CN114428319A - Near-surface modeling method and system for geological structure constraint - Google Patents

Abstract

The invention provides a near-surface modeling method and system for geological structure constraint, and belongs to the field of geophysical exploration. The method comprises the steps of firstly extracting construction information according to digital construction information obtained in an earlier stage or seismic data after offset, then introducing a constraint term for a target function in near-surface modeling to carry out constraint, and solving a near-surface modeling equation after constraint to obtain a high-precision near-surface velocity model. The method solves the problems of low precision, low resolution, inconsistent speed trend with the real condition and the like of the speed model obtained by the conventional near-surface modeling method, and obtains the high-precision and high-resolution near-surface speed model more stably and efficiently by introducing the structural constraint regularization term, thereby providing the high-quality near-surface speed model for the subsequent offset imaging. The invention has good application prospect in the field of seismic signal processing.

Description

A near-surface modeling method and system for geological structure constraints

technical field

The invention belongs to the field of geophysical exploration, and in particular relates to a near-surface modeling method and system constrained by geological structures.

Background technique

With the continuous development of China's oil and gas exploration, the focus of onshore seismic exploration has gradually shifted to complex and undulating surface areas such as mountains, Gobi Desert, Loess Plateau, and desert in the west. In these undulating surface exploration areas, the rapid changes in near-surface elevation and velocity have a serious impact on the acquisition, processing and interpretation of seismic data. The exploration areas in western my country and Sichuan have the dual complex characteristics of complex surface and complex underground structures, which are the hotspots of current seismic exploration and processing research. When dealing with seismic data with complex near-surface feature exploration areas, an accurate near-surface velocity model is the premise and foundation for high-precision migration imaging to achieve good results, and is the key to implementing the structure.

The traditional approach of near-surface velocity modeling is to first pick up the first arrival, then establish an initial velocity model, perform ray tracing on the basis of the initial model, calculate the traveltime residual, and then update the near-surface velocity. Near-surface models. For example, Chinese patent publication CN102590864A discloses a two-step tomographic inversion near-surface modeling method, the specific steps of which include: (1) picking and inputting the first arrival of small refraction; (2) performing tomographic inversion on the first arrival of small refraction, outputting Interpretation results that can reflect the fine changes of velocity in the very shallow layer; 3. Pick up the first arrival time of the cannon record; 4. Interpolate the inversion results of small refraction combined with the interpretation results of micro-logging in the work area using the kriging method to establish the ultra-shallow layer. ⑤ The initial velocity model of mesh division and constrained inversion is established, and the super shallow near-surface velocity body obtained in the previous step is replaced by the shallow part of the initial model generated by the first arrival of the cannon. Loss of very shallow velocity caused by too large displacement; ⑥ Generation of constraint weight field; ⑦ Inversion of near-surface velocity model by constrained tomography, and pick a high velocity model based on the velocity-depth model obtained by tomography inversion. Layer top interface, calculate the static correction of shot point and receiver point.

However, the tomographic equations in conventional near-surface modeling are ill-conditioned, so regularization constraints and large smoothing need to be added, resulting in problems such as low near-surface velocity accuracy and resolution, and even the velocity trend is opposite to the tectonic trend.

SUMMARY OF THE INVENTION

The purpose of the present invention is to solve the above-mentioned problems in the prior art, and to provide a near-surface modeling method and system constrained by geological structure, and introduce geological structure constraints when solving tomographic equations, so that the near-surface velocity and the real structure are more closely related. to improve the accuracy and stability of the inversion.

The present invention is achieved through the following technical solutions:

A first aspect of the present invention provides a near-surface modeling method constrained by geological structures. The method first extracts structural information according to digital structural information obtained in the early stage or migrated seismic data, and then builds a near-surface modeling method. In , a constraint term is introduced for the objective function to constrain, and a high-precision near-surface velocity model is obtained by solving the constrained near-surface modeling equation.

A further improvement of the present invention is that the constraint term is a construction constraint regularization term.

A further improvement of the method of the present invention is that the method comprises:

(1) Input digital structural information or migrated seismic data, prepare the initial near-surface velocity v ₀ , set the number of iterations N, i=0;

(2) Extract the structure tensor from the input digital structural information or the migrated seismic data;

(3) Perform ray tracing according to the current near-surface velocity to obtain the tomographic kernel function K;

(4) Calculate the difference Δt between the theoretical travel time and the real travel time;

(5) Add a regularization term of construction constraint to the tomographic kernel function K to obtain a constrained tomographic equation, then solve the constrained tomographic equation to obtain a slowness update amount, and obtain a new round of iterations according to the slowness update amount speed;

(6) judge whether i is equal to N, if not, then i=i+1, return to the (3) step, if so, then turn to the (7) step;

(7) Output the final speed v.

A further improvement of the method of the present invention is that the digital structural information and the migrated seismic data in the step (1) are all three-dimensional data volumes.

A further improvement of the method of the present invention is that the operation of the step (3) includes:

The path information from the shot point to the detection point is calculated by the current near-surface velocity, and then the tomographic kernel function K is obtained according to the path information.

A further improvement of the method of the present invention is that the tomographic kernel function K in the step (3) is a two-dimensional matrix, wherein the number of rows represents the number of rays, and the number of columns represents the number of velocity grids.

A further improvement of the method of the present invention is that the theoretical travel time in the step (4) is the travel time from the shot point to the detection point calculated according to the current velocity model, and the real travel time is the travel time obtained by field observation;

Use the following formula to calculate the difference between the theoretical travel time and the real travel time

Δt = real travel time - theoretical travel time.

A further improvement of the method of the present invention is that the operation of the step (5) includes:

Add the regularization term of construction constraint to the tomographic kernel function K to obtain the constrained tomographic equation:

是构造约束正则化项；D是预条件算子，如下：Among them, L(s) is the error function in the sense of the two-norm, Δs is the slowness update amount, ε is the regularization parameter,

is the construction constraint regularization term; D is the preconditioner, as follows:

Among them, I represents the identity matrix, ▽ is the Laplacian operator, and T is the structure tensor.

A second aspect of the present invention provides a geological structure-constrained near-surface modeling system, the system comprising: a memory, a processor, and a computer program stored on the memory, the computer program being The processor runs the following steps:

(1) Input digital structural information or migrated seismic data, prepare the initial near-surface velocity v ₀ , set the number of iterations N, i=0;

(2) Extract the structure tensor from the input digital structural information or the migrated seismic data;

(3) Perform ray tracing according to the current near-surface velocity to obtain the tomographic kernel function K;

(4) Calculate the difference Δt between the theoretical travel time and the real travel time;

(5) Add a regularization term of construction constraint to the tomographic kernel function K to obtain a constrained tomographic equation, then solve the constrained tomographic equation to obtain a slowness update amount, and obtain a new round of iterations according to the slowness update amount speed;

(6) judge whether i is equal to N, if not, then i=i+1, return to the (3) step, if so, then turn to the (7) step;

(7) Output the final speed v.

In a third aspect of the present invention, there is provided a computer-readable storage medium, where the computer-readable storage medium stores at least one program executable by a computer, and when the at least one program is executed by the computer, causes the computer to The steps in the described geological structure-constrained near-surface modeling method are performed.

Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention solves the problems of low precision, low resolution, and inconsistency between the velocity trend and the real situation of the velocity model obtained by the conventional near-surface modeling method, and introduces structural constraints by introducing structural constraints. Regularization, more stable and efficient acquisition of high-precision and high-resolution near-surface velocity models, providing high-quality near-surface velocity models for subsequent migration imaging.

The invention has a good application prospect in the field of seismic signal processing.

Description of drawings

Fig. 1 is a block diagram of the steps of the method of the present invention.

Figure 2-1 is the true speed model;

Figure 2-2 is the initial model;

Figure 2-3 shows the near-surface velocity calculated by conventional near-surface modeling methods;

Figure 2-4 is the near-surface velocity calculated by the method of the present invention;

Figure 3 shows the trend of the average non-zero number of sparse coefficients in the iterative process.

Detailed ways

Below in conjunction with accompanying drawing, the present invention is described in further detail:

In order to solve the problems of low modeling accuracy, low resolution and low stability of conventional near-surface modeling methods, as well as velocity trends that do not match the real situation, and inaccurate structural characterization, the present invention uses prior information such as geological outcrops and small refractions. (digital structural information is obtained from geological outcrops), convert it into digital structural information, and constrain the velocity model in the process of near-surface modeling, so as to obtain a high-precision velocity model that conforms to the near-surface geological structure. Better results have been achieved in mid-deep velocity modeling and seismic data migration imaging.

The method of the invention inputs digital structural information or migration imaging profiles, and calculates the structural tensor to extract structural features on the input data. In the process of near-surface modeling, structural constraint regularization is introduced to improve the accuracy and stability of the inversion. Finally, a high-precision near-surface velocity model is obtained.

Examples of the method of the present invention are as follows:

[Example 1]

As shown in Figure 1, the method of the present invention comprises:

(1) Input the digital structural information or the migrated seismic data Y, prepare the initial near-surface velocity v ₀ , set the number of iterations N, i=0;

Either digital structural information or migrated seismic data Y may be used, wherein the digital structural information is the structural information manually described by observing geological outcrops, and the migrated seismic data is based on data-driven structural information. Digital structural information and seismic data are all three-dimensional data volumes.

In general, the initial velocity is the gradient model used, using the velocity field of m x n x o, m is the depth z direction, n is the horizontal y direction, o is the horizontal x direction, at z=0, the velocity is v1, and the velocity gradient is a , the velocity function is v(x, y, z)=v1+z*a.

(2) Extract the structure tensor from the input digital structural information or the migrated seismic data;

The method of extracting the structure tensor adopts the existing method, which is briefly introduced as follows:

For a three-dimensional data volume, the structure tensor T(x) of a certain point can be represented by a 3x3 symmetric positive definite matrix:

Among them, dx, dy, and dz are the gradients of the data along the three directions of x, y, and z, respectively. Through the eigenvalue decomposition of the matrix, the above formula can be expressed as

T(x)=λ _u uu ^T +λ _v vv ^T +λ _w ww ^T (2)

where λ _u , λ _v and λ _w are positive real numbers distributed from 0 to 1, representing the eigenvalues of the matrix T(x), and u, v and w are the corresponding eigenvectors. Usually λ _u ≥ λ _v ≥ λ _w ≥ 0, in this case, the eigenvector u represents the direction with the largest gradient, that is, the direction perpendicular to the structural dip. The eigenvectors v and w represent directions parallel to the structure, respectively, and a larger λ means a larger smoothing scale along the corresponding eigendirection.

(3) Perform ray tracing according to the current near-surface velocity to obtain the tomographic kernel function K;

The existing method is used to achieve, and the introduction is as follows: velocity modeling is a process of continuous updating and iteration, the current near-surface velocity is vi, ray tracing is to calculate the path from the shot point to the detection point through the current near-surface velocity, and then according to the current near-surface velocity Path information, get the tomographic kernel function K.

The tomographic kernel function K is the ray path obtained by ray tracing, and K is a two-dimensional matrix, in which the number of rows represents the number of rays, and the number of columns represents the number of velocity grids. The length of the velocity grid, so K can be calculated from the ray path.

(4) Calculate the difference Δt between the theoretical travel time and the real travel time;

The theoretical travel time is the travel time from the shot point to the receiver point calculated according to the current velocity model, and the real travel time refers to the travel time observed in the field.

(5) Add a regularization term of construction constraint to the tomographic kernel function K to obtain a constrained tomographic equation, then solve the constrained tomographic equation to obtain a slowness update amount, and obtain a new round of iterations according to the slowness update amount speed.

details as follows:

The analytical solution of the anisotropic diffusion equation is introduced, and its implicit expression is the following partial differential equation:

where f(x) is the input seismic data, g(x) is the output seismic data, α is a positive real number, used to control the smoothing force; T(x) is the structural tensor, including the structural strike, inclination and normal gradient information.

Introducing the finite difference approximation, the preconditioner D can be written as:

是拉普拉斯算子，是对称矩阵，整体表示解在沿构造梯度的一个平滑。where I represents the identity matrix,

is the Laplacian operator, is a symmetric matrix, the overall representation of the solution is a smoothing along the structural gradient.

The conventional near-surface modeling formula is as follows:

Among them, K is the tomographic kernel function, Δt is the difference between the pick-up travel time and the calculated travel time, the present invention introduces the regularization term of construction constraint, and obtains the constrained tomographic equation, as shown in formula (6):

是构造约束正则化项。Among them, L(s) is the error function in the sense of two norm, Δs is the slowness update amount to be calculated, ε is the regularization parameter,

is the construction constraint regularizer.

Conventional near-surface modeling is formula (5). In the case of some high and steep structures, the velocity model calculated by the conventional method of formula (5) cannot describe the velocity model well. The process of the present invention in the near-surface modeling In , a structural constraint regularization term is introduced, and after introducing structural information, a near-surface velocity model that is closer to the true velocity can be obtained.

A high-precision near-surface velocity model can be obtained by solving formula (6): formula (6) is solved by using existing methods, such as LSCG, and formula (6) is solved by least squares conjugate gradient method. L(s) is the error function in the sense of the two-norm. By taking the derivative of L(s) and setting the derivative = 0, the slowness update amount is calculated by the current slowness (the reciprocal of the current speed) and the slowness Update amount, get new slowness, get new speed through new slowness.

Each time formula (6) is solved, the update amount of slowness (that is, the reciprocal of speed) is obtained, and the speed of a new round of iteration can be obtained according to the update amount of slowness.

(6) judge whether i is equal to N, if not, then i=i+1, return to the (3) step, if so, then turn to the (7) step;

(7) Output the final speed v.

The present invention also provides a near-surface modeling system constrained by geological structure, and the embodiment of the system is as follows:

[Example 2]

The system includes: a memory, a processor, and a computer program stored on the memory, the computer program executing the following steps when executed by the processor:

(1) Input digital structural information or migrated seismic data, prepare the initial near-surface velocity v ⁰ , set the number of iterations N, i=0;

(2) Extract the structure tensor from the input digital structural information or the migrated seismic data;

(3) Perform ray tracing according to the current near-surface velocity to obtain the tomographic kernel function K;

(4) Calculate the difference Δt between the theoretical travel time and the real travel time;

(5) Add a regularization term of construction constraint to the tomographic kernel function K to obtain a constrained tomographic equation, then solve the constrained tomographic equation to obtain a slowness update amount, and obtain a new round of iterations according to the slowness update amount speed;

(6) judge whether i is equal to N, if not, then i=i+1, return to the (3) step, if so, then turn to the (7) step;

(7) Output the final speed v.

The digital structural information and the migrated seismic data in the step (1) are all three-dimensional data volumes.

The operation of the step (3) includes: calculating the path information from the shot point to the detection point according to the current near-surface velocity, and then obtaining the tomographic kernel function K according to the path information.

The tomographic kernel function K in the step (3) is a two-dimensional matrix, wherein the number of rows represents the number of rays, and the number of columns represents the number of velocity grids.

The theoretical travel time in the described step (4) is the travel time from the shot point to the detection point calculated according to the current velocity model, and the real travel time is the travel time obtained by field observation;

Use the following formula to calculate the difference between the theoretical travel time and the real travel time

Δt = real travel time - theoretical travel time.

The operation of the step (5) includes:

Add the regularization term of construction constraint to the tomographic kernel function K to obtain the constrained tomographic equation:

是构造约束正则化项；D是预条件算子，如下：Among them, L(s) is the error function in the sense of the two-norm, Δs is the slowness update amount, ε is the regularization parameter,

is the construction constraint regularization term; D is the preconditioner, as follows:

是拉普拉斯算子，T是结构张量。where I represents the identity matrix,

is the Laplace operator and T is the structure tensor.

The present invention also provides a computer-readable storage medium. Examples of the computer-readable storage medium are as follows:

[Example 3]

The computer-readable storage medium stores at least one program executable by a computer, and when executed by the computer, the at least one program causes the computer to perform the steps in the geological structure-constrained near-surface modeling method.

Examples of the application of the present invention are as follows:

[Example 4]

Compare the results obtained by damping dictionary learning and conventional K-SVD dictionary learning. Figure 2-1 is the dictionary obtained by damping dictionary learning, Figure 2-2 is the dictionary obtained by the conventional K-SVD algorithm, and Figure 3 is the iterative dictionary. The change trend of the average non-zero number of sparse coefficients in the process, wherein the solid line is the method of the present invention, and the dotted line is the traditional K-SVD algorithm. On the one hand, it can be seen from the comparison between Fig. 2-1 and Fig. 2-2 that the dictionary obtained by the improved dictionary learning algorithm of the present invention contains fewer noise atoms. On the other hand, the average non-zero number in Fig. 3 is compared It also reflects that the improved algorithm can represent seismic data more sparsely, indicating that the improved algorithm can better adapt to seismic data.

Comparing Figures 2-1 to 2-4, it can be found that the inversion results obtained by near-surface modeling based on structural constraints are generally better than those obtained by conventional near-surface modeling; especially the velocity inversion at the valley, which clearly reflects In view of the advantages of tectonic constraints, near the surface on the right side, near-surface modeling based on tectonic constraints can well describe the near-surface velocity changes with high resolution. The comparison of the curves extracted in the depth direction in Figure 3 shows the ability of near-surface modeling based on structural constraints to effectively restore the true velocity near the surface, illustrating the advantages of the near-surface modeling algorithm based on structural constraints.

Finally, it should be noted that the above technical solution is only an embodiment of the present invention. For those skilled in the art, on the basis of the application methods and principles disclosed in the present invention, various types of improvements can be easily made. It is not limited to the methods described in the above-mentioned specific embodiments of the present invention, so the above-described methods are only preferred and have no restrictive meaning.

Claims (10)

1. A near-surface modeling method for geological structure constraint is characterized in that: the method comprises the steps of firstly extracting construction information according to digital construction information obtained in an earlier stage or seismic data after offset, then introducing a constraint term for a target function in near-surface modeling to carry out constraint, and solving a near-surface modeling equation after constraint to obtain a high-precision near-surface velocity model.

2. The near-surface modeling method of geological structure constraints as defined in claim 1, wherein: the constraint term is a construction constraint regularization term.

3. The near-surface modeling method of geological structure constraints as defined in claim 2, wherein: the method comprises the following steps:

(1) inputting digital construction information or seismic data after deflection to prepare initial near-surface velocity v₀Setting the iteration number N, i is 0;

(2) extracting a structure tensor from the input digital construction information or the seismic data after the offset;

(3) carrying out ray tracing according to the current near-surface speed to obtain a chromatography kernel function K;

(4) calculating the difference delta t between the theoretical travel time and the real travel time;

(5) adding a structural constraint regularization term into the chromatographic kernel function K to obtain a constrained chromatographic equation, solving the constrained chromatographic equation to obtain a slowness updating quantity, and obtaining the speed of a new iteration according to the slowness updating quantity;

(6) judging whether i is equal to N, if not, i is i +1, returning to the step (3), and if yes, switching to the step (7);

(7) the final velocity v is output.

4. The near-surface modeling method of geological structure constraints as defined in claim 3, wherein: and (2) the digital construction information and the seismic data after the offset in the step (1) are three-dimensional data volumes.

5. The near-surface modeling method of geological structure constraints as defined in claim 4, wherein: the operation of the step (3) comprises:

and calculating path information from the shot point to the wave detection point according to the current near-surface velocity, and then obtaining a chromatography kernel function K according to the path information.

6. The near-surface modeling method of geological structure constraints as defined in claim 5, wherein: the tomographic kernel function K in step (3) is a two-dimensional matrix, where the number of rows represents the number of rays and the number of columns represents the number of velocity grids.

7. The near-surface modeling method of geological structure constraints as defined in claim 6, wherein: the theoretical travel time in the step (4) is calculated according to a current speed model to obtain travel time from a shot point to a wave detection point, and the real travel time is observed in the field;

and calculating the difference between the theoretical travel time and the real travel time by using the following formula:

Δ t is the true travel time — the theoretical travel time.

8. The near-surface modeling method of geological structure constraints as defined in claim 7, wherein: the operation of the step (5) comprises the following steps:

adding a structural constraint regularization term into the chromatographic kernel function K to obtain a constrained chromatographic equation:

where L(s) is an error function in the sense of a two-norm, Δ s is the slowness update, ε is the regularization parameter,

is to construct a constraint regularization term; d is a preconditioner, as follows:

wherein, I represents an identity matrix,

is the laplacian operator and T is the structure tensor.

9. A near-surface modeling system of geological structure constraints, characterized by: the system comprises: a memory, a processor, and a computer program stored on the memory, the computer program when executed by the processor performing the steps of:

(1) inputting digital construction information or seismic data after deflection to prepare initial near-surface velocity v₀Setting the iteration number N, i is 0;

(2) extracting a structure tensor from the input digital construction information or the seismic data after the offset;

(3) carrying out ray tracing according to the current near-surface speed to obtain a chromatography kernel function K;

(4) calculating the difference delta t between the theoretical travel time and the real travel time;

(5) adding a structural constraint regularization term into the chromatographic kernel function K to obtain a constrained chromatographic equation, solving the constrained chromatographic equation to obtain a slowness updating quantity, and obtaining the speed of a new iteration according to the slowness updating quantity;

(6) judging whether i is equal to N, if not, i is i +1, returning to the step (3), and if yes, switching to the step (7);

(7) the final velocity v is output.

10. A computer-readable storage medium characterized by: the computer-readable storage medium stores at least one program executable by a computer, the at least one program when executed by the computer causing the computer to perform the steps in the method of near-surface modeling of geological structure constraints as claimed in any of claims 1-8.

Images