A time-frequency domain pre-stack seismic inversion method and system of a stratum attenuation factor

Abstract

The present application is suitable for oil and gas geophysical exploration technical field, provide a kind of stratum attenuation factor time-frequency domain prestack seismic inversion method and system, the method includes the following steps: construct about longitudinal, transverse wave attenuation factor frequency domain AVO formula, propose the collaborative inversion strategy of joint time-frequency domain seismic data real part and imaginary part information;Input partial stack seismic data and carry out spectral decomposition, obtain time-frequency domain seismic data;According to the real part of time-frequency domain data, the frequency variable longitudinal, transverse wave modulus reflectivity is calculated;According to the imaginary part of time-frequency domain data, construct the time-frequency spectrum of imaginary part reflectivity difference;With the real part inversion result as prior information, in combination with the spectrum of seismic wavelet, the frequency variable longitudinal, transverse wave attenuation factor frequency gradient attribute is inverted, and the longitudinal, transverse wave attenuation factor is calculated by frequency integration.This method uses the real part and imaginary part information of time-frequency domain seismic data, and provides an effective method for calculating stratum attenuation factor by "real part-imaginary part" hierarchical inversion strategy.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of oil and gas geophysical exploration, and particularly relates to a time-frequency domain pre-stack seismic inversion method and system for stratum attenuation factors. BACKGROUND

[0002] Due to the viscoelasticity of underground medium, energy loss of seismic wave occurs in the propagation process, which can be characterized by the attenuation factor Q -1 The seismic wave attenuation factor is closely related to the fluid properties, pores and cracks of rocks and has been widely applied to reservoir identification and physical property parameter evaluation. Therefore, accurate calculation of the attenuation factor is of great significance for reservoir prediction.

[0003] The conventional Q value estimation methods (such as statistical analysis method, time domain method, frequency domain method and time-frequency domain method) generally have limited precision and insufficient resolution, which are difficult to meet the requirements of high-resolution reservoir prediction. In comparison, the pre-stack seismic inversion method based on the quantitative relationship between the reflection coefficient and the Q value in the viscoelastic medium has obvious advantages in resolution. However, most of these methods fail to accurately depict the frequency variation characteristics of the attenuation factor. Rock physics experiments and theoretical modeling show that the attenuation caused by the propagation of seismic wave in the medium has significant frequency dependence. Therefore, it is particularly necessary to develop an inversion method capable of revealing the variation law of the attenuation factor with frequency.

[0004] Complex seismic trace analysis is a common method in seismic attribute extraction. Existing researches introduce the imaginary part information of seismic signals for the estimation of the attenuation factor. However, in the field of pre-stack seismic inversion, complex seismic data have not been fully utilized. In fact, the imaginary part of the reflection coefficient not only causes phase shift, but also may lead to wavelet polarity inversion, and the information contained therein should be fully valued and utilized.

[0005] Therefore, it is urgent to develop a time-frequency domain pre-stack seismic inversion method for stratum attenuation factors, which fully utilizes the real and imaginary part information of seismic data in the time-frequency domain, calculates the stratum attenuation factors through the hierarchical inversion strategy of "real part-imaginary part" to characterize the attenuation characteristics of reservoir medium and improves the accuracy and reliability of reservoir prediction. SUMMARY

[0006] The application aims to provide a time-frequency domain pre-stack seismic inversion method for stratum attenuation factors, which aims to solve the above technical problems.

[0007] The application is implemented as follows: a time-frequency domain pre-stack seismic inversion method for stratum attenuation factors, comprising the following steps:

[0008] Constructing the frequency domain AVO formula about the attenuation factors of P and S waves, and proposing a collaborative inversion strategy based on the joint time-frequency domain pre-stack seismic data real and imaginary part information based on the frequency domain AVO formula;

[0009] The input partial stack seismic data, extracts the seismic wavelet and calculates its spectrum;

[0010] Based on the inversion spectrum decomposition method, the partial stack seismic data is spectrum decomposed to obtain the time-frequency domain seismic data;

[0011] Based on the deconvolution method, the wavelet interference effect contained in the real part information of the time-frequency domain seismic data is removed to obtain the real part reflection coefficient, and the frequency-dependent longitudinal and transverse wave modulus reflectivity is calculated;

[0012] According to the imaginary part information of the time-frequency domain seismic data, the time-frequency spectrum of the imaginary part reflection coefficient difference is constructed;

[0013] Taking the frequency-dependent longitudinal and transverse wave modulus reflectivity as prior information, and combining the spectrum of the seismic wavelet, the frequency-dependent longitudinal and transverse wave attenuation factor frequency gradient attribute is calculated based on the FS-AVO inversion method;

[0014] According to the frequency-dependent longitudinal and transverse wave attenuation factor frequency gradient attribute, the prediction result of the formation attenuation factor is determined.

[0015] Further, the frequency domain AVO formula is specifically as follows:

[0016] ;

[0017] In the formula, is the reflection coefficient; θ is the incidence angle, is the frequency; ΔM and Δμ are the longitudinal and transverse wave modulus of the uniform perturbed medium, and M and μ are the longitudinal and transverse wave modulus of the corresponding uniform background medium; 、 A p (ω) and A s (ω) are the frequency-dependent longitudinal and transverse wave attenuation factors; γ sat is the ratio of the saturated rock longitudinal and transverse wave velocity; j is the imaginary unit.

[0018] Further, the joint time-frequency domain pre-stack seismic data real part and imaginary part information collaborative inversion strategy is based on the frequency domain AVO formula of the longitudinal and transverse wave attenuation factors, and the inversion theoretical formula of the real part and the imaginary part is respectively established; The inversion theoretical formula of the real part and the imaginary part is specifically as follows:

[0019] ;

[0020] ;

[0021] ;

[0022] In the formula, is a real part reflection coefficient; is a real part reflection coefficient; p is a real part reflection coefficient; s are respectively longitudinal and transverse wave attenuation factors; ω0 is a reference frequency.

[0023] Further, the frequency-variable longitudinal and transverse wave modulus reflectivity is obtained by sparse pulse deconvolution of the real part of the time-frequency domain seismic data to obtain an accurate reflection coefficient sequence, and is calculated by applying an inversion theory formula to the real part.

[0024] Further, the FS-AVO inversion method is a frequency scanning method, in which a frequency window is continuously moved in the seismic frequency band with a variable reference frequency as the center, and then the frequency-variable longitudinal and transverse wave attenuation factor frequency gradient attribute is calculated.

[0025] Further, the step of determining the prediction result of the formation attenuation factor according to the frequency-variable longitudinal and transverse wave attenuation factor frequency gradient attribute specifically comprises:

[0026] integrating and calculating the frequency-variable longitudinal and transverse wave attenuation factor frequency gradient attribute to obtain a frequency-variable longitudinal and transverse wave attenuation factor;

[0027] taking a peak value of the frequency-variable longitudinal and transverse wave attenuation factor as the final prediction result.

[0028] Another object of the present application is to provide a time-frequency domain pre-stack seismic inversion system of a formation attenuation factor, which is used for realizing the time-frequency domain pre-stack seismic inversion method of the formation attenuation factor, and comprises:

[0029] a cooperative inversion strategy determination module, which is used for constructing a frequency domain AVO formula about the longitudinal and transverse wave attenuation factors, and proposing a cooperative inversion strategy of joint time-frequency domain pre-stack seismic data real part and imaginary part information based on the frequency domain AVO formula;

[0030] a spectrum extraction module, which is used for inputting partial stack seismic data, extracting a seismic wavelet and calculating a spectrum thereof;

[0031] a time-frequency domain seismic data acquisition module, which is used for performing spectrum decomposition on the partial stack seismic data based on an inversion spectrum decomposition method to obtain time-frequency domain seismic data;

[0032] a real part inversion module, which is used for removing a wavelet interference effect contained in the real part information in the time-frequency domain seismic data based on a deconvolution method to obtain a real part reflection coefficient, and calculating a frequency-variable longitudinal and transverse wave modulus reflectivity;

[0033] The imaginary part inversion module is used for constructing a time-frequency spectrum of imaginary part reflection coefficient difference according to imaginary part information of the time-frequency domain seismic data; the frequency-variable longitudinal wave and transverse wave modulus reflectivity is taken as prior information, and the frequency-variable longitudinal wave and transverse wave attenuation factor frequency gradient attribute is calculated based on the FS-AVO inversion method in combination with the spectrum of the seismic wavelet;

[0034] The prediction result determination module is used for determining the prediction result of the formation attenuation factor according to the frequency-variable longitudinal wave and transverse wave attenuation factor frequency gradient attribute.

[0035] The time-frequency domain prestack seismic inversion method for the formation attenuation factor provided by the application proposes a collaborative inversion strategy of combining the real part and imaginary part information of the time-frequency domain prestack seismic data by expanding the AVO approximation formula containing the longitudinal wave and transverse wave attenuation factors to the frequency domain. BRIEF DESCRIPTION OF DRAWINGS

[0036] Figure 1 The flowchart of the time-frequency domain prestack seismic inversion method for the formation attenuation factor provided by the embodiment of the application.

[0037] Figure 2 The partial stack seismic profile of the well A and the well B is shown in the figure; in the figure, (a) is a near-angle (0-8°) stack profile; (b) is a middle-angle (8-16°) stack profile; and (c) is a far-angle (16-24°) stack profile.

[0038] Figure 3 The profile of the longitudinal wave and transverse wave attenuation factor frequency gradient attribute at different frequencies is shown in the figure; in the figure, (a) is the longitudinal wave attenuation factor frequency gradient attribute; and (b) is the transverse wave attenuation factor frequency gradient attribute.

[0039] Figure 4 The profile of the longitudinal wave and transverse wave attenuation factor at different frequencies is shown in the figure; in the figure, (a) is the longitudinal wave attenuation factor; and (b) is the transverse wave attenuation factor.

[0040] Figure 5The figures show the peak values ​​of the P-wave and S-wave attenuation factors as a function of frequency; in the figures, (a) shows the peak value of the P-wave attenuation factor as a function of frequency; and (b) shows the peak value of the S-wave attenuation factor as a function of frequency. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0042] like Figure 1 As shown, in one embodiment of the present invention, a time-frequency domain pre-stack seismic inversion method for formation attenuation factors is provided, comprising the following steps:

[0043] S1. Construct frequency domain AVO formulas for P-wave and S-wave attenuation factors, and propose a collaborative inversion strategy based on the frequency domain AVO formulas that combines real and imaginary information from pre-stack seismic data in the time and frequency domains.

[0044] S2. Input partially superimposed seismic data, extract seismic wavelets and calculate their spectrum;

[0045] S3. Based on the Inverse Spectral Decomposition (ISD) method, perform spectral decomposition on some stacked seismic data to obtain time-frequency domain seismic data;

[0046] S4. Based on the deconvolution method, the wavelet interference effect contained in the real part information of the time-frequency domain seismic data is removed to obtain the real part reflection coefficient, and the frequency-varying P-wave and S-wave modulus reflectivity is calculated.

[0047] S5. Based on the imaginary part information of the time-frequency domain seismic data, construct the time spectrum of the imaginary part reflection coefficient difference;

[0048] S6. Using the frequency-varying P-wave and S-wave modulus reflectivity as prior information, and combining it with the seismic wavelet spectrum, calculate the frequency gradient properties of the frequency-varying P-wave and S-wave attenuation factors based on the FS-AVO inversion method.

[0049] S7. Based on the frequency gradient properties of the frequency-varying P-wave and S-wave attenuation factors, determine the predicted results of the formation attenuation factor.

[0050] In a preferred embodiment of the present invention, the frequency domain AVO formula is as follows:

[0051] ;

[0052] In the formula, θ is the reflection coefficient; θ is the angle of incidence. is frequency; ΔM, Δμ are respectively the longitudinal and transverse wave modulus of the uniform perturbation medium, and M, μ are respectively the longitudinal and transverse wave modulus of the corresponding uniform background medium; 、 are respectively the frequency-dependent longitudinal and transverse wave modulus reflectivity; p (ω) and A s (ω) are respectively the frequency-dependent longitudinal and transverse wave attenuation factor; γ sat is the ratio of the longitudinal and transverse wave velocity of the saturated rock; j is the imaginary unit.

[0053] In a preferred embodiment of the present application, the joint time-frequency domain pre-stack seismic data real and imaginary part information collaborative inversion strategy (including the real part inversion and imaginary part inversion stages) is realized based on the frequency domain AVO formula of the longitudinal and transverse wave attenuation factors, and the inversion theoretical formula of the real part and the imaginary part is respectively established; the inversion theoretical formula of the real part and the imaginary part is specifically as follows:

[0054] ;

[0055] ;

[0056] ;

[0057] In the formula, is the real part reflection coefficient; is the imaginary part reflection coefficient; A p and A s are respectively the longitudinal and transverse wave attenuation factor; ω0 is the reference frequency.

[0058] In step S2, the spectrum of the extracted seismic wavelet is applied to the imaginary part inversion stage to eliminate the superposition effect of the wavelet.

[0059] In a preferred embodiment of the present application, the frequency-dependent longitudinal and transverse wave modulus reflectivity is obtained by sparse pulse deconvolution of the real part of the time-frequency domain seismic data to obtain the accurate reflection coefficient sequence, and is applied to the inversion theoretical formula of the real part to obtain.

[0060] In a preferred embodiment of the present application, the FS-AVO inversion method is realized by frequency scanning, and the frequency window is continuously moved in the seismic frequency band with the changing reference frequency as the center, and then the frequency-dependent longitudinal and transverse wave attenuation factor frequency gradient attribute is calculated.

[0061] In a preferred embodiment of the present application, according to the frequency-dependent longitudinal and transverse wave attenuation factor frequency gradient attribute, the step of determining the prediction result of the formation attenuation factor specifically includes:

[0062] The frequency-dependent longitudinal and transverse wave attenuation factor frequency gradient attribute is integrated and calculated to obtain the frequency-dependent longitudinal and transverse wave attenuation factor.

[0063] The peak value of the frequency-dependent longitudinal and transverse wave attenuation factors is taken as the final prediction result.

[0064] In addition, the prediction result can be analyzed and verified in combination with the logging interpretation result of the research area. If the high-value abnormal position of the peak value of the frequency-dependent longitudinal and transverse wave attenuation factors is consistent with the position of the gas layer in the well, it indicates that the proposed inversion method is effective and reliable in reservoir identification.

[0065] In practical application, the principle and implementation process of the above-mentioned time-frequency domain prestack seismic inversion method of the formation attenuation factor are as follows:

[0066] Considering the viscoelasticity of the underground medium, based on the perturbation method, the AVO approximate formula containing the longitudinal and transverse wave attenuation factors and is derived based on the Born integral and the steady-state approximation assumption:

[0067] (1);

[0068] Where θ is the incidence angle; ΔM, Δμ and Δρ are the longitudinal, transverse wave modulus and density of the uniform perturbation medium, respectively. M, μ and ρ correspond to the longitudinal, transverse wave modulus and density of the uniform background medium, respectively; γ sat is the ratio of the longitudinal and transverse wave velocities.

[0069] The above formula (1) is extended to the frequency domain:

[0070] (2);

[0071] Where , are the frequency-dependent longitudinal and transverse wave attenuation factors, respectively.

[0072] The reflection coefficient can be expressed as the sum of the real part and the imaginary part:

[0073] (3);

[0074] (4);

[0075] (5);

[0076] Where:

[0077] (6);

[0078] (7);

[0079] (8);

[0080] (9);

[0081] The real part inversion formula is established according to formula (4):

[0082] (10);

[0083] Wherein:

[0084] (11);

[0085] (12);

[0086] (13);

[0087] The formula (10) is solved by using the least square method, and the frequency-varying longitudinal and transverse wave modulus reflectivity And .

[0088] The first order Taylor expansion is made to formula (5) at the reference frequency ω0:

[0089] (14);

[0090] The formula (14) can be further arranged as:

[0091] (15);

[0092] Wherein, the imaginary part reflection coefficient difference ; the frequency gradient attribute of the longitudinal and transverse wave attenuation factors D Ap And D As Is respectively expressed as:

[0093] (16);

[0094] The time-frequency spectrum of the reflection coefficient difference in formula (15) is used as the input of the imaginary part inversion, and the spectrum of the seismic wavelet is introduced to eliminate the superposition effect of the wavelet, and the imaginary part inversion formula is established as follows:

[0095] (17);

[0096] Wherein:

[0097] (18);

[0098] (19); ​​

[0099] (20);

[0100] Solve equation (17) by using the damped least square method:

[0101] (21);

[0102] In the formula, ε is the damping factor, is the unit matrix.

[0103] Most of the existing frequency-dependent inversion methods use constant reference frequency ω0 and fixed frequency band range for inversion calculation. Considering the rationality of Taylor expansion, the imaginary part inversion needs to be carried out in a relatively narrow frequency band. The embodiments of the present application calculate the frequency gradient attributes of the frequency-dependent longitudinal and transverse wave attenuation factors in the moving frequency window centered on different reference frequencies ω0 and .

[0104] The frequency integral calculation is carried out on and to obtain the frequency-dependent longitudinal and transverse wave attenuation factors A p (ω) and A s (ω) with more practical significance:

[0105] (22);

[0106] (23);

[0107] Wherein, k is the index, and n is the total number of frequency points.

[0108] Finally, the peak values of the frequency-dependent longitudinal and transverse wave attenuation factors A p-max and A s-max are extracted and applied to reservoir prediction.

[0109] The embodiment of the present application mainly researches on formation attenuation factor, and develops a time-frequency domain prestack seismic inversion method of formation attenuation factor. That is, by constructing the frequency domain AVO formula (formula 2) about longitudinal and transverse wave attenuation factors, a collaborative inversion strategy of combining the real part and the imaginary part information of the time-frequency domain prestack seismic data is proposed. First, the time-frequency domain seismic data with high resolution is obtained by spectrum decomposition of the prestack seismic data through the inversion spectrum decomposition method, and the frequency-dependent longitudinal and transverse wave modulus reflectivity (formula 10) is obtained by using the real part of the time-frequency domain data inversion; then, the calculation result is used as prior information, the frequency-dependent longitudinal and transverse wave attenuation factor frequency gradient attribute (formula 17) is calculated by using the imaginary part of the time-frequency domain data by applying the FS-AVO inversion method, and the frequency-dependent longitudinal and transverse wave attenuation factors (formula 22-23) are further calculated by integration, and finally the peak attenuation factor is extracted and applied to reservoir prediction. The above method is applied to actual seismic data, and the inversion result is shown in Figures 3-5 .

[0110] wherein, Figure 2 Fig. 1 shows the partial stack seismic profiles of wells A and B; Figure 2 Fig. 1 shows the partial stack seismic profiles of wells A and B;

[0111] Figure 3 Fig. 2 shows the imaginary part inversion result, the profiles of longitudinal and transverse wave attenuation factor frequency gradient attribute at different frequencies; Figure 3 Fig. 2 shows the imaginary part inversion result, the profiles of longitudinal and transverse wave attenuation factor frequency gradient attribute at different frequencies; Ap Fig. 2 shows the imaginary part inversion result, the profiles of longitudinal and transverse wave attenuation factor frequency gradient attribute at different frequencies; As Fig. 2 shows the imaginary part inversion result, the profiles of longitudinal and transverse wave attenuation factor frequency gradient attribute at different frequencies; Ap Fig. 2 shows the imaginary part inversion result, the profiles of longitudinal and transverse wave attenuation factor frequency gradient attribute at different frequencies; As Fig. 2 shows the imaginary part inversion result, the profiles of longitudinal and transverse wave attenuation factor frequency gradient attribute at different frequencies;

[0112] Figure 4 Fig. 3 shows the profiles of longitudinal and transverse wave attenuation factors at different frequencies; Figure 4 Fig. 3 shows the profiles of longitudinal and transverse wave attenuation factors at different frequencies; p Fig. 3 shows the profiles of longitudinal and transverse wave attenuation factors at different frequencies; s Fig. 3 shows the profiles of longitudinal and transverse wave attenuation factors at different frequencies; Ap Fig. 3 shows the profiles of longitudinal and transverse wave attenuation factors at different frequencies; As Fig. 3 shows the profiles of longitudinal and transverse wave attenuation factors at different frequencies; p Fig. 3 shows the profiles of longitudinal and transverse wave attenuation factors at different frequencies; s Fig. 3 shows the profiles of longitudinal and transverse wave attenuation factors at different frequencies; p Fig. 3 shows the profiles of longitudinal and transverse wave attenuation factors at different frequencies; sThe high value abnormal response of A s The ratio of A p The resolution of A

[0113] Figure 5 The peak value A p-max and A s-max of the profile of the frequency-variable longitudinal wave attenuation factor and the frequency-variable transverse wave attenuation factor are shown. Figure 5 The peak value A p-max of the frequency-variable longitudinal wave attenuation factor is (a) and the peak value A s-max of the frequency-variable transverse wave attenuation factor is (b). The results show that the high value abnormal response of A p-max and A s-max is highly consistent with the interpreted gas layer position in the well. These results prove that the peak value A p-max and A s-max of the frequency-variable longitudinal wave attenuation factor and the frequency-variable transverse wave attenuation factor calculated by the method proposed in the embodiment of the present application can effectively identify the favorable reservoir and provide technical support for the exploration and development of the reservoir.

[0114] In another embodiment of the present application, a time-frequency domain pre-stack seismic inversion system of formation attenuation factors is also provided for implementing the time-frequency domain pre-stack seismic inversion method of formation attenuation factors, which comprises:

[0115] A collaborative inversion strategy determination module is configured to construct a frequency domain AVO formula about the longitudinal wave attenuation factor and the transverse wave attenuation factor, and propose a collaborative inversion strategy of jointly using the real part and the imaginary part information of the time-frequency domain pre-stack seismic data based on the frequency domain AVO formula;

[0116] A spectrum extraction module is configured to input the partial stack seismic data, extract the seismic wavelet and calculate the spectrum thereof;

[0117] A time-frequency domain seismic data acquisition module is configured to perform spectrum decomposition on the partial stack seismic data based on the inversion spectrum decomposition method to obtain the time-frequency domain seismic data;

[0118] A real part inversion module is configured to remove the wavelet interference effect contained in the real part information of the time-frequency domain seismic data based on the deconvolution method to obtain the real part reflection coefficient and calculate the frequency-variable longitudinal wave modulus reflectivity and the frequency-variable transverse wave modulus reflectivity;

[0119] An imaginary part inversion module is configured to construct a time-frequency spectrum of the imaginary part reflection coefficient difference according to the imaginary part information of the time-frequency domain seismic data, take the frequency-variable longitudinal wave modulus reflectivity and the frequency-variable transverse wave modulus reflectivity as the prior information, combine the spectrum of the seismic wavelet, and calculate the frequency-variable longitudinal wave attenuation factor frequency gradient attribute and the frequency-variable transverse wave attenuation factor frequency gradient attribute based on the FS-AVO inversion method;

[0120] A prediction result determination module is configured to determine the prediction result of the formation attenuation factors according to the frequency-variable longitudinal wave attenuation factor frequency gradient attribute and the frequency-variable transverse wave attenuation factor frequency gradient attribute.

[0121] It should be noted that the above modules can be implemented in the form of a computer program, which can run on a computer device, and the computer program stored in the memory of the computer device can constitute the modules to enable the processor to execute the steps of the above method.

[0122] It should be understood that, although each step in the flowchart of each embodiment of the present application is displayed in sequence according to the arrow, these steps are not necessarily executed in sequence according to the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other orders. Moreover, at least part of the steps in each embodiment can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these sub-steps or stages is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or sub-steps or stages of other steps.

[0123] It can be understood by those skilled in the art that all or part of the above-mentioned embodiment methods can be completed by a computer program instructing related hardware. The above-mentioned program can be stored in a non-volatile computer readable storage medium, and the program can include the above-mentioned flow of each embodiment of the method when executed.

[0124] The above embodiments only express several embodiments of the present application, which are described in detail and specifically, but should not be understood as a limitation on the scope of the patent of the present application. It should be noted that, for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are all within the scope of protection of the present application. Therefore, the scope of protection of the patent of the present application should be subject to the appended claims.

Claims

1. A time-frequency domain prestack seismic inversion method for attenuation factors of a formation, characterized in that, The method comprises the following steps: constructing a frequency domain AVO formula about P-wave and S-wave attenuation factors, and proposing a collaborative inversion strategy of combining real and imaginary parts of pre-stack seismic data in time-frequency domain based on the frequency domain AVO formula; inputting partial stack seismic data, extracting a seismic wavelet and calculating a spectrum thereof; performing spectrum decomposition on the partial stack seismic data based on an inversion spectrum decomposition method to obtain time-frequency domain seismic data; removing wavelet interference effects contained in the real part information of the time-frequency domain seismic data based on a deconvolution method to obtain real part reflection coefficients and calculate frequency-variable P-wave and S-wave modulus reflectivities; constructing a time-frequency spectrum of imaginary part reflection coefficient difference according to the imaginary part information of the time-frequency domain seismic data; taking the frequency-variable P-wave and S-wave modulus reflectivities as prior information and combining the spectrum of the seismic wavelet to calculate frequency-variable P-wave and S-wave attenuation factor frequency gradient attributes based on an FS-AVO inversion method; determining a prediction result of the formation attenuation factor according to the frequency-variable P-wave and S-wave attenuation factor frequency gradient attributes; the frequency domain AVO formula is specifically as follows: ； where, is the reflection coefficient; θ is the incidence angle, is the frequency; ΔM, Δμ are the perturbation of the longitudinal and shear wave moduli of the medium, respectively, and M, μ are the corresponding moduli of the background medium; , are the frequency-dependent longitudinal and shear wave modulus reflectivities, respectively; A p (ω) and A s (ω) are the frequency-dependent longitudinal and shear wave attenuation factors, respectively; γ sat is the ratio of the longitudinal and shear wave velocities of the saturated rock; j is the imaginary unit. the collaborative inversion strategy of combining real and imaginary parts of pre-stack seismic data in time-frequency domain is realized by respectively establishing inversion theoretical formulas of the real and imaginary parts based on the frequency domain AVO formula about P-wave and S-wave attenuation factors; the inversion theoretical formulas of the real and imaginary parts are specifically as follows: ； ； ； where is the real part of the reflection coefficient; is the imaginary part of the reflection coefficient; A p and A s are the longitudinal and transverse wave attenuation factors, respectively; and ω0is the reference frequency.

2. The time-frequency domain pre-stack seismic inversion method of formation attenuation factor according to claim 1, characterized in that, the frequency-variable P-wave and S-wave modulus reflectivities are obtained by performing sparse pulse deconvolution on the real part of the time-frequency domain seismic data to obtain an accurate reflection coefficient sequence and applying the accurate reflection coefficient sequence to the inversion theoretical formula of the real part to obtain the frequency-variable P-wave and S-wave modulus reflectivities.

3. The time-frequency domain pre-stack seismic inversion method for formation attenuation factors according to claim 1, wherein, the FS-AVO inversion method is to continuously move a frequency window in a seismic frequency band with a variable reference frequency as the center by means of frequency scanning, and then calculate the frequency-variable P-wave and S-wave attenuation factor frequency gradient attributes.

4. The time-frequency domain pre-stack seismic inversion method of formation attenuation factor according to claim 1, characterized in that, The step of determining a prediction result of the formation attenuation factor according to the frequency-variable P-wave and S-wave attenuation factor frequency gradient attributes specifically comprises: integrally calculating the frequency-variable P-wave and S-wave attenuation factor frequency gradient attributes to obtain frequency-variable P-wave and S-wave attenuation factors; taking a peak value of the frequency-variable P-wave and S-wave attenuation factors as the final prediction result.

5. A time-frequency domain prestack seismic inversion system for attenuating factors of a formation, for implementing the time-frequency domain prestack seismic inversion method of any one of claims 1-4, characterized in that, It comprises: a collaborative inversion strategy determination module for constructing a frequency domain AVO formula about P-wave and S-wave attenuation factors and proposing a collaborative inversion strategy of combining real and imaginary parts of pre-stack seismic data in time-frequency domain based on the frequency domain AVO formula; a spectrum extraction module for inputting partial stack seismic data, extracting a seismic wavelet and calculating a spectrum thereof; a time-frequency domain seismic data acquisition module for performing spectrum decomposition on the partial stack seismic data based on an inversion spectrum decomposition method to obtain time-frequency domain seismic data; a real part inversion module for removing wavelet interference effects contained in the real part information of the time-frequency domain seismic data based on a deconvolution method to obtain real part reflection coefficients and calculate frequency-variable P-wave and S-wave modulus reflectivities; an imaginary part inversion module for constructing a time-frequency spectrum of imaginary part reflection coefficient difference according to the imaginary part information of the time-frequency domain seismic data; taking the frequency-variable P-wave and S-wave modulus reflectivities as prior information and combining the spectrum of the seismic wavelet to calculate frequency-variable P-wave and S-wave attenuation factor frequency gradient attributes based on an FS-AVO inversion method; A prediction result determining module is configured to determine a prediction result of the formation attenuation factor according to the frequency-variable longitudinal wave and transverse wave attenuation factor frequency gradient attribute.

Images