Earthquake high-resolution inversion method and device

Abstract

The invention provides a seismic high-resolution inversion method and device. The method comprises the following steps: acquiring angle domain pre-stack seismic trace gather data, seismic horizon data, seismic wavelet data, initial low-frequency longitudinal wave impedance, initial low-frequency transverse wave impedance and initial low-frequency density; obtaining a reflection coefficient according to the angle domain pre-stack seismic trace set data, the seismic horizon data, the seismic wavelet data and a Fatti longitudinal wave reflection coefficient equation; establishing an inversion objective function according to the initial low-frequency longitudinal wave impedance, the initial low-frequency transverse wave impedance, the initial low-frequency density and the reflection coefficient, and obtaining an inversion equation through the inversion objective function; presetting an inversion iteration frequency and an inversion error, and carrying out iterative solution on the inversion equation to obtain intermediate-frequency longitudinal wave impedance, intermediate-frequency transverse wave impedance and intermediate-frequency density; and processing the intermediate-frequency longitudinal wave impedance, the intermediate-frequency transverse wave impedance and the intermediate-frequency density by using a preset first formula to obtain final high-resolution longitudinal wave impedance, transverse wave impedance and density. According to the invention, the precision and resolution of seismic inversion can be effectively improved.

Description

Technical Field

[0001] The invention belongs to the technical field of geological exploration, and in particular relates to a high-resolution seismic inversion method and device. Background Art

[0002] As the main technical means for thin reservoir prediction or thin interlayer prediction, seismic high-resolution inversion technology mainly includes deterministic inversion and random inversion. Deterministic inversion is faithful to the seismic data itself, using low-frequency logging to participate in the inversion, with relatively low requirements for the number of wells, and the resolution of the inversion results is relatively low. Seismic random inversion combines random simulation with seismic inversion, using high-frequency data of wells to count their lateral and vertical changes, requiring a certain number of wells, and the resolution of the inversion results is relatively high. When there are few wells in the study area and the target layer is thin, how to improve the inversion accuracy and resolution is one of the difficulties and hot spots of seismic inversion. Summary of the invention

[0003] In order to enrich the seismic high-resolution inversion method to be applicable to more application scenarios, the present invention proposes a multi-scale progressive high-resolution inversion method, including:

[0004] In a first aspect, an embodiment of the present invention provides a seismic high-resolution inversion method, comprising:

[0005] Obtain angle domain pre-stack seismic gather data, seismic layer data, seismic wavelet data, initial low-frequency longitudinal wave impedance, initial low-frequency shear wave impedance, and initial low-frequency density;

[0006] Obtaining a reflection coefficient using the Fatti longitudinal wave reflection coefficient equation according to the angle domain pre-stack seismic gather data, the seismic layer data and the seismic wavelet data;

[0007] An inversion objective function is established according to the initial low-frequency longitudinal wave impedance, the initial low-frequency shear wave impedance, the initial low-frequency density and the reflection coefficient to obtain an intermediate-frequency longitudinal wave impedance, an intermediate-frequency shear wave impedance and an intermediate-frequency density;

[0008] The intermediate frequency longitudinal wave impedance, the intermediate frequency shear wave impedance and the intermediate frequency density are processed by using a preset first formula to obtain high-resolution longitudinal wave impedance, shear wave impedance and density.

[0009] In an optional embodiment, the inversion objective function is established according to the initial low-frequency longitudinal wave impedance, the initial low-frequency shear wave impedance, the initial low-frequency density and the reflection coefficient to obtain the intermediate-frequency longitudinal wave impedance, the intermediate-frequency shear wave impedance and the intermediate-frequency density, including:

[0010] Obtaining an initial model of elastic parameters according to the initial low-frequency longitudinal wave impedance, the initial low-frequency shear wave impedance, and the initial low-frequency density;

[0011] Establishing an inversion objective function using the elastic parameter initial model and the reflection coefficient;

[0012] Derivative the inversion objective function to obtain an inversion equation;

[0013] The number of inversion iterations and the inversion error are preset, and the inversion equation is iteratively solved to obtain the intermediate frequency longitudinal wave impedance, the intermediate frequency shear wave impedance, and the intermediate frequency density.

[0014] In an optional embodiment, an elastic parameter initial model is obtained according to the initial low-frequency longitudinal wave impedance, the initial low-frequency shear wave impedance, and the initial low-frequency density, and the elastic parameter initial model m0 is:

[0015] m0=(ln(Z P0 ),ln(Z S0 ),ln(ρ0))

[0016] Among them, Z P0 is the initial low-frequency longitudinal wave impedance, Z S0 is the initial low-frequency shear wave impedance, ρ0 is the initial low-frequency density, and m0 is the initial model of elastic parameters.

[0017] In an optional embodiment, the establishing of an inversion objective function using the elastic parameter initial model and the reflection coefficient includes:

[0018] Using the seismic wavelet data and the reflection coefficient to simulate and synthesize pre-stack seismic angle gathers according to a preset second formula;

[0019] An inversion objective function is obtained by using the elastic parameter initial model, the simulated synthesized pre-stack seismic angle gather and the acquired angle domain pre-stack seismic gather data.

[0020] In an optional embodiment, the preset second formula is:

[0021] d(θ)=W(t)*R pp (θ)

[0022] Where W(t) is the seismic wavelet data, R pp is the reflection coefficient, and d(θ) is the simulated synthetic pre-stack seismic angle gather.

[0023] In an optional embodiment, the inversion objective function is:

[0024]

[0025] Among them, u is the white noise coefficient, m0 is the initial model of elastic parameters, m is the elastic parameter model, G is the kernel function, α is the parameter for adjusting the model scale, d is the angle domain prestack seismic gather data, Φ iis the intermediate parameter, Ns is the data dimension of the elastic parameter model, and i' is the index number.

[0026] In an optional embodiment, the inversion objective function is derived to obtain the inversion equation:

[0027] According to the inversion objective function, the inversion equation is obtained:

[0028] (G T G+αΨ -1 +μQ)m=G T d+αΨ -1 m0

[0029] Among them, u is the white noise coefficient, m0 is the initial model of elastic parameters, m is the elastic parameter model, G is the kernel function, α is the adjustment model scale parameter, Ψ -1 is the covariance matrix, Q is the sparse constraint parameter, and d is the pre-stack seismic gather data in the angle domain.

[0030] In an optional embodiment, the preset number of inversion iterations and inversion error, iteratively solving the inversion equation to obtain the intermediate frequency longitudinal wave impedance, intermediate frequency shear wave impedance, and intermediate frequency density, include:

[0031] Preset the number of inversion iterations and inversion error;

[0032] Calculating the intermediate parameter using a preset third formula;

[0033] Calculating the sparse constraint parameter using a preset fourth formula;

[0034] Calculating the elastic parameter model by using the intermediate parameters and sparse constraint parameters using iterative least square method and conjugate gradient method;

[0035] Iterating the elastic parameter model until a preset number of inversion iterations and an inversion error are reached to obtain an inverted elastic parameter model;

[0036] The inverted elastic parameter model is subjected to exponential operation to obtain the intermediate frequency longitudinal wave impedance, intermediate frequency shear wave impedance and intermediate frequency density.

[0037] In an optional embodiment, the preset third formula for calculating the intermediate parameter includes:

[0038]

[0039] Among them, Ψ is the covariance matrix of model parameters;

[0040] The matrix S is:

[0041]

[0042] Where n represents the number of inversion parameters, l is the number of inversion sample points, i' is the index number, the dimension of the matrix S is 3*3Ns, and Ns is the data dimension of the elastic parameter model;

[0043] The matrix D1 is:

[0044]

[0045] The dimension of matrix D is 3Ns*3Ns, where Ns is the data dimension of the elastic parameter model.

[0046] In an optional embodiment, the sparse constraint parameter is calculated using a preset fourth formula, including:

[0047]

[0048] Where i is the number of iterations, i' is the number of indexes, Ns is the model size, Q is the sparse constraint parameter, Φ i’ is the intermediate parameter, m i is the elastic parameter model of the i-th iteration.

[0049] In an optional embodiment, calculating the elastic parameter model using the intermediate parameters and the sparse constraint parameters by adopting the iterative least square method and the conjugate gradient method includes:

[0050] m i+1 =(G T G+μQ(i)+αΨ -1 ) -1 (Gd+αΨ -1 m0)

[0051] Among them, μ is the white noise parameter, α is the adjustment model scale parameter, and i is the number of iterations.

[0052] In an optional embodiment, the preset first formula is:

[0053]

[0054] Among them, X(u) is the estimated value of the main variable at location u, Y(u) is the data of the covariate at location u, and λ represents the cokriging weighting coefficient.

[0055] In a second aspect, an embodiment of the present invention provides a seismic high-resolution inversion device, comprising:

[0056] Parameter acquisition module, used to obtain angle domain pre-stack seismic gather data, seismic layer data, seismic wavelet data, initial low-frequency longitudinal wave impedance, initial low-frequency shear wave impedance, and initial low-frequency density;

[0057] A reflection coefficient acquisition module, which obtains the reflection coefficient by using the Fatti longitudinal wave reflection coefficient equation according to the angle domain pre-stack seismic gather data, the seismic layer data and the seismic wavelet data;

[0058] An inversion equation establishment module is used to establish an inversion objective function according to the initial low-frequency longitudinal wave impedance, the initial low-frequency shear wave impedance, the initial low-frequency density and the reflection coefficient to obtain the intermediate-frequency longitudinal wave impedance, the intermediate-frequency shear wave impedance and the intermediate-frequency density;

[0059] The high-resolution parameter acquisition module is used to process the intermediate frequency longitudinal wave impedance, the intermediate frequency shear wave impedance, and the intermediate frequency density using a preset first formula to obtain high-resolution longitudinal wave impedance, shear wave impedance, and density.

[0060] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium, in which instructions are stored. When the instructions are executed on a terminal, the terminal executes the method established as described above.

[0061] In a fourth aspect, an embodiment of the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described above when executing the computer program.

[0062] Based on the above technical solution, the beneficial effects of the present invention compared with the prior art are as follows:

[0063] The invention establishes a high-resolution seismic inversion method. First, angle-domain pre-stack seismic gather data, seismic layer data, seismic wavelet data, initial low-frequency longitudinal wave impedance, initial low-frequency shear wave impedance and initial low-frequency density are obtained; then, the Fatti longitudinal wave reflection coefficient equation is used to obtain the reflection coefficient according to the angle-domain pre-stack seismic gather data, seismic layer data and seismic wavelet data; then, an inversion target function is established according to the initial low-frequency longitudinal wave impedance, the initial low-frequency shear wave impedance, the initial low-frequency density and the reflection coefficient, and an inversion equation is obtained through the inversion target function; the number of inversion iterations and the inversion error are preset, and the inversion equation is iteratively solved to obtain intermediate-frequency longitudinal wave impedance, intermediate-frequency shear wave impedance and intermediate-frequency density; finally, the intermediate-frequency longitudinal wave impedance, the intermediate-frequency shear wave impedance and the intermediate-frequency density are processed by a preset first formula to obtain high-resolution longitudinal wave impedance, shear wave impedance and density. This method first obtains the intermediate frequency longitudinal wave impedance, intermediate frequency shear wave impedance, and intermediate frequency density through inversion, and then uses collaborative Kriging to process the intermediate frequency longitudinal wave impedance, intermediate frequency shear wave impedance, and intermediate frequency density to obtain the final high-resolution longitudinal wave impedance, shear wave impedance, and density. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0065] Figure 1 A schematic diagram of a high-resolution seismic inversion method provided by an embodiment of the present invention;

[0066] Figure 2 A schematic diagram of establishing an inversion objective function provided by an embodiment of the present invention;

[0067] Figure 3 A schematic diagram of establishing an inversion objective function using the elastic parameter initial model and the reflection coefficient provided in an embodiment of the present invention;

[0068] Figure 4 A schematic diagram of iteratively solving the inversion equation provided in an embodiment of the present invention;

[0069] Figure 5 An impedance effect diagram obtained by conventional inversion provided in an embodiment of the present invention;

[0070] Figure 6 The impedance effect diagram obtained by the method provided by the embodiment of the present invention;

[0071] Figure 7 A schematic diagram of a multi-scale progressive high-resolution inversion device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0072] The inventors found that the research technology has limitations. As for the problem of few wells and thin layers in the research area, how to improve the accuracy and resolution of seismic inversion is still a difficult problem. Therefore, the inventors made the present invention after further research and development.

[0073] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0074] Embodiment 1

[0075] The embodiment of the present invention provides a high-resolution seismic inversion method, combining Figure 1 As shown, it includes steps S110 to S140:

[0076] Step S110, obtaining angle domain pre-stack seismic gather data, seismic layer data, seismic wavelet data, initial low-frequency longitudinal wave impedance, initial low-frequency shear wave impedance, and initial low-frequency density.

[0077] In some embodiments, it is necessary to first obtain logging data of wells drilled in the work area, and obtain 3-10 Hz low-frequency trend information based on the longitudinal wave velocity, shear wave velocity, and density logging data of the wells drilled in the work area. Under the constraints of seismic layer data, the initial low-frequency longitudinal wave impedance, initial low-frequency shear wave impedance, and initial low-frequency density are obtained by interpolation and extrapolation.

[0078] Step S120, obtaining the reflection coefficient by using the Fatti longitudinal wave reflection coefficient equation according to the angle domain pre-stack seismic gather data, the seismic layer data and the seismic wavelet data.

[0079] Specifically, the Fatti longitudinal wave reflection coefficient equation is:

[0080] R PP (θ)=c1R P +c2R S +c3R D

[0081] Among them, R pp is the reflection coefficient;

[0082] c1, c2, c3 are:

[0083] c1=1+tan 2 θ

[0084] c2=-8γ 2 sin 2 θ

[0085] c3=-0.5tan 2 θ+2γ 2 sin 2 θ

[0086]

[0087] Wherein, Vs and Vp are the preset shear wave velocity and longitudinal wave velocity respectively;

[0088] Rp, Rs, R D for:

[0089]

[0090]

[0091] R D =Δln(ρ)

[0092] Among them, Z p is the initial low-frequency longitudinal wave impedance; Z s is the initial low-frequency shear wave impedance; ρ is the initial low-frequency density.

[0093] Step S130, establishing an inversion objective function according to the initial low-frequency longitudinal wave impedance, the initial low-frequency shear wave impedance, the initial low-frequency density and the reflection coefficient to obtain the intermediate-frequency longitudinal wave impedance, the intermediate-frequency shear wave impedance and the intermediate-frequency density.

[0094] Specifically, the inversion objective function is established according to the initial low-frequency longitudinal wave impedance, the initial low-frequency shear wave impedance, the initial low-frequency density and the reflection coefficient to obtain the intermediate-frequency longitudinal wave impedance, the intermediate-frequency shear wave impedance and the intermediate-frequency density, such as Figure 2 As shown, it includes steps S1301 to S1304:

[0095] Step S1301, obtaining an initial model of elastic parameters according to the initial low-frequency longitudinal wave impedance, the initial low-frequency shear wave impedance, and the initial low-frequency density.

[0096] Specifically, an elastic parameter initial model is obtained according to the initial low-frequency longitudinal wave impedance, the initial low-frequency shear wave impedance, and the initial low-frequency density. The elastic parameter initial model m0 is:

[0097] m0=(ln(Z P0 ),ln(Z S0 ),ln(ρ0))

[0098] Among them, Z P0 is the initial low-frequency longitudinal wave impedance, Z S0 is the initial low-frequency shear wave impedance, ρ0 is the initial low-frequency density, and m0 is the initial model of elastic parameters.

[0099] Step S1302: establishing an inversion objective function using the elastic parameter initial model and the reflection coefficient.

[0100] Specifically, the elastic parameter initial model and the reflection coefficient are used to establish the inversion objective function, such as Figure 3 As shown, it includes steps S13021 to S13022:

[0101] Step S13021, using the seismic wavelet data and the reflection coefficient to simulate and synthesize pre-stack seismic angle gathers according to a preset second formula.

[0102] Specifically, the preset second formula is:

[0103] d(θ)=W(t)*R pp (θ)

[0104] Where W(t) is the seismic wavelet data, R pp is the reflection coefficient, and d(θ) is the simulated synthetic pre-stack seismic angle gather.

[0105] Step S13022, using the elastic parameter initial model, the simulated pre-stack seismic angle gather and the acquired angle domain pre-stack seismic gather data, to obtain an inversion objective function.

[0106] Specifically, the inversion objective function is:

[0107]

[0108] Among them, u is the white noise coefficient, m0 is the initial model of elastic parameters, m is the elastic parameter model, G is the kernel function, α is the parameter for adjusting the model scale, d is the angle domain prestack seismic gather data, Φ i is the intermediate parameter, Ns is the data dimension of the elastic parameter model, and i' is the index number.

[0109] Step S1303, deriving the inversion objective function to obtain an inversion equation.

[0110] Specifically, according to the inversion objective function, the inversion equation is obtained:

[0111] (G T G+αΨ -1 +μQ)m=G T d+αΨ -1 m0

[0112] Among them, u is the white noise coefficient, m0 is the initial model of elastic parameters, m is the elastic parameter model, G is the kernel function, α is the adjustment model scale parameter, Ψ -1 is the covariance matrix, Q is the sparse constraint parameter, and d is the pre-stack seismic gather data in the angle domain.

[0113] Step S1304, presetting the number of inversion iterations and the inversion error, iteratively solving the inversion equation, and obtaining the intermediate frequency longitudinal wave impedance, the intermediate frequency shear wave impedance, and the intermediate frequency density.

[0114] Specifically, the preset inversion iteration number and inversion error are used to iteratively solve the inversion equation to obtain the intermediate frequency longitudinal wave impedance, intermediate frequency shear wave impedance, and intermediate frequency density, such as Figure 4 As shown, it includes steps S13041 to S13046:

[0115] Step S13041, preset the number of inversion iterations and the inversion error.

[0116] Step S13042: Calculate the intermediate parameter using a preset third formula.

[0117] Specifically, the preset third formula for calculating the intermediate parameter includes:

[0118] Φ i′ =(S) TD1TΨ -1 D1

[0119] Where Ψ is the covariance matrix of the model parameters.

[0120] The matrix S is:

[0121]

[0122] Where n represents the number of inversion parameters, l is the number of inversion sample points, i' is the index number, the dimension of the matrix S is 3*3Ns, and Ns is the data dimension of the elastic parameter model;

[0123] The matrix D1 is:

[0124]

[0125] The dimension of matrix D is 3Ns*3Ns, where Ns is the data dimension of the elastic parameter model.

[0126] Step S13043, using the preset fourth formula to calculate the sparse constraint parameters.

[0127] Specifically, the calculating the sparse constraint parameter by using the preset fourth formula includes:

[0128]

[0129] Where i is the number of iterations, i' is the number of indexes, Ns is the model size, Q is the sparse constraint parameter, Φ i’ is the intermediate parameter, m i is the elastic parameter model of the i-th iteration.

[0130] Step S13044, using the intermediate parameters and sparse constraint parameters to calculate the elastic parameter model using iterative least squares method and conjugate gradient method.

[0131] Step S13045, iterating the elastic parameter model until a preset number of inversion iterations and an inversion error are reached, thereby obtaining an inverted elastic parameter model.

[0132] Step S13046, performing exponential operation on the inverted elastic parameter model to obtain intermediate frequency longitudinal wave impedance, intermediate frequency shear wave impedance, and intermediate frequency density.

[0133] Step S140, using a preset first formula to process the intermediate frequency longitudinal wave impedance, the intermediate frequency shear wave impedance, and the intermediate frequency density to obtain high frequency longitudinal wave impedance, high frequency shear wave impedance, and high frequency density. The formation medium resolution is higher, and the formation information reflected is more comprehensive and clear.

[0134] Specifically, the intermediate frequency longitudinal wave impedance, the intermediate frequency shear wave impedance, and the intermediate frequency density are processed using the collaborative Kriging method to obtain final high-precision longitudinal wave impedance, shear wave impedance, and density.

[0135] Specifically, the first formula is preset as the co-kriging method:

[0136]

[0137] X(u) is the estimated value of the primary variable at location u, Y(u) is the data of the covariate at location u, and λ represents the cokriging weighting coefficient.

[0138] The corresponding co-kriging equations only need to know the covariance function C of the main variable X X (h), the cross-covariance function C of the main variable X and the covariate Y xy (h), the latter can be approximated by the following model C XY (h) = βC X (h), where C X (0) and C Y (0) is the variance function of X and Y, P XY (0) is the linear correlation coefficient of the co-located X and Y data.

[0139] Specifically, using actual data, the impedance effect obtained by conventional inversion method and the impedance effect obtained by this method are compared. Figure 5 is the impedance effect obtained by conventional inversion method, Figure 6 The impedance effect obtained by this method is compared Figure 5 and Figure 6 It can be seen that the stratigraphic medium inversion obtained by this method has higher resolution and reflects more comprehensive and clear stratigraphic information.

[0140] The present invention establishes a high-resolution seismic inversion method. First, angle domain pre-stack seismic gather data, seismic layer data, seismic wavelet data, initial low-frequency longitudinal wave impedance, initial low-frequency shear wave impedance, and initial low-frequency density are obtained; then, the Fatti longitudinal wave reflection coefficient equation is used to obtain the reflection coefficient according to the angle domain pre-stack seismic gather data, seismic layer data, and seismic wavelet data; then, the inversion objective function is established according to the initial low-frequency longitudinal wave impedance, the initial low-frequency shear wave impedance, the initial low-frequency density, and the reflection coefficient, and the inversion equation is obtained through the inversion objective function; the number of inversion iterations and the inversion error are preset, and the inversion equation is iteratively solved to obtain the intermediate frequency longitudinal wave impedance, the intermediate frequency shear wave impedance, and the intermediate frequency density; finally, the intermediate frequency longitudinal wave impedance, the intermediate frequency shear wave impedance, and the intermediate frequency density are processed by the preset first formula to obtain high-resolution longitudinal wave impedance, shear wave impedance, and density. The method can effectively improve the accuracy and resolution of the inversion by performing seismic inversion on the target area with few wells and thin layers.

[0141] Embodiment 2

[0142] The embodiment of the present invention provides a high-resolution seismic inversion device, such as Figure 7 As shown, the device comprises:

[0143] The parameter acquisition module 201 is used to acquire angle domain pre-stack seismic gather data, seismic layer data, seismic wavelet data, initial low-frequency longitudinal wave impedance, initial low-frequency shear wave impedance, and initial low-frequency density.

[0144] The reflection coefficient acquisition module 202 obtains the reflection coefficient by using the Fatti longitudinal wave reflection coefficient equation according to the angle domain pre-stack seismic gather data, the seismic layer data and the seismic wavelet data.

[0145] Specifically, the Fatti longitudinal wave reflection coefficient equation is:

[0146] R PP (θ)=c1R P +c2R S +c3R D

[0147] Among them, R pp is the reflection coefficient;

[0148] c1, c2, c3 are:

[0149] c1=1+tan 2 θ

[0150] c2=-8γ 2 sin 2 θ

[0151] c3=-0.5tan 2 θ+2γ2 sin 2 θ

[0152]

[0153] Wherein, Vs and Vp are the preset shear wave velocity and longitudinal wave velocity respectively;

[0154] Rp, Rs, R D for:

[0155]

[0156]

[0157] R D =Δln(ρ)

[0158] Among them, Z p is the initial low-frequency longitudinal wave impedance; Z s is the initial low-frequency shear wave impedance; ρ is the initial low-frequency density.

[0159] The inversion equation establishment module 203 is used to establish an inversion objective function according to the initial low-frequency longitudinal wave impedance, the initial low-frequency shear wave impedance, the initial low-frequency density and the reflection coefficient to obtain the intermediate frequency longitudinal wave impedance, the intermediate frequency shear wave impedance and the intermediate frequency density.

[0160] Specifically, the inversion objective function is established according to the initial low-frequency longitudinal wave impedance, the initial low-frequency shear wave impedance, the initial low-frequency density and the reflection coefficient to obtain the intermediate-frequency longitudinal wave impedance, the intermediate-frequency shear wave impedance and the intermediate-frequency density, such as Figure 2 As shown, it includes steps S2301 to S2304:

[0161] Step S2301, obtaining an initial model of elastic parameters according to the initial low-frequency longitudinal wave impedance, the initial low-frequency shear wave impedance, and the initial low-frequency density.

[0162] Specifically, an elastic parameter initial model is obtained according to the initial low-frequency longitudinal wave impedance, the initial low-frequency shear wave impedance, and the initial low-frequency density. The elastic parameter initial model m0 is:

[0163] m0=(ln(Z P0 ),ln(Z S0 ),ln(ρ0))

[0164] Among them, Z P0 is the initial low-frequency longitudinal wave impedance, Z S0 is the initial low-frequency shear wave impedance, ρ0 is the initial low-frequency density, and m0 is the initial model of elastic parameters.

[0165] Step S2302: establishing an inversion objective function using the elastic parameter initial model and the reflection coefficient.

[0166] Specifically, the elastic parameter initial model and the reflection coefficient are used to establish the inversion objective function, such as Figure 3 As shown, it includes steps S23021 to S23022:

[0167] Step S23021, using the seismic wavelet data and the reflection coefficient to simulate and synthesize pre-stack seismic angle gathers according to a preset second formula.

[0168] Specifically, the preset second formula is:

[0169] d(θ)=W(t)*R pp (θ)

[0170] Where W(t) is the seismic wavelet data, R pp is the reflection coefficient, and d(θ) is the simulated synthetic pre-stack seismic angle gather.

[0171] Step S23022, using the elastic parameter initial model, the simulated pre-stack seismic angle gather and the acquired angle domain pre-stack seismic gather data to obtain an inversion objective function.

[0172] Specifically, the inversion objective function is:

[0173]

[0174] Among them, u is the white noise coefficient, m0 is the initial model of elastic parameters, m is the elastic parameter model, G is the kernel function, α is the parameter for adjusting the model scale, d is the angle domain prestack seismic gather data, Φ i is the intermediate parameter, Ns is the data dimension of the elastic parameter model, and i' is the index number.

[0175] Step S2303, deriving the inversion objective function to obtain an inversion equation.

[0176] Specifically, according to the inversion objective function, the inversion equation is obtained:

[0177] (G T G+αΨ -1 +μQ)m=G T d+αΨ -1 m0

[0178] Among them, u is the white noise coefficient, m0 is the initial model of elastic parameters, m is the elastic parameter model, G is the kernel function, α is the adjustment model scale parameter, Ψ -1 is the covariance matrix, Q is the sparse constraint parameter, and d is the pre-stack seismic gather data in the angle domain.

[0179] Step S2304, preset the number of inversion iterations and the inversion error, iteratively solve the inversion equation, and obtain the intermediate frequency longitudinal wave impedance, intermediate frequency shear wave impedance, and intermediate frequency density.

[0180] Specifically, the preset inversion iteration number and inversion error are used to iteratively solve the inversion equation to obtain the intermediate frequency longitudinal wave impedance, intermediate frequency shear wave impedance, and intermediate frequency density, such as Figure 4 As shown, it includes steps S23041 to S23046:

[0181] Step S23041, preset the number of inversion iterations and the inversion error.

[0182] Step S23042, using a preset third formula to calculate the intermediate parameter.

[0183] Specifically, the preset third formula for calculating the intermediate parameter includes:

[0184]

[0185] Where Ψ is the covariance matrix of the model parameters.

[0186] The matrix S is:

[0187]

[0188] Where n represents the number of inversion parameters, l is the number of inversion sample points, i' is the index number, the dimension of the matrix S is 3*3Ns, and Ns is the data dimension of the elastic parameter model;

[0189] The matrix D1 is:

[0190]

[0191] The dimension of matrix D is 3Ns*3Ns, where Ns is the data dimension of the elastic parameter model.

[0192] Step S23043, using the preset fourth formula to calculate the sparse constraint parameters.

[0193] Specifically, the calculating the sparse constraint parameter by using the preset fourth formula includes:

[0194]

[0195] Where i is the number of iterations, i' is the number of indexes, Ns is the model size, Q is the sparse constraint parameter, Φ i’ is the intermediate parameter, m i is the elastic parameter model of the i-th iteration.

[0196] Step S23044, using the intermediate parameters and sparse constraint parameters to calculate the elastic parameter model using iterative least squares method and conjugate gradient method.

[0197] Step S23045, iterating the elastic parameter model until a preset number of inversion iterations and an inversion error are reached to obtain an inverted elastic parameter model.

[0198] Step S23046, performing exponential operation on the inverted elastic parameter model to obtain intermediate frequency longitudinal wave impedance, intermediate frequency shear wave impedance, and intermediate frequency density.

[0199] The high-resolution parameter acquisition module 204 is used to process the intermediate frequency longitudinal wave impedance, the intermediate frequency shear wave impedance, and the intermediate frequency density using a preset first formula to obtain high-frequency longitudinal wave impedance, high-frequency shear wave impedance, and high-frequency density. The formation medium resolution is higher, and the formation information reflected is more comprehensive and clear.

[0200] Specifically, the intermediate frequency longitudinal wave impedance, the intermediate frequency shear wave impedance, and the intermediate frequency density are processed using the collaborative Kriging method to obtain final high-precision longitudinal wave impedance, shear wave impedance, and density.

[0201] Specifically, the first formula is preset as the co-kriging method:

[0202]

[0203] X(u) is the estimated value of the primary variable at location u, Y(u) is the data of the covariate at location u, and λ represents the cokriging weighting coefficient.

[0204] The corresponding co-kriging equations only need to know the covariance function C of the main variable X X (h), the cross-covariance function C of the main variable X and the covariate Y xy (h), the latter can be approximated by the following model C XY (h) = βC X (h), where C X (0) and C Y (0) is the variance function of X and Y, P XY (0) is the linear correlation coefficient of the co-located X and Y data.

[0205] An embodiment of the present invention further provides a computer-readable storage medium, in which instructions are stored. When the instructions are executed on a terminal, the terminal executes the high-resolution seismic inversion method as described in the first embodiment.

[0206] An embodiment of the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the high-resolution seismic inversion method as described in the first embodiment is implemented.

[0207] In the above detailed description, various features are grouped together in a single embodiment to simplify the disclosure. This method of disclosure should not be interpreted as reflecting an intention that the embodiments of the claimed subject matter require more features than are clearly stated in each claim. On the contrary, as reflected in the appended claims, the invention is in a state of having less than all the features of the disclosed individual embodiments. Therefore, the appended claims are hereby expressly incorporated into the detailed description, with each claim standing on its own as a separate preferred embodiment of the invention.

[0208] The above description includes examples of one or more embodiments. Of course, it is not possible to describe all possible combinations of components or methods for the purpose of describing the above embodiments, but it should be recognized by those skilled in the art that the various embodiments may be further combined and arranged. Therefore, the embodiments described herein are intended to cover all such changes, modifications and variations that fall within the scope of protection of the appended claims. In addition, with respect to the term "comprising" used in the specification or claims, the word is encompassed in a manner similar to the term "including", as explained by "including," used as a transitional word in the claims. In addition, any term "or" used in the specification of the claims is intended to mean "non-exclusive or".

Claims

1. A high-resolution seismic inversion method, characterized in that: include: Obtain angle domain pre-stack seismic gather data, seismic layer data, seismic wavelet data, initial low-frequency longitudinal wave impedance, initial low-frequency shear wave impedance, and initial low-frequency density; Obtaining a reflection coefficient using the Fatti longitudinal wave reflection coefficient equation according to the angle domain pre-stack seismic gather data, the seismic layer data and the seismic wavelet data; An inversion objective function is established according to the initial low-frequency longitudinal wave impedance, the initial low-frequency shear wave impedance, the initial low-frequency density and the reflection coefficient to obtain an intermediate-frequency longitudinal wave impedance, an intermediate-frequency shear wave impedance and an intermediate-frequency density; The intermediate frequency longitudinal wave impedance, the intermediate frequency shear wave impedance, and the intermediate frequency density are processed using a preset first formula to obtain high frequency longitudinal wave impedance, high frequency shear wave impedance, and high frequency density.

2. The method according to claim 1, characterized in that The inversion objective function is established according to the initial low-frequency longitudinal wave impedance, the initial low-frequency shear wave impedance, the initial low-frequency density and the reflection coefficient to obtain the intermediate-frequency longitudinal wave impedance, the intermediate-frequency shear wave impedance and the intermediate-frequency density, including: Obtaining an initial model of elastic parameters according to the initial low-frequency longitudinal wave impedance, the initial low-frequency shear wave impedance, and the initial low-frequency density; Establishing an inversion objective function using the elastic parameter initial model and the reflection coefficient; Derivative the inversion objective function to obtain an inversion equation; The number of inversion iterations and the inversion error are preset, and the inversion equation is iteratively solved to obtain the intermediate frequency longitudinal wave impedance, the intermediate frequency shear wave impedance, and the intermediate frequency density.

3. The method according to claim 2, characterized in that The elastic parameter initial model is obtained according to the initial low-frequency longitudinal wave impedance, the initial low-frequency shear wave impedance, and the initial low-frequency density. The elastic parameter initial model m0 is: m0=(ln(Z P0 ),ln(Z S0 ),ln(ρ0)) Among them, Z P0 is the initial low-frequency longitudinal wave impedance, Z S0 is the initial low-frequency shear wave impedance, ρ0 is the initial low-frequency density, and m0 is the initial model of elastic parameters.

4. The method according to claim 2, characterized in that The method of establishing an inversion objective function by using the elastic parameter initial model and the reflection coefficient comprises: Using the seismic wavelet data and the reflection coefficient to simulate and synthesize pre-stack seismic angle gathers according to a preset second formula; An inversion objective function is obtained by using the elastic parameter initial model, the simulated synthesized pre-stack seismic angle gather and the acquired angle domain pre-stack seismic gather data.

5. The method according to claim 4, characterized in that The preset second formula is: d(θ)=W(t)*R pp (i) Where W(t) is the seismic wavelet data, R pp is the reflection coefficient, and d(θ) is the simulated synthetic pre-stack seismic angle gather.

6. The method according to claim 4, characterized in that The obtained inversion objective function is: Among them, u is the white noise coefficient, m0 is the initial model of elastic parameters, m is the elastic parameter model, G is the kernel function, α is the parameter for adjusting the model scale, d is the angle domain prestack seismic gather data, Φ i′ is the intermediate parameter, Ns is the data dimension of the elastic parameter model, and i' is the index number.

7. The method according to claim 6, characterized in that The inversion objective function is derived to obtain the inversion equation: According to the inversion objective function, the inversion equation is obtained: (G T G+αΨ -1 +μQ)m=G T d+aΨ -1 m0 Among them, u is the white noise coefficient, m0 is the initial model of elastic parameters, m is the elastic parameter model, G is the kernel function, α is the adjustment model scale parameter, Ψ -1 is the covariance matrix, Q is the sparse constraint parameter, and d is the pre-stack seismic gather data in the angle domain.

8. The method according to claim 7, characterized in that The preset inversion iteration number and inversion error are used to iteratively solve the inversion equation to obtain the intermediate frequency longitudinal wave impedance, intermediate frequency shear wave impedance, and intermediate frequency density, including: Preset the number of inversion iterations and inversion error; Calculating the intermediate parameter using a preset third formula; Calculating the sparse constraint parameter using a preset fourth formula; Calculating the elastic parameter model by using the intermediate parameters and sparse constraint parameters using iterative least square method and conjugate gradient method; Iterating the elastic parameter model until a preset number of inversion iterations and an inversion error are reached to obtain an inverted elastic parameter model; The inverted elastic parameter model is subjected to exponential operation to obtain the intermediate frequency longitudinal wave impedance, intermediate frequency shear wave impedance and intermediate frequency density.

9. The method according to claim 8, characterized in that The preset third formula for calculating the intermediate parameter includes: Among them, Ψ is the covariance matrix of model parameters, Φ i′ is the intermediate parameter; The matrix S is: Where n represents the number of inversion parameters, l is the number of inversion sample points, i' is the index number, the dimension of the matrix S is 3*3Ns, and Ns is the data dimension of the elastic parameter model; The matrix D1 is: The dimension of matrix D is 3Ns*3Ns, where Ns is the data dimension of the elastic parameter model.

10. The method according to claim 8, characterized in that The calculating the sparse constraint parameter by using a preset fourth formula includes: Among them, i is the number of iterations, i' is the index number, Ns is the data dimension of the elastic parameter model, Q is the sparse constraint parameter, Φ i’ is the intermediate parameter, m i is the elastic parameter model of the i-th iteration.

11. The method according to claim 8, wherein the elastic parameter model is calculated by using the intermediate parameters and the sparse constraint parameters using the iterative least square method and the conjugate gradient method, comprising: m i+1 =(G T G+μQ(i)+αΨ -1 ) -1 (Gd+αΨ -1 m0) Among them, μ is the white noise parameter, α is the adjustment model scale parameter, and i is the number of iterations.

12. The method according to claim 1, wherein the first preset formula is: in, X(u) is the estimated value of the primary variable at location u, Y(u) is the data of the covariate at location u, and λ represents the cokriging weighting coefficient.

13. A high-resolution seismic inversion device, characterized in that: It includes parameter acquisition module, reflection coefficient acquisition module, inversion equation establishment module and high-resolution parameter acquisition module, among which: The parameter acquisition module is used to acquire angle domain pre-stack seismic gather data, seismic layer data, seismic wavelet data, initial low-frequency longitudinal wave impedance, initial low-frequency shear wave impedance, and initial low-frequency density; The reflection coefficient acquisition module is used to obtain the reflection coefficient using the Fatti longitudinal wave reflection coefficient equation according to the angle domain pre-stack seismic gather data, the seismic layer data and the seismic wavelet data; The inversion equation establishment module is used to establish an inversion objective function according to the initial low-frequency longitudinal wave impedance, the initial low-frequency shear wave impedance, the initial low-frequency density and the reflection coefficient to obtain the intermediate-frequency longitudinal wave impedance, the intermediate-frequency shear wave impedance and the intermediate-frequency density; The high-resolution parameter acquisition module is used to process the intermediate-frequency longitudinal wave impedance, the intermediate-frequency shear wave impedance, and the intermediate-frequency density using a preset first formula to obtain high-frequency longitudinal wave impedance, high-frequency shear wave impedance, and high-frequency density.

14. A computer-readable storage medium, wherein instructions are stored in the computer-readable storage medium, and when the instructions are executed on a terminal, the terminal executes the method for establishing as claimed in any one of claims 1 to 12.

15. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 12 when executing the computer program.

Images