the correlation between model parameters; Huber distribution, t distribution, etc. have also been introduced into Bayesian inversion. The above prior distributions all assume that the reflection coefficient is a random variable, and its distribution satisfies a specific distribution characterized by a certain parameter. When the distribution of the reflection coefficient does not satisfy the specific distribution, the estimation result of the reflection coefficient may be erroneous, and the accuracy and resolution of prestack seismic inversion of elastic parameters are low.

[0004] CN114428301A discloses a pre-stack elastic parameter inversion method, which includes obtaining pre-stack seismic data; obtaining reflection coefficients of elastic parameters; constructing a forward modeling matrix of pre-stack seismic data; obtaining a likelihood function of seismic data based on the forward modeling matrix; obtaining a target function of elastic parameter inversion based on the likelihood function and prior information; converting the likelihood function into a maximized marginal likelihood function based on a maximum likelihood estimation method; solving the maximized marginal likelihood function to obtain an estimation factor; inverting the elastic parameters based on the estimation factor and the target function, and integrating the prior information of the elastic parameters into the inversion by introducing automatic correlation discrimination prior information.

[0005] CN110857997A discloses a step-by-step prestack elastic parameter inversion method based on lateral constraints, the method comprising the following steps:

[0006] 1) Based on the PP wave and PS wave angle gathers that are consistent with the event axis matching, the PP wave and PS wave angle domain reflection coefficients are obtained;

[0007] 2) Use the Stewart reflection coefficient formula to perform multi-wave joint inversion, and use the Kalman filter algorithm to constrain the inversion equation laterally to obtain the longitudinal wave velocity change rate and the shear wave velocity change rate;

[0008] 3) Based on the longitudinal wave velocity change rate and the shear wave velocity change rate, the Aki-Richards three-parameter reflection coefficient formula is substituted, and the Kalman filter algorithm is used to perform lateral constraints on the inversion equation to obtain the density change rate;

[0009] 4) Perform low-frequency compensation on the inversion results obtained in step 2) and step 3) to obtain the absolute values ​​of the final pre-stack elastic parameters.

[0010] However, the Young's modulus and Poisson's ratio obtained by the existing pre-stack elastic parameter inversion method have low accuracy, resulting in poor crack prediction effect. Therefore, an inversion method is needed to overcome the above problems. Summary of the invention

[0011] The main purpose of the present invention is to provide a method for predicting prestack fractures in lacustrine carbonate rocks based on fractional-order total variation. The method of the present invention can improve the accuracy and stability of the inversion results of Young's modulus and Poisson's ratio, and further obtain high-precision fracture prediction results through the inversion results of Young's modulus and Poisson's ratio with high accuracy, thus overcoming the shortcomings of the prior art.

[0012] To achieve the above object, the present invention adopts the following technical solutions:

[0013] The present invention provides a method for predicting prestack fractures in lacustrine carbonate rocks based on fractional-order total variation, the method comprising the following steps:

[0014] Step 1: Preprocess the seismic data to obtain the initial model of the parameters to be inverted;

[0015] Step 2: According to the processed seismic data, Young's modulus, Poisson's ratio and density are combined into a logarithmic matrix L0, and a prestack elastic parameter forward model and objective function based on fractional total variation are constructed;

[0016] Step 3: Combine the alternating direction multiplication method, the extreme value of the gradient being zero, and the eigenvalue decomposition to inversely update the logarithmic matrix to obtain the updated logarithmic matrix L i ;

[0017] Step 4: Determine whether the values ​​before and after the update satisfy ||L i+1 -L i ||2 / ||L i ||2>tol, if yes, return to step 3 to loop; if no, according to Young's modulus, Poisson's ratio and density and elastic three-parameter logarithmic matrix L i+1The relationship between Young's modulus, Poisson's ratio and density is obtained;

[0018] Step 5: Slice the inversion results of Young's modulus and Poisson's ratio by layer to obtain the fracture prediction results.

[0019] Furthermore, in step 1, the seismic data includes seismic records, wavelet data and well logging data; the layer information is extracted from the seismic records and the well logging data is interpolated and filtered to obtain the initial model of Young's modulus, Poisson's ratio and density.

[0020] Furthermore, in step 2, a prestack elastic parameter forward model and an objective function based on fractional-order total variation are constructed, which specifically includes the following steps:

[0021] The vertical and horizontal fractional-order difference vectors are used to construct the vertical and horizontal fractional-order difference matrix and the vertical first-order difference matrix.

[0022]

[0023]

[0024] D=d*E (3i-1)×(3i-1)

[0025] Among them, * represents the convolution operation, represents the transverse fractional difference matrix, represents the longitudinal fractional-order difference matrix, D represents the longitudinal first-order difference matrix, d=[-11] represents the longitudinal first-order difference vector, i represents the total number of rows of a single parameter to be inverted, and j represents the total number of columns of a single parameter to be inverted;

[0026] Construct a forward model based on fractional-order total variation prestack elastic three-parameter:

[0027]

[0028] Among them, ||||1 represents the L1 norm, ||||2 represents the L2 norm, * represents the convolution operation; S is the seismic record matrix, W is the wavelet matrix, and G represents the coefficient matrix of the three elastic parameters to be inverted;

[0029] Based on the forward model, the Lagrange multiplier term R is introduced. y , R x and the dual term C y , C x , the objective function of prestack elastic three-parameter seismic inversion based on fractional-order total variation is obtained:

[0030]

[0031] Among them, η is the weight coefficient of the dual term.

[0032] Furthermore, the relationship between the elastic three-parameter logarithmic matrix and Young's modulus, Poisson's ratio and density is expressed by the following formula:

[0033] L=[lnE lnσ lnρ] T

[0034] Among them, L represents the elastic three-parameter logarithmic matrix, E represents Young's modulus, σ represents Poisson's ratio, ρ represents density, ln represents logarithmic operation, and T represents the transpose of the matrix;

[0035] Furthermore, the vertical and horizontal fractional-order difference vectors are expressed by the following formula:

[0036]

[0037] Among them, ψ a (k) = (-1) k Γ(a+1) / [Γ(k+1)Γ(a-k+1)];

[0038] Furthermore, S is composed of earthquake records at three angles: near, middle and far. θ1 , S θ2 and S θ3 The composition is expressed by the following formula:

[0039] S=[S θ1 ,S θ2 ,S θ3 ] T

[0040] W is expressed by the following formula:

[0041]

[0042] where w q represents the qth data of the seismic wavelet, where q represents the length of the wavelet;

[0043] G is expressed by the following formula:

[0044]

[0045] Among them, G E The coefficient matrix representing Young's modulus, G σ The coefficient matrix representing Poisson's ratio, G ρ The coefficient matrix representing the density.

[0046] Furthermore, the specific steps of step 3 are:

[0047] The initial elastic three-parameter logarithmic matrix L0 is used as the initial model L for prestack elastic three-parameter seismic inversion of fractional total variation 1 , and introduce the initial value R of the Lagrange multiplierx =R y = 0 and its dual term C x =C y =0;

[0048] The elastic three-parameter logarithmic matrix L is updated using the alternating direction multiplier algorithm, the criterion of taking the extreme value of the gradient when it is zero, and the eigenvalue decomposition. i+1 :

[0049]

[0050] Where U and V are square matrices and The orthogonal matrix obtained after eigenvalue decomposition:

[0051]

[0052] Where SVD represents eigenvalue decomposition, Λ and Σ are the eigenvalue diagonal matrices obtained after decomposition;

[0053] The update formula is as follows:

[0054]

[0055] Where a represents the number of eigenvalues ​​in the eigenvalue diagonal matrix Λ, b represents the number of eigenvalues ​​in the eigenvalue diagonal matrix Σ,

[0056] Update the Lagrange multiplier term R according to the alternating direction multiplier algorithm and the soft threshold shrinkage algorithm x i+1 , R y i+1 ;

[0057] Update the dual term according to the alternating direction multiplier algorithm and Fermat's lemma

[0058] Furthermore, the updated Lagrange multiplier term R x i+1 , R y i+1 It is expressed by the following formula:

[0059]

[0060] Furthermore, after updating the dual term This is shown in the following formula:

[0061]

[0062] Furthermore, in step 4, the parameters to be inverted and the elastic three-parameter logarithmic matrix Li+1 The relationship is expressed by the following formula:

[0063]

[0064] Compared with the prior art, the present invention has the following advantages:

[0065] The present invention adopts fractional-order total variation to establish a sparse regularization term and construct an objective function of prestack elastic three-parameter seismic inversion, thereby improving the accuracy and stability of the inversion result and thus improving the fracture prediction accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 This is a flowchart of the method described in an embodiment of the present invention.

[0067] Figure 2 The seismic profile containing noise according to the method of the embodiment of the present invention: (1) represents S θ1 , (2) represents S θ2 , (3) represents S θ3 .

[0068] Figure 3 Schematic cross-sectional views of the initial models of Young's modulus, Poisson's ratio and density in the method according to an embodiment of the present invention: (1) represents the initial model of Young's modulus, (2) represents the initial model of Poisson's ratio, and (3) represents the initial model of density;

[0069] Figure 4 The cross-sectional diagrams are schematic diagrams of the inversion results of Young's modulus, Poisson's ratio and density in the method described in the embodiment of the present invention: (1) represents the inversion result of Young's modulus, (2) represents the inversion result of Poisson's ratio, and (3) represents the inversion result of density;

[0070] Figure 5 The result of crack prediction along the slice of the layer by Young's modulus in the method described in the embodiment of the present invention;

[0071] Figure 6 This is the result of Poisson's ratio crack prediction along the layer slice in the method described in the embodiment of the present invention. DETAILED DESCRIPTION

[0072] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0073] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations and / or combinations thereof.

[0074] In order to enable those skilled in the art to more clearly understand the technical solution of the present invention, the technical solution of the present invention will be described in detail below in conjunction with specific embodiments.

[0075] Example 1

[0076] Step 1: Preprocess the lacustrine carbonate seismic data and obtain the initial model of the parameters to be inverted:

[0077] Step 1.1: Collect seismic records S, wavelet data w and well logging data;

[0078] Step 1.2: Extract the layer information through the seismic records and perform interpolation filtering on the well logging data to obtain the initial model of Young's modulus, Poisson's ratio and density.

[0079] Step 2: Combining Young's modulus, Poisson's ratio and density into an elastic three-parameter matrix according to the processed seismic data, and calculating the logarithmic matrix L0 of the elastic three-parameter matrix, calculating the fractional-order difference matrix to construct a pre-stack elastic parameter forward model and objective function based on fractional-order total variation, the steps are specifically as follows:

[0080] Step 2.1: Based on the initial model of Young's modulus, Poisson's ratio and density, an elastic three-parameter matrix is ​​combined, and the initial elastic three-parameter logarithmic matrix L0 is obtained according to the recursive relationship between the elastic three-parameter matrix and its logarithm; the relationship between the elastic three-parameter logarithmic matrix and Young's modulus, Poisson's ratio and density is shown in Formula 1:

[0081] L=[lnE lnσ lnρ] T Formula (1)

[0082] Among them, L represents the elastic three-parameter logarithmic matrix, E represents Young's modulus, σ represents Poisson's ratio, ρ represents density, ln represents logarithmic operation, and T represents the transpose of the matrix;

[0083] Step 2.2: Calculate the vertical and horizontal fractional difference vectors:

[0084] The calculation formula is shown in formula 2 and 3:

[0085] ψ a (k) = (-1) kΓ(a+1) / [Γ(k+1)Γ(a-k+1)] Formula (2)

[0086]

[0087] k, a are fractional-order difference coefficients.

[0088] Step 2.3: Use the vertical and horizontal fractional-order difference vectors to construct the vertical and horizontal fractional-order difference matrix and the vertical first-order difference matrix, as shown in formulas 4, 5, and 6.

[0089]

[0090]

[0091] D=d*E (3i-1)×(3i-1) Formula (6)

[0092] Among them, * represents the convolution operation, represents the transverse fractional difference matrix, represents the longitudinal fractional-order difference matrix, D represents the longitudinal first-order difference matrix, d=[-1 1] represents the longitudinal first-order difference vector, E represents the unit matrix, i represents the total number of rows of a single parameter to be inverted, and j represents the total number of columns of a single parameter to be inverted.

[0093] Step 2.4: Construct a forward model based on the fractional-order total variation prestack elastic three-parameter, as shown in Formula 7:

[0094]

[0095] Among them, ||||1 represents the L1 norm, ||||2 represents the L2 norm, * represents the convolution operation, λ is the difference regularization factor, and μ is the weight coefficient of the fidelity term. S is composed of the earthquake records S at three angles: near, medium and far. θ1 , S θ2 and S θ3 Composition, as shown in formula 8

[0096] S=[S θ1 ,S θ2 ,S θ3 ] T Formula (8)

[0097] W is the wavelet matrix, as shown in Formula 9

[0098]

[0099] where w q Represents the qth data of the seismic wavelet, where q represents the length of the wavelet.

[0100] G represents the coefficient matrix of the three elastic parameters to be inverted, as shown in Formula 10

[0101]

[0102] Among them, G E The coefficient matrix representing Young's modulus, G σ The coefficient matrix representing Poisson's ratio, G ρ The coefficient matrix representing the density.

[0103] Step 2.5: Introduce the Lagrange multiplier term R based on the forward model y , R x and the dual term C y , C x , the objective function of prestack elastic three-parameter seismic inversion based on fractional-order total variation is obtained as shown in formula 11:

[0104]

[0105] Among them, η is the weight coefficient of the dual term.

[0106] Step 3: Update L0 by combining the alternating direction multiplier method and the forward model to obtain the updated elastic three-parameter logarithmic matrix L i+1 ; The steps are specifically:

[0107] Step 3.1: Use the initial elastic three-parameter logarithmic matrix L0 as the initial model L for the prestack elastic three-parameter seismic inversion of fractional total variation 1 , and introduce the initial value R of the Lagrange multiplier x =R y = 0 and its dual term C x =C y =0;

[0108] Step 3.2: Update the elastic three-parameter logarithmic matrix L using the alternating direction multiplier algorithm, the criterion of taking the extreme value of the gradient to zero, and the eigenvalue decomposition i+1 , the calculation principle is shown in formula 12:

[0109]

[0110] Where U and V are square matrices and The orthogonal matrix obtained after eigenvalue decomposition is shown in formula 13

[0111]

[0112] Where SVD represents eigenvalue decomposition, and Λ and Σ are the eigenvalue diagonal matrices obtained after decomposition.

[0113] In formula 12, The update formula is shown in Formula 14

[0114]

[0115] Where a represents the number of eigenvalues ​​in the eigenvalue diagonal matrix Λ, b represents the number of eigenvalues ​​in the eigenvalue diagonal matrix Σ,

[0116] Step 3.3: Update the Lagrange multiplier term R according to the alternating direction multiplier algorithm and the soft threshold shrinkage algorithm x i+1 , R y i+1 The calculation formula is shown in Formula 15:

[0117]

[0118] Step 3.4: Update the dual term according to the alternating direction multiplier algorithm and Fermat's lemma The calculation formula is shown in Formula 16:

[0119]

[0120] Step 4: Determine whether the values ​​before and after the update satisfy ||L i+1 -L i ||2 / ||L i ||2>tol, if yes, return to step 3 to loop, if no, according to Young's modulus, Poisson's ratio and density and elastic three-parameter logarithmic matrix L i+1 The relationship between Young's modulus, Poisson's ratio and density is obtained:

[0121] According to the parameters to be inverted and the elastic three-parameter logarithmic matrix L i+1 The relationship between and is used to obtain the parameters to be inverted, as shown in Formula 17:

[0122]

[0123] Step 5: Slice the inversion results of Young's modulus and Poisson's ratio by layer to obtain the fracture prediction results.

[0124] like Figure 4 As shown in Figure 2, the inversion results of the three parameters of prestack elasticity can better conform to the trend of seismic records than the initial model, proving the correctness of this method. Figure 5, the left side shows the crack prediction result corresponding to the inversion result of Young's modulus, and the right side shows the crack prediction result corresponding to the inversion result of Poisson's ratio. The present invention uses fractional total variation to establish sparse regularization terms and construct the objective function of prestack elastic three-parameter seismic inversion, further solves the objective function by alternating direction multiplier method, obtains the inversion results of Young's modulus and Poisson's ratio, and obtains the crack prediction result by slicing along the layer through the inversion results, and proposes a prestack crack prediction method for lacustrine carbonate based on fractional total variation. The present invention improves the accuracy and stability of crack prediction.

[0125] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. A method for predicting prestack fractures in lacustrine carbonate rocks based on fractional-order total variation, characterized in that: The following steps are involved: Step 1: Preprocess the seismic data to obtain the initial model of the parameters to be inverted; Step 2: According to the processed seismic data, Young's modulus, Poisson's ratio and density are combined into a logarithmic matrix L0, and a prestack elastic parameter forward model and objective function based on fractional total variation are constructed; Step 3: Combine the alternating direction multiplication method, the extreme value of the gradient being zero, and the eigenvalue decomposition to inversely update the logarithmic matrix to obtain the updated logarithmic matrix L i ; Step 4: Determine whether the values ​​before and after the update satisfy L i+1 -L i 2 / L i 2>tol, if yes, return to step 3 to loop; if no, calculate the logarithmic matrix L of the three parameters of Young's modulus, Poisson's ratio, density and elasticity according to the logarithmic matrix L i+1 The relationship between Young's modulus, Poisson's ratio and density is obtained; Step 5: Slice the inversion results of Young's modulus and Poisson's ratio by layer to obtain the fracture prediction results.

2. The method according to claim 1, characterized in that: In step 1, the seismic data includes seismic records, wavelet data and well logging data; the layer information is extracted from the seismic records and the well logging data is interpolated and filtered to obtain the initial model of Young's modulus, Poisson's ratio and density.

3. The method according to claim 1, characterized in that: In step 2, a prestack elastic parameter forward model and objective function based on fractional-order total variation are constructed, which specifically includes the following steps: The vertical and horizontal fractional-order difference vectors are used to construct the vertical and horizontal fractional-order difference matrix and the vertical first-order difference matrix. D=d*E (3i -1)×(3i-1) Among them, * represents the convolution operation, represents the transverse fractional difference matrix, represents the longitudinal fractional-order difference matrix, D represents the longitudinal first-order difference matrix, d=[-11] represents the longitudinal first-order difference vector, i represents the total number of rows of a single parameter to be inverted, and j represents the total number of columns of a single parameter to be inverted; Construct a forward model based on fractional-order total variation prestack elastic three-parameter: Among them, ||||1 represents the L1 norm, ||||2 represents the L2 norm, * represents the convolution operation; S is the seismic record matrix, W is the wavelet matrix, and G represents the coefficient matrix of the three elastic parameters to be inverted; Based on the forward model, the Lagrange multiplier term R is introduced. y , R x and the dual term C y , C x , the objective function of prestack elastic three-parameter seismic inversion based on fractional-order total variation is obtained: Among them, η is the weight coefficient of the dual term.

4. The method according to claim 1 or 3, characterized in that: The relationship between the elastic three-parameter logarithmic matrix and Young's modulus, Poisson's ratio and density is expressed by the following formula: L=[lnE lnσ lnρ] T Among them, L represents the elastic three-parameter logarithmic matrix, E represents Young's modulus, σ represents Poisson's ratio, ρ represents density, ln represents logarithmic operation, and T represents the transpose of the matrix.

5. The method according to claim 3, characterized in that: The vertical and horizontal fractional-order difference vectors are expressed by the following formula: among them, ps a (k)=(-1) k C(a+1) / [C(k+1)C(a-k+1)]; 6. The method according to claim 3, characterized in that: S is composed of earthquake records at three angles: near, medium and far. θ1 , S θ2 and S θ3 The composition is expressed by the following formula: S=[S θ1 ,S θ2 ,S θ3 ] T W is expressed by the following formula: where w q represents the qth data of the seismic wavelet, where q represents the length of the wavelet; G is expressed by the following formula: Among them, G E The coefficient matrix representing Young's modulus, G σ The coefficient matrix representing Poisson's ratio, G ρ The coefficient matrix representing the density.

7. The method according to claim 1, characterized in that: Step 3 The specific steps are: The initial elastic three-parameter logarithmic matrix L0 is used as the initial model L for prestack elastic three-parameter seismic inversion of fractional total variation 1 , and introduce the initial value R of the Lagrange multiplier x =R y = 0 and its dual term C x =C y =0; The elastic three-parameter logarithmic matrix L is updated using the alternating direction multiplier algorithm, the criterion of taking the extreme value of the gradient when it is zero, and the eigenvalue decomposition. i+1 : Where U and V are square matrices and The orthogonal matrix obtained after eigenvalue decomposition: Where SVD represents eigenvalue decomposition, Λ and Σ are the eigenvalue diagonal matrices obtained after decomposition; The update formula is as follows: Where a represents the number of eigenvalues ​​in the eigenvalue diagonal matrix Λ, b represents the number of eigenvalues ​​in the eigenvalue diagonal matrix Σ, Update the Lagrange multiplier term R according to the alternating direction multiplier algorithm and the soft threshold shrinkage algorithm x i+1 , R y i+1 ; Update the dual term according to the alternating direction multiplier algorithm and Fermat's lemma 8. The method according to claim 7, characterized in that: Updated Lagrange multiplier term R x i+1 , R y i+1 It is expressed by the following formula:

9. The method according to claim 7, characterized in that: Updated dual term This is shown in the following formula:

10. The method according to claim 1, characterized in that: In step 4, the inversion parameters and the elastic three-parameter logarithmic matrix L i+1 The relationship is expressed by the following formula:

Citation Information

Patent Citations

  • Formation anisotropy predominant direction predication method based on location Young's modulis

    CN104749619A

  • Method for precisely inverting Young modulus and Poisson's ratio

    CN106597537A

  • Two-term type seism pre-stack inversion method based on Young modulus and Poisson ratio

    CN108398720A

  • Quantitative prediction method and device for crack density of carbonate rocks

    CN108897066A

  • Seismic inversion method based on hybrid high order and fractional order Anisotropic Total p-variation (ATpV) sparse regularization

    CN110208862A