Method and device for seismic inversion of fracture density and azimuth of deep coal seams

Abstract

The present invention provides a method and device for azimuthal seismic inversion of deep coal seam fracture density, which relates to the field of seismic exploration technology, including: obtaining pre-stack seismic gather data and well logging data of a study area; constructing an elastic stiffness matrix of each formation as a VTI medium based on the well logging data; processing the porosity of the coalbed methane reservoir in the well logging data and the elastic stiffness matrix of the coalbed methane reservoir as a VTI medium according to the linear slip theory to obtain the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium, so as to determine the initial model parameter vector, and constructing a target function for inversion model parameters in combination with an approximate formula of reflection coefficient based on pre-stack seismic gather data and orthotropic anisotropy; solving the target function to obtain a target model parameter vector of the study area; and processing the target model parameters of the coalbed methane reservoir using a target prediction model to obtain the fracture density of the coalbed methane reservoir, thereby providing an effective technical means for inversion of deep coal seam fracture density.

Description

Technical Field

[0001] The present invention relates to the field of seismic exploration technology, and in particular to a method and device for seismic inversion of deep coal seam fracture density and azimuth. Background Art

[0002] Coalbed methane (CBM), a cleaner energy source compared to traditional fossil fuels, is becoming increasingly prominent in the energy mix, and its exploration and development has become a research hotspot. Due to the low porosity and permeability of deep coal seams, CBM is stored in large quantities within fractures, requiring hydraulic fracturing for extraction. Fracture density has become a key parameter in predicting the engineering sweet spot in CBM exploration and development, as well as in CBM hydraulic fracturing.

[0003] As research into deep coal seams continues to deepen, researchers are expanding their understanding of coal seams beyond isotropic media and vertically symmetrical transversely isotropic (VTI) media. Studies have shown that deep coal seams develop a complex dual-pore structure, consisting of matrix pores and fractures. Due to the strong in-situ stresses in deep coal seams, horizontal fractures tend to close, resulting in the development of horizontal bedding and vertical fractures. Consequently, the coal rock is considered an orthotropic (OA) medium.

[0004] Conventional Amplitude Variation and Migration (AVO) inversion for coal-measure gas reservoirs assumes isotropy. Even when anisotropy is considered, it is limited to VTI media, with little consideration given to OA media. This simplistic characterization of coalbed methane reservoirs (isotropy or VTI) reduces the assessment of CBM production potential. Therefore, a method for predicting fracture density that considers coal as an orthotropic (OA) medium is urgently needed. Summary of the Invention

[0005] The purpose of the present invention is to provide a method and device for azimuthal seismic inversion of fracture density in deep coal seams, so as to accurately predict the fracture density in orthotropic deep coal seams.

[0006] In a first aspect, the present invention provides a method for azimuthal seismic inversion of deep coal seam fracture density, comprising: obtaining pre-stack seismic track data and well logging data of a study area; constructing an elastic stiffness matrix of each stratum in the study area as a VTI medium based on the well logging data; processing the porosity of the coalbed methane reservoir in the well logging data and the elastic stiffness matrix of the coalbed methane reservoir as a VTI medium according to the linear slip theory to obtain an elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium; constructing an initial model parameter vector of the study area based on the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium and the elastic stiffness matrices of other strata as VTI media; wherein the model parameters include : azimuthal anisotropy gradient, the difference in longitudinal wave velocity propagating along the normal direction of the fracture surface and the direction of the fracture strike; construct an objective function for inversion model parameters based on pre-stack seismic gather data, the approximate formula of the reflection coefficient of orthogonal anisotropy and the initial model parameter vector; solve the objective function to obtain the target model parameter vector of the study area; use the target prediction model to process the target model parameters of the coalbed methane reservoir in the target model parameter vector to obtain the fracture density of the coalbed methane reservoir; among them, each model parameter corresponds to a fracture density prediction model, and the target prediction model represents the model with the highest goodness of fit among the fracture density prediction models corresponding to the two model parameters.

[0007] In an optional embodiment, the porosity of the coalbed methane reservoir in the logging data and the elastic stiffness matrix of the coalbed methane reservoir as a VTI medium are processed according to the linear slip theory to obtain the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium, including: processing the porosity of the coalbed methane reservoir based on the empirical relationship between porosity and fracture density to obtain an estimated value of the fracture density of the coalbed methane reservoir; based on the estimated value of fracture density, the bulk modulus of the preset fracture filling material, the shear modulus of the preset fracture filling material, the shear wave and compression wave velocity of the coalbed methane reservoir The square of the ratio of the elastic stiffness matrix of the coalbed methane reservoir as a VTI medium is used to determine the flexibility matrix of the fractures in the coalbed methane reservoir; the inverse matrix of the elastic stiffness matrix of the coalbed methane reservoir as a VTI medium is calculated to obtain the flexibility matrix of the coalbed methane reservoir as a VTI medium; based on the flexibility matrix of the fractures in the coalbed methane reservoir and the flexibility matrix of the coalbed methane reservoir as a VTI medium, the flexibility matrix of the coalbed methane reservoir as a medium with fractures is determined; the inverse matrix of the flexibility matrix of the medium with fractures is calculated to obtain the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium.

[0008] In an optional embodiment, based on the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium and the elastic stiffness matrix of other formations as VTI media, an initial model parameter vector of the study area is constructed, including: substituting the elastic stiffness coefficients in the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium into the calculation formula of the model parameters to obtain the model parameters of the coalbed methane reservoir; substituting the elastic stiffness coefficients in the elastic stiffness matrix of the target formation as a VTI medium into the calculation formula of the model parameters to obtain the model parameters of the target formation; wherein the target formation represents any formation among the other formations; and performing sliding average processing on the model parameters of all formations in the study area to obtain the initial model parameter vector of the study area.

[0009] In an optional embodiment, solving the objective function includes: calculating the first-order derivative of the objective function with respect to the model parameter vector to obtain a first function; constructing a pseudo-Hessian matrix based on the first function and the model parameter vector; constructing a second function for solving the model parameter vector based on the first function and the pseudo-Hessian matrix; iteratively solving the second function based on the initial model parameter vector and pre-stack seismic trace data until a preset iteration termination condition is reached to obtain the target model parameter vector.

[0010] In an optional embodiment, the second function is iteratively solved based on the initial model parameter vector and pre-stack seismic trace data, including: extracting non-90-degree azimuth seismic records from the pre-stack seismic trace data to obtain a first seismic record; extracting 90-degree azimuth seismic records from the pre-stack seismic trace data to obtain a second seismic record; calculating the difference between the first seismic record and the second seismic record to obtain a third seismic record; using the third seismic record as the real seismic record in the second function, and iteratively solving the second function in combination with the initial model parameter vector.

[0011] In an optional embodiment, it also includes: obtaining a first training sample set and a second training sample set; wherein the first training sample set includes multiple groups of first training samples, each group of first training samples includes: the fracture density and azimuthal anisotropy gradient of the coalbed methane reservoir; the second training sample set includes multiple groups of second training samples, each group of second training samples includes: the fracture density of the coalbed methane reservoir and the difference in longitudinal wave velocity propagating along the normal direction of the fracture surface and the direction of the fracture strike; training the first initial fracture density prediction model based on the first training sample set to obtain a first fracture density prediction model; training the second initial fracture density prediction model based on the second training sample set to obtain a second fracture density prediction model; calculating the goodness of fit of the first fracture density prediction model and the second fracture density prediction model respectively, and determining the target prediction model from the first fracture density prediction model and the second fracture density prediction model based on the calculation results of the goodness of fit.

[0012] In an optional embodiment, the first crack density prediction model and the second crack density prediction model are both support vector regression models.

[0013] In a second aspect, the present invention provides a deep coal seam fracture density azimuthal seismic inversion device, comprising: an acquisition module for acquiring pre-stack seismic track data and logging data of a research area; a first construction module for constructing an elastic stiffness matrix of each stratum in the research area as a VTI medium based on the logging data; a first processing module for processing the porosity of the coalbed methane reservoir and the elastic stiffness matrix of the coalbed methane reservoir as a VTI medium in the logging data according to the linear slip theory to obtain the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium; a second construction module for constructing an initial model parameter vector of the research area based on the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium and the elastic stiffness matrix of other strata as VTI media; wherein, the model The model parameters include: azimuthal anisotropy gradient, the difference in longitudinal wave velocity propagating along the normal direction of the fracture surface and the direction of the fracture strike; the third construction module is used to construct an objective function for inversion model parameters based on pre-stack seismic trace data, the reflection coefficient approximate formula of orthogonal anisotropy and the initial model parameter vector; the solution module is used to solve the objective function to obtain the target model parameter vector of the study area; the second processing module is used to use the target prediction model to process the target model parameters of the coalbed methane reservoir in the target model parameter vector to obtain the fracture density of the coalbed methane reservoir; wherein each model parameter corresponds to a fracture density prediction model, and the target prediction model represents the model with the highest goodness of fit among the fracture density prediction models corresponding to the two model parameters.

[0014] In a third aspect, the present invention provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and when the processor executes the computer program, it implements the deep coal seam fracture density azimuth seismic inversion method described in any one of the aforementioned embodiments.

[0015] In a fourth aspect, the present invention provides a computer-readable storage medium storing computer instructions, which, when executed by a processor, implement the deep coal seam fracture density azimuth seismic inversion method described in any one of the aforementioned embodiments.

[0016] The present invention fully considers the characteristics of deep coal seams containing orthotropic media with vertical fractures. Using linear slip theory, vertical fractures are added to the VTI medium to construct the elastic stiffness matrix of the orthotropic medium. On this basis, an initial model parameter vector is constructed for the study area. The objective function for inverting model parameters is constructed by combining prestack seismic gather data and an approximate formula for the orthotropic reflection coefficient. The model parameters include the following two azimuthal attribute parameters affected by fractures: the azimuthal anisotropy gradient and the difference in longitudinal wave velocity propagating along the fracture surface normal and along the fracture strike. Next, after solving the objective function to obtain the target model parameter vector for the study area, the target model prediction model is used to process the target model parameters of the coalbed methane reservoir in the target model parameter vector to obtain the fracture density of the coalbed methane reservoir. The method of first inverting the model parameters and then using the model to predict the fracture density can effectively avoid the complex linearization operations required to express and directly invert the fracture density, providing an effective technical means for inverting the fracture density of deep coal seams. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 A flow chart of a method for azimuthal seismic inversion of deep coal seam fracture density provided by an embodiment of the present invention;

[0019] Figure 2 A schematic diagram of the inversion effect of model parameters D and E provided in an embodiment of the present invention;

[0020] Figure 3 An intersection diagram of parameter D and crack density provided by an embodiment of the present invention;

[0021] Figure 4 An intersection diagram of a parameter E and crack density provided by an embodiment of the present invention;

[0022] Figure 5 A diagram showing the prediction effect of a model trained with parameter D on crack density in a validation set provided by an embodiment of the present invention;

[0023] Figure 6 A diagram showing the prediction effect of a model trained with parameter E on crack density in a validation set provided by an embodiment of the present invention;

[0024] Figure 7Schematic diagram of the effect of predicting fracture density based on well bypass data using the model trained with parameter E in an embodiment of the present invention;

[0025] Figure 8 A functional module diagram of a deep coal seam fracture density azimuth seismic inversion device provided by an embodiment of the present invention;

[0026] Figure 9 A schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations.

[0028] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention as claimed, but rather merely represents selected embodiments of the present invention. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without creative effort shall fall within the scope of protection of the present invention.

[0029] The following embodiments of the present invention are described in detail with reference to the accompanying drawings. In the absence of conflict, the following embodiments and features in the embodiments may be combined with each other.

[0030] Example 1

[0031] Figure 1 A flowchart of a deep coal seam fracture density azimuth seismic inversion method provided by an embodiment of the present invention is shown in FIG. Figure 1 As shown, the method specifically includes the following steps:

[0032] Step S102: Acquire pre-stack seismic gather data and well logging data in the study area.

[0033] In order to determine the fracture density of deep coal seams in the study area, the embodiment of the present invention needs to use the pre-stack seismic gather data and well logging data of the study area, wherein the well logging data includes the anisotropic parameters of each layer in the study area measured at the wellhead. , , , longitudinal wave velocity , shear wave velocity ,density and porosity data.

[0034] After obtaining prestack seismic gather information for the study area, wavelet information can be extracted and combined with well logging data for well-seismic calibration. Well-seismic calibration is used to map well logging depth domain information to time domain information. Therefore, by comparing the depth position of coal seams on the well logging stratigraphic interpretation map, the time range corresponding to the coalbed methane reservoir on the seismic record can be determined.

[0035] Step S104: constructing an elastic stiffness matrix of each formation in the research area as a VTI medium based on the well logging data.

[0036] After acquiring the well logging data, the present embodiment first uses this data to preliminarily calculate the elastic stiffness matrix for each formation in the study area as a VTI medium (i.e., transversely isotropic). The following formulas calculate the elements in the elastic stiffness matrix for VTI media. Elements not shown in the matrix are assumed to have a value of 0. By substituting the well logging data for each formation into the following formulas, the elastic stiffness matrix for each formation in the study area as a VTI medium can be obtained.

[0037] ,in, , , , , , , , , . That is, express The elastic stiffness coefficient of the i-th row and j-th column in .

[0038] Step S106 , processing the porosity of the coalbed methane reservoir in the logging data and the elastic stiffness matrix of the coalbed methane reservoir as a VTI medium according to the linear slip theory to obtain the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium.

[0039] Existing research shows that deep coal seams have complex pore structures. Therefore, based on the coalbed methane reservoir being considered as a VTI medium, the fracture density of the coalbed methane reservoir can be roughly calculated based on the empirical formula and the porosity of the coalbed methane reservoir obtained by logging. Then, using the linear slip theory (such as Schoenberg's linear slip theory), vertical fractures are added to the VTI medium to characterize the coalbed methane reservoir as an orthotropic medium. Therefore, based on the porosity of the coalbed methane reservoir and the elastic stiffness matrix of the coalbed methane reservoir as a VTI medium, the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium can be constructed. .

[0040] Step S108 : constructing an initial model parameter vector of the research area based on the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium and the elastic stiffness matrix of other formations as VTI media.

[0041] The model parameters include: azimuthal anisotropy gradient, and the difference in P-wave velocity between the normal direction of the fracture surface and the direction of the fracture strike.

[0042] In the embodiment of the present invention, except for the coalbed methane reservoir which is an orthotropic medium, all other strata in the study area are regarded as VTI media. In the orthotropic medium, there are the following two azimuthal attribute parameters affected by fractures: azimuthal anisotropy gradient D, and the difference E in the longitudinal wave velocity propagating along the normal direction of the fracture surface and the direction of the fracture strike. Therefore, the embodiment of the present invention uses the above two attribute parameters as model parameters to be inverted.

[0043] In the above, the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium has been calculated. For other formations, since there are no fractures, the elastic stiffness matrix of other formations as orthotropic media is equivalent to the elastic stiffness matrix of them as VTI media.

[0044] It is known that the parameters D and E are determined based on the elastic stiffness coefficients in the elastic stiffness matrix of the orthotropic medium. Therefore, after determining that each stratum is an orthotropic medium, the parameters D and E of each stratum can be calculated by a preset formula. Next, the parameters D and E of all strata are subjected to sliding average processing. For example, a sliding average of a span of 100 points is performed to construct the initial model parameter vector of the study area. The initial model parameter vector represents the vector constructed by the initial model parameters of each stratum.

[0045] Step S110 : constructing an objective function for inversion model parameters based on pre-stack seismic gather data, an orthotropic reflection coefficient approximation formula, and an initial model parameter vector.

[0046] By substituting the model parameters of each formation in the initial model parameter vector into the orthotropic reflection coefficient approximation formula, the reflection coefficients at different incident angles and azimuths at the logging location can be obtained, and then the coefficient matrix of the model parameters can be constructed.

[0047] The formula for forward modeling synthetic seismic records is known to be: ,in, represents the forward synthetic seismic record, represents the forward operator, represents the model parameter vector, , represents noise, represents the wavelet convolution matrix, The coefficient matrix representing the model parameters, represents the difference matrix, .

[0048] In the Bayesian framework, the embodiment of the present invention assumes that the prior model distribution and likelihood function obey Gaussian distribution, and the posterior probability and prior probability and the likelihood function Positive correlation, , based on the maximum a posteriori probability estimate, the objective function for inversion model parameters is established: ,in, represents pre-stack seismic gather data, that is, real earthquake records, represents the noise covariance diagonal matrix, represents the initial model parameter vector, represents the model prior covariance matrix, which can be obtained from well logs or rock physics relationships.

[0049] Step S112, solving the objective function to obtain the target model parameter vector of the study area.

[0050] Optionally, the Gauss-Newton method is used to iteratively invert the model parameters to obtain the target model parameter vector for the study area. This is a vector constructed from the target model parameters for each stratum. This embodiment of the present invention does not specifically limit the method for solving the objective function; users can adapt the method based on their actual needs.

[0051] Step S114: using the target prediction model to process the target model parameters of the coalbed methane reservoir in the target model parameter vector to obtain the fracture density of the coalbed methane reservoir.

[0052] Among them, each model parameter corresponds to a crack density prediction model, and the target prediction model represents the model with the highest goodness of fit among the crack density prediction models corresponding to the two model parameters.

[0053] It is known that the model parameters of the coalbed methane reservoir are related to the fracture density. Therefore, the embodiment of the present invention pre-trains two fracture density prediction models. One is that the input data is the parameter D and the output data is the fracture density The other is a prediction model where the input data is parameter E and the output data is crack density. prediction model.

[0054] In order to select a prediction model suitable for the research area, the embodiment of the present invention refers to the above method of calculating the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium, and the method of calculating the model parameters according to the elastic stiffness coefficient. By adding vertical cracks with different crack densities to the coalbed methane reservoir of VTI type medium, a large number of orthotropic coal rock elastic stiffness matrices with different crack densities can be created. , known According to the formula of parameters D and E, the corresponding model parameters can be calculated, thereby establishing the training set and validation set of parameters D, E and crack density.

[0055] After the two models are trained, their respective goodness of fit can be calculated, such as the parameters , and then select the larger The corresponding model is used as the target prediction model for predicting fracture density. Therefore, after obtaining the target model parameter vector, the target model parameters of the coalbed methane reservoir are also obtained. If the target prediction model is a prediction model with input data as parameter D and output data as fracture density, then the target prediction model is used to process the target model parameter D of the coalbed methane reservoir to obtain the fracture density of the coalbed methane reservoir; if the target prediction model is a prediction model with input data as parameter E and output data as fracture density, then the target prediction model is used to process the target model parameter E of the coalbed methane reservoir to obtain the fracture density of the coalbed methane reservoir.

[0056] The azimuthal seismic inversion method for deep coal seam fracture density provided by the embodiment of the present invention fully considers the characteristics of deep coal seams containing orthogonal anisotropic media with vertical fractures, and uses linear slip theory to add vertical fractures to the VTI medium to construct the elastic stiffness matrix of the orthotropic medium. On this basis, the initial model parameter vector of the study area is constructed, and the objective function for inversion model parameters is constructed by combining pre-stack seismic data and the orthotropic reflection coefficient approximation formula, wherein the model parameters include the following two azimuthal attribute parameters affected by fractures: azimuthal anisotropy gradient, and the difference in longitudinal wave velocity propagating along the normal direction of the fracture surface and the direction of the fracture strike. Next, after solving the objective function to obtain the target model parameter vector of the study area, the target prediction model is used to process the target model parameters of the coalbed methane reservoir in the target model parameter vector to obtain the fracture density of the coalbed methane reservoir. The method adopted in the embodiment of the present invention of first inverting model parameters and then using the model to predict fracture density can effectively avoid the complex linearization operations of expressing and directly inverting fracture density, and provides an effective technical means for the inversion of fracture density in deep coal seams.

[0057] In an optional embodiment, step S106 is to process the porosity of the coalbed methane reservoir in the well logging data and the elastic stiffness matrix of the coalbed methane reservoir as a VTI medium according to the linear slip theory to obtain the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium, specifically including the following steps:

[0058] Step S1061 : Processing the porosity of the coalbed methane reservoir based on the empirical relationship between porosity and fracture density to obtain an estimated fracture density of the coalbed methane reservoir.

[0059] Specifically, the empirical relationship between porosity and fracture density is: ,in, represents the fracture porosity, represents the crack aspect ratio, Indicates the crack density. Because the current calculation serves the subsequent initial model, accurate crack density is not required, so the crack aspect ratio A rough estimate can be made based on empirical values. Substituting the porosity of the coalbed methane reservoir into the above formula, the estimated fracture density of the coalbed methane reservoir can be obtained.

[0060] Step S1062, based on the estimated value of the fracture density, the bulk modulus of the preset fracture filler, the shear modulus of the preset fracture filler, the square of the ratio of the shear wave velocity to the longitudinal wave velocity of the coalbed methane reservoir, and the elastic stiffness matrix of the coalbed methane reservoir as a VTI medium, determine the flexibility matrix of the fracture in the coalbed methane reservoir.

[0061] In the embodiment of the present invention, the flexibility matrix of the crack is expressed as: ,in, represents the normal crack flexibility, represents the vertical tangential flexibility, represents the horizontal tangential flexibility, It represents the square of the ratio of shear wave and longitudinal wave velocities in the coalbed methane reservoir. represents the bulk modulus of the pre-set crack filler, represents the shear modulus of the pre-set crack filler, The elastic stiffness matrix representing the coalbed methane reservoir as a VTI medium The elastic stiffness coefficient of the i-th row and j-th column in .

[0062] Step S1063, calculate the inverse matrix of the elastic stiffness matrix of the coalbed methane reservoir as a VTI medium to obtain the flexibility matrix of the coalbed methane reservoir as a VTI medium. That is, the flexibility matrix of the coalbed methane reservoir as a VTI medium is expressed as: .

[0063] Step S1064: Based on the flexibility matrix of the fractures in the coalbed methane reservoir and the flexibility matrix of the coalbed methane reservoir as a VTI medium, the flexibility matrix of the coalbed methane reservoir as a fractured medium is determined. The flexibility matrix of the coalbed methane reservoir as a fractured medium is expressed as: .

[0064] Step S1065, calculate the inverse matrix of the flexibility matrix of the fractured medium to obtain the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium. That is, the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium is expressed as . It can be deduced that: ,in, represents the normal crack weakness, Indicates the vertical tangential crack weakness, Indicates the weakness of horizontal tangential cracks.

[0065] In an optional embodiment, the above step S108, based on the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium and the elastic stiffness matrix of other formations as VTI media, constructs the initial model parameter vector of the study area, specifically including the following steps:

[0066] Step S1081: Substitute the elastic stiffness coefficient in the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium into the calculation formula of the model parameters to obtain the model parameters of the coalbed methane reservoir.

[0067] Step S1082 , substituting the elastic stiffness coefficient in the elastic stiffness matrix of the target formation being the VTI medium into the calculation formula of the model parameters to obtain the model parameters of the target formation; wherein the target formation represents any formation among the other formations.

[0068] Step S1083: Perform sliding average processing on the model parameters of all strata in the study area to obtain the initial model parameter vector of the study area.

[0069] Specifically, the embodiment of the present invention refers to the approximate formula of the orthotropic reflection coefficient: ,in, , , represents the angle of incidence, represents the azimuth angle relative to the crack normal plane, represents the longitudinal wave impedance, , represents the anisotropic shear modulus, , represents the longitudinal wave phase velocity along the crack surface, , represents the azimuthal anisotropy gradient, , , , is positively correlated with the crack density, so it is often used to express relative crack density or horizontal stress anisotropy. It represents the difference in the longitudinal wave velocity propagating along the normal direction of the crack surface and the crack strike direction, , subscript i represents the parameters of the i-th underground layer.

[0070] in, are orthotropic parameters, , , , , , In the above expression of orthotropic parameters, It represents the elastic stiffness coefficient in the orthotropic elastic stiffness matrix. From the above description, it can be seen that, except for the coalbed methane reservoir, the elastic stiffness matrix of other strata as VTI media is the elastic stiffness matrix of other strata as orthotropic media. Therefore, by substituting the orthotropic elastic stiffness coefficient of each stratum into the formula of parameters D and E, the model parameters of each stratum can be obtained. Finally, the initial model parameter vector of the study area is obtained through sliding average processing.

[0071] In an optional implementation manner, the above step S112, solving the objective function, specifically includes the following steps:

[0072] Step S1121 , calculating the first-order derivative of the objective function with respect to the model parameter vector to obtain a first function.

[0073] Known , find the first-order derivative of the model parameter vector, that is, .

[0074] Step S1122: construct a pseudo Hessian matrix based on the first function and the model parameter vector.

[0075] In the embodiment of the present invention, the pseudo-Hessian matrix is ​​expressed as: .

[0076] Step S1123: construct a second function for solving the model parameter vector based on the first function and the pseudo-Hessian matrix.

[0077] Specifically, the second function is expressed as: ,in, Represents the model parameter vector of the kth iteration. When k=0, .

[0078] Step S1124 , based on the initial model parameter vector and pre-stack seismic gather data, the second function is iteratively solved until a preset iteration termination condition is reached to obtain a target model parameter vector.

[0079] Optionally, the preset iteration termination condition is reaching a specified number of iterations.

[0080] In an optional implementation, the above step S1124, iteratively solving the second function based on the initial model parameter vector and pre-stack seismic gather data, specifically includes the following steps:

[0081] Step S11241: extracting non-90-degree azimuth seismic records from pre-stack seismic gather data to obtain a first seismic record.

[0082] Step S11242: extract the 90-degree azimuth seismic record from the pre-stack seismic gather data to obtain a second seismic record.

[0083] Step S11243, calculating the difference between the first earthquake record and the second earthquake record to obtain a third earthquake record.

[0084] Step S11244: Use the third earthquake record as the real earthquake record in the second function, and iteratively solve the second function in combination with the initial model parameter vector.

[0085] Specifically, it can be seen from the reflection coefficient approximation formula that it consists of two parts: and composition, It only depends on the angle of incidence. It is related to both the incident angle and the azimuth angle. In order to reduce the ill-posedness of the inversion, the embodiment of the present invention adopts a step-by-step inversion strategy for both.

[0086] against Part, according to From the approximate formula, we can see that when , , , reflection coefficient Only related to the angle of incidence Since the seismic record is synthesized by the convolution of wavelet and reflection coefficient, the seismic record with an azimuth angle of 90 degrees eliminates Part of the information can be used to invert parameters A, B, and C. The inversion steps of parameters A, B, and C are: each seismic record with different incidence angles of 90 degrees Composition vector , ; The model parameter vector composed of parameters A, B, and C is expressed as , Initial value of The construction method refers to the initial model parameter vector above The calculation method of After that, refer to the iteration formula of the model parameter vector and iterate a certain number of times to obtain the parameters A, B, and C.

[0087] against part, due to the same incident angle part Same, but 90 degrees azimuth , so the reflection coefficient at the non-90 degree azimuth angle minus the reflection coefficient at the same incident angle of 90 degrees can be obtained . That is, .

[0088] Therefore, according to the distribution rate of convolution, the difference between the earthquake record with non-90 degree azimuth and the earthquake record with the same incident angle of 90 degrees can eliminate the influence of the convolution. The earthquake records affected by The affected earthquake record is recorded as the third earthquake record .

[0089] Therefore, the initial model parameter vector and the third earthquake record Substituting into the second function, we can calculate , and then and the third earthquake record Substitute into the second function and calculate , and so on, iterate continuously until the preset iteration termination condition is reached, and the target model parameter vector can be obtained.

[0090] In order to verify the inversion effect of the embodiment of the present invention on the model parameters, the initial inversion model is generated using a sliding average. Figure 2 A schematic diagram of the inversion effect of model parameters D and E provided in an embodiment of the present invention, through Figure 2 It can be seen that the inversion results of the model parameters can match the actual values ​​well, except that some high-frequency parts cannot be fully fitted due to factors such as the frequency limitation of the seismic wavelet.

[0091] In an optional implementation manner, the embodiment of the present invention further includes the following steps:

[0092] Step S201, obtaining a first training sample set and a second training sample set; wherein the first training sample set includes multiple groups of first training samples, each group of first training samples includes: the fracture density and azimuthal anisotropy gradient of the coalbed methane reservoir; the second training sample set includes multiple groups of second training samples, each group of second training samples includes: the fracture density of the coalbed methane reservoir and the difference in longitudinal wave velocity propagating along the normal direction of the fracture surface and the direction of the fracture strike.

[0093] Step S202 : training a first initial crack density prediction model based on a first training sample set to obtain a first crack density prediction model.

[0094] Step S203 : training the second initial crack density prediction model based on the second training sample set to obtain a second crack density prediction model.

[0095] Step S204 , respectively calculating the goodness of fit of the first crack density prediction model and the second crack density prediction model, so as to determine a target prediction model from the first crack density prediction model and the second crack density prediction model based on the calculation results of the goodness of fit.

[0096] Optionally, the first crack density prediction model and the second crack density prediction model are both support vector regression models.

[0097] In one embodiment, Figure 3 and Figure 4 The intersection diagrams of parameters D and E and crack density are used as the training set for support vector regression to train the model. Figure 5 and Figure 6 The prediction effect of the model trained with parameters D and E on the crack density in the validation set is shown in Figure 2. For parameter D, >0.8, relative to the data quality of the training data, it can be considered that the prediction effect is good, and it can roughly fit the changing trend of D relative to the crack density. At the same time, it is noted that in the part where the crack density is <0.1, the smaller D is, the better the prediction of the crack density is, and the linear relationship between the two is more obvious; compared with parameter D, parameter E has a better linear relationship with crack density, and the prediction effect is better. >0.9, by comparison A prediction model with better effect is selected. In this example, parameter E is finally selected as the target model parameter for predicting crack density. Figure 7 This is a schematic diagram of the effect of predicting fracture density based on well bypass data using the model trained with parameter E in an embodiment of the present invention. That is, parameter E is selected as the target model parameter for predicting fracture density to train a support vector regression model, and then the fracture density is predicted based on the E parameter obtained by inversion of the actual well bypass. Figure 7 The predicted curve of fracture density is well fitted with the actual curve, which shows that the target prediction model can effectively help predict the fracture density distribution of underground coalbed methane reservoirs from the model parameters inverted from the actual seismic profile.

[0098] In summary, the embodiments of the present invention fully consider the characteristics of orthogonal anisotropic (OA) media containing vertical cracks in deep coal seams, use Schoenberg's linear slip theory to add vertical cracks to construct the rock elastic stiffness matrix of the OA medium, and on this basis use the AVOAz inversion technology to fully utilize the amplitude information of the seismic incident angle and azimuth angle, derive the partial derivatives of the objective function constructed based on the reflection coefficient of azimuthal anisotropy with respect to the model parameters, use the distributed inversion strategy to reduce the instability of multi-parameter inversion, and obtain a prediction model for the crack density of each model parameter through training using the machine learning support vector regression method, which provides a more effective means for the inversion of crack density in deep coal seams.

[0099] Example 2

[0100] An embodiment of the present invention also provides a deep coal seam fracture density azimuthal seismic inversion device, which is mainly used to execute the deep coal seam fracture density azimuthal seismic inversion method provided in the above-mentioned embodiment 1. The deep coal seam fracture density azimuthal seismic inversion device provided in the embodiment of the present invention is specifically introduced below.

[0101] Figure 8 A functional module diagram of a deep coal seam fracture density azimuth seismic inversion device provided by an embodiment of the present invention is shown in FIG. Figure 8 As shown, the device mainly includes: an acquisition module 11, a first construction module 12, a first processing module 13, a second construction module 14, a third construction module 15, a solution module 16, and a second processing module 17, wherein:

[0102] The acquisition module 11 is used to acquire pre-stack seismic gather data and well logging data in the study area.

[0103] The first construction module 12 is used to construct an elastic stiffness matrix of each formation in the research area as a VTI medium based on the well logging data.

[0104] The first processing module 13 is used to process the porosity of the coalbed methane reservoir and the elastic stiffness matrix of the coalbed methane reservoir as a VTI medium in the logging data according to the linear slip theory to obtain the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium.

[0105] The second construction module 14 is used to construct the initial model parameter vector of the study area based on the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium and the elastic stiffness matrix of other formations as VTI media; wherein the model parameters include: azimuthal anisotropy gradient, and the difference in longitudinal wave velocity propagating along the normal direction of the fracture surface and the direction of the fracture strike.

[0106] The third construction module 15 is used to construct an objective function for inversion model parameters based on pre-stack seismic gather data, an orthotropic reflection coefficient approximation formula and an initial model parameter vector.

[0107] The solution module 16 is used to solve the objective function and obtain the target model parameter vector of the research work area.

[0108] The second processing module 17 is used to use the target prediction model to process the target model parameters of the coalbed methane reservoir in the target model parameter vector to obtain the fracture density of the coalbed methane reservoir; wherein each model parameter corresponds to a fracture density prediction model, and the target prediction model represents the model with the highest goodness of fit among the fracture density prediction models corresponding to the two model parameters.

[0109] The embodiments of the present invention fully consider the characteristics of orthotropic media containing vertical fractures in deep coal seams. Using linear slip theory, vertical fractures are added to the VTI medium to construct the elastic stiffness matrix of the orthotropic medium. Based on this, an initial model parameter vector is constructed for the study area. The objective function for inverting model parameters is constructed by combining prestack seismic gather data and an approximate formula for the orthotropic reflection coefficient. The model parameters include the following two azimuthal attribute parameters affected by fractures: the azimuthal anisotropy gradient and the difference in P-wave velocity propagating along the fracture surface normal and along the fracture strike. Next, after solving the objective function to obtain the target model parameter vector for the study area, the target model parameters of the coalbed methane reservoir in the target model parameter vector are processed using a target prediction model to obtain the fracture density of the coalbed methane reservoir. The approach of first inverting the model parameters and then using the model to predict fracture density, adopted by the embodiments of the present invention, effectively avoids the complex linearization operations required to express and directly invert fracture density, providing an effective technical approach for inverting fracture density in deep coal seams.

[0110] Optionally, the first processing module 13 is specifically configured to:

[0111] The porosity of the coalbed methane reservoir is processed based on the empirical relationship between porosity and fracture density to obtain the estimated value of the fracture density of the coalbed methane reservoir.

[0112] The flexibility matrix of the fractures in the coalbed methane reservoir is determined based on the estimated value of the fracture density, the bulk modulus of the preset fracture filling material, the shear modulus of the preset fracture filling material, the square of the ratio of the shear wave and compression wave velocities of the coalbed methane reservoir, and the elastic stiffness matrix of the coalbed methane reservoir as a VTI medium.

[0113] The inverse matrix of the elastic stiffness matrix of the coalbed methane reservoir as VTI medium is calculated to obtain the flexibility matrix of the coalbed methane reservoir as VTI medium.

[0114] Based on the flexibility matrix of the fractures in the CBM reservoir and the flexibility matrix of the CBM reservoir as a VTI medium, the flexibility matrix of the CBM reservoir as a medium with fractures is determined.

[0115] The inverse matrix of the flexibility matrix of the fractured medium is calculated to obtain the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium.

[0116] Optionally, the second building module 14 is specifically configured to:

[0117] The elastic stiffness coefficients in the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium are substituted into the calculation formula of the model parameters to obtain the model parameters of the coalbed methane reservoir.

[0118] The elastic stiffness coefficient in the elastic stiffness matrix of the target formation being the VTI medium is substituted into the calculation formula of the model parameters to obtain the model parameters of the target formation; wherein the target formation represents any formation among the other formations.

[0119] The model parameters of all strata in the study area are processed by sliding average to obtain the initial model parameter vector of the study area.

[0120] Optionally, the solution module 16 includes:

[0121] The calculation unit is used to calculate the first-order derivative of the objective function with respect to the model parameter vector to obtain a first function.

[0122] The first construction unit is used to construct a pseudo Hessian matrix based on the first function and the model parameter vector.

[0123] The second construction unit is used to construct a second function for solving the model parameter vector based on the first function and the pseudo-Hessian matrix.

[0124] The solving unit is used to iteratively solve the second function based on the initial model parameter vector and pre-stack seismic gather data until a preset iteration termination condition is reached to obtain the target model parameter vector.

[0125] Optionally, the solving unit is specifically used to:

[0126] Seismic records with non-90-degree azimuth angles are extracted from pre-stack seismic gather data to obtain the first seismic record.

[0127] The 90-degree azimuth earthquake record is extracted from the pre-stack seismic gather data to obtain the second earthquake record.

[0128] The difference between the first earthquake record and the second earthquake record is calculated to obtain a third earthquake record.

[0129] The third earthquake record is used as the real earthquake record in the second function, and the second function is iteratively solved in combination with the initial model parameter vector.

[0130] Optionally, the device is further used to:

[0131] A first training sample set and a second training sample set are obtained; wherein the first training sample set includes multiple groups of first training samples, and each group of first training samples includes: the fracture density and azimuthal anisotropy gradient of the coalbed methane reservoir; the second training sample set includes multiple groups of second training samples, and each group of second training samples includes: the fracture density of the coalbed methane reservoir and the difference in longitudinal wave velocity propagating along the normal direction of the fracture surface and the direction of the fracture strike.

[0132] The first initial crack density prediction model is trained based on the first training sample set to obtain a first crack density prediction model.

[0133] The second initial crack density prediction model is trained based on the second training sample set to obtain a second crack density prediction model.

[0134] The goodness of fit of the first crack density prediction model and the second crack density prediction model are calculated respectively, so as to determine a target prediction model from the first crack density prediction model and the second crack density prediction model based on the calculation results of the goodness of fit.

[0135] Optionally, the first crack density prediction model and the second crack density prediction model are both support vector regression models.

[0136] Example 3

[0137] See also Figure 9 An embodiment of the present invention provides an electronic device, which includes: a processor 60, a memory 61, a bus 62 and a communication interface 63, wherein the processor 60, the communication interface 63 and the memory 61 are connected via the bus 62; the processor 60 is used to execute an executable module stored in the memory 61, such as a computer program.

[0138] Memory 61 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk drive. Communication between the system network element and at least one other network element is achieved via at least one communication interface 63 (which may be wired or wireless), and may utilize the Internet, a wide area network, a local area network, a metropolitan area network, or the like.

[0139] The bus 62 may be an ISA bus, a PCI bus, or an EISA bus. The bus may be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, Figure 9 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0140] Among them, the memory 61 is used to store programs, and the processor 60 executes the program after receiving the execution instruction. The method executed by the device defined by the process disclosed in any embodiment of the above-mentioned embodiment of the present invention can be applied to the processor 60 or implemented by the processor 60.

[0141] The processor 60 may be an integrated circuit chip with signal processing capabilities. During implementation, the steps of the above method may be performed by hardware integrated logic circuits or software instructions within the processor 60. The processor 60 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It may implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of the present invention. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of the present invention may be directly executed by a hardware decoding processor or by a combination of hardware and software modules within the decoding processor. The software modules may be located in storage media well-known in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or the like. The storage medium is located in the memory 61 , and the processor 60 reads the information in the memory 61 and completes the steps of the above method in combination with its hardware.

[0142] The computer program product of a method and apparatus for azimuthal seismic inversion of deep coal seam fracture density provided in an embodiment of the present invention includes a computer-readable storage medium storing non-volatile program code executable by a processor. The instructions included in the program code can be used to execute the method described in the previous method embodiment. For specific implementation, please refer to the method embodiment and will not be repeated here.

[0143] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0144] If the functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage media include various media capable of storing program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0145] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings.

[0146] In the description of the present invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer" and the like indicate positions or locations based on the positions shown in the accompanying drawings, or the positions or locations in which the inventive product is typically placed when in use. These terms are intended solely to facilitate the description of the present invention and to simplify the description, and are not intended to indicate or imply that the devices or components referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limitations on the present invention. Furthermore, the terms "first," "second," and "third," etc., are used solely to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0147] Furthermore, terms such as "horizontal," "vertical," and "overhanging" do not necessarily imply that a component must be absolutely horizontal or overhanging, but rather that it can be slightly tilted. For example, "horizontal" simply means that its direction is more horizontal than "vertical," and does not mean that the structure must be completely horizontal, but rather that it can be slightly tilted.

[0148] In the description of the present invention, it should also be noted that, unless otherwise expressly specified or limited, the terms "disposed," "installed," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; they may refer to mechanical connections or electrical connections; they may refer to direct connections or indirect connections through an intermediate medium; and they may refer to internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0149] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for azimuthal seismic inversion of deep coal seam fracture density, characterized in that: include: Obtain pre-stack seismic gather data and well logging data in the study area; Constructing an elastic stiffness matrix of each formation in the study area as a VTI medium based on the well logging data; The porosity of the coalbed methane reservoir in the well logging data and the elastic stiffness matrix of the coalbed methane reservoir as a VTI medium are processed according to the linear slip theory to obtain the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium; Based on the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium and the elastic stiffness matrices of other formations as VTI media, an initial model parameter vector for the study area is constructed; wherein the model parameters include: azimuthal anisotropy gradient, the difference in P-wave velocity propagating along the fracture surface normal direction and the fracture strike direction; constructing an objective function for inverting the model parameters based on the pre-stack seismic gather data, an orthotropic reflection coefficient approximation formula, and the initial model parameter vector; Calculating the first-order derivative of the objective function with respect to the model parameter vector to obtain a first function; constructing a pseudo-Hessian matrix based on the first function and the model parameter vector; Constructing a second function for solving the model parameter vector based on the first function and the pseudo-Hessian matrix; Extracting non-90-degree azimuth seismic records from the pre-stack seismic gather data to obtain a first seismic record; Extracting a 90-degree azimuth seismic record from the pre-stack seismic gather data to obtain a second seismic record; calculating a difference between the first seismic record and the second seismic record to obtain a third seismic record; Using the third earthquake record as a real earthquake record in the second function, and combining it with the initial model parameter vector to iteratively solve the second function until a preset iteration termination condition is reached, thereby obtaining a target model parameter vector for the study area; The target model parameters of the coalbed methane reservoir in the target model parameter vector are processed using a target prediction model to obtain the fracture density of the coalbed methane reservoir; wherein each model parameter corresponds to a fracture density prediction model, and the target prediction model represents the model with the highest goodness of fit among the fracture density prediction models corresponding to the two model parameters.

2. The deep coal seam fracture density azimuth seismic inversion method according to claim 1, characterized in that: The porosity of the coalbed methane reservoir in the well logging data and the elastic stiffness matrix of the coalbed methane reservoir as a VTI medium are processed according to the linear slip theory to obtain the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium, including: Processing the porosity of the coalbed methane reservoir based on an empirical relationship between porosity and fracture density to obtain an estimated fracture density value of the coalbed methane reservoir; Determining a compliance matrix of fractures in the coalbed methane reservoir based on the estimated fracture density, the bulk modulus of a predetermined fracture filler, the shear modulus of the predetermined fracture filler, the square of the ratio of shear wave to longitudinal wave velocities in the coalbed methane reservoir, and the elastic stiffness matrix of the coalbed methane reservoir as a VTI medium; Calculating the inverse matrix of the elastic stiffness matrix of the coalbed methane reservoir as a VTI medium to obtain the flexibility matrix of the coalbed methane reservoir as a VTI medium; Determining a flexibility matrix of the coalbed methane reservoir as a medium with fractures based on a flexibility matrix of fractures in the coalbed methane reservoir and a flexibility matrix of the coalbed methane reservoir as a VTI medium; The inverse matrix of the flexibility matrix of the fractured medium is calculated to obtain the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium.

3. The deep coal seam fracture density azimuth seismic inversion method according to claim 1, characterized in that: Based on the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium and the elastic stiffness matrix of other formations as VTI media, the initial model parameter vector of the study area is constructed, including: Substituting the elastic stiffness coefficient in the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium into the calculation formula of the model parameters to obtain the model parameters of the coalbed methane reservoir; Substituting the elastic stiffness coefficient in the elastic stiffness matrix of the target formation being a VTI medium into the calculation formula of the model parameters to obtain the model parameters of the target formation; wherein the target formation represents any formation among the other formations; The model parameters of all strata in the study area are processed by sliding average to obtain an initial model parameter vector of the study area.

4. The deep coal seam fracture density azimuth seismic inversion method according to claim 1, characterized in that: Also includes: Obtain a first training sample set and a second training sample set; wherein the first training sample set includes multiple groups of first training samples, each group of first training samples includes: the fracture density and azimuthal anisotropy gradient of the coalbed methane reservoir; the second training sample set includes multiple groups of second training samples, each group of second training samples includes: the fracture density of the coalbed methane reservoir and the difference in the longitudinal wave velocity propagating along the normal direction of the fracture surface and the fracture strike direction; Training a first initial crack density prediction model based on the first training sample set to obtain a first crack density prediction model; Training a second initial crack density prediction model based on the second training sample set to obtain a second crack density prediction model; The goodness of fit of the first crack density prediction model and the second crack density prediction model are calculated respectively, so as to determine the target prediction model from the first crack density prediction model and the second crack density prediction model based on the calculation results of the goodness of fit.

5. The deep coal seam fracture density azimuth seismic inversion method according to claim 4, characterized in that: The first crack density prediction model and the second crack density prediction model are both support vector regression models.

6. A deep coal seam fracture density azimuth seismic inversion device, characterized by: include: Acquisition module, used to obtain pre-stack seismic gather data and well logging data in the study area; A first construction module is used to construct an elastic stiffness matrix of each formation in the research area as a VTI medium based on the well logging data; A first processing module is configured to process the porosity of the coalbed methane reservoir in the well logging data and the elastic stiffness matrix of the coalbed methane reservoir as a VTI medium according to the linear slip theory to obtain the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium; The second construction module is used to construct an initial model parameter vector for the study area based on the elastic stiffness matrix of the coalbed methane reservoir as an orthotropic medium and the elastic stiffness matrix of other formations as VTI media; wherein the model parameters include: azimuthal anisotropy gradient, and the difference in P-wave velocity propagating along the normal direction of the fracture surface and the direction of the fracture strike; A third construction module is configured to construct an objective function for inverting the model parameters based on the pre-stack seismic gather data, an orthotropic reflection coefficient approximation formula, and the initial model parameter vector; Solver modules include: a calculation unit, configured to calculate a first-order derivative of the objective function with respect to a model parameter vector to obtain a first function; A first construction unit is configured to construct a pseudo-Hessian matrix based on the first function and the model parameter vector; A second construction unit is configured to construct a second function for solving the model parameter vector based on the first function and the pseudo-Hessian matrix; A solving unit, configured to extract a non-90-degree azimuth seismic record from the pre-stack seismic gather data to obtain a first seismic record; Extracting a 90-degree azimuth seismic record from the pre-stack seismic gather data to obtain a second seismic record; calculating a difference between the first seismic record and the second seismic record to obtain a third seismic record; Using the third earthquake record as a real earthquake record in the second function, and combining it with the initial model parameter vector to iteratively solve the second function until a preset iteration termination condition is reached, thereby obtaining a target model parameter vector for the study area; The second processing module is used to use the target prediction model to process the target model parameters of the coalbed methane reservoir in the target model parameter vector to obtain the fracture density of the coalbed methane reservoir; wherein each model parameter corresponds to a fracture density prediction model, and the target prediction model represents the model with the highest goodness of fit among the fracture density prediction models corresponding to the two model parameters.

7. An electronic device comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, wherein: When the processor executes the computer program, the deep coal seam fracture density azimuth seismic inversion method described in any one of claims 1 to 5 is implemented.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, the deep coal seam fracture density azimuth seismic inversion method according to any one of claims 1 to 5 is implemented.

Images