CN112147681A - Pre-stack inversion method and system based on gamma _ Zoeppritz equation - Google Patents
Pre-stack inversion method and system based on gamma _ Zoeppritz equation Download PDFInfo
- Publication number
- CN112147681A CN112147681A CN201910571021.1A CN201910571021A CN112147681A CN 112147681 A CN112147681 A CN 112147681A CN 201910571021 A CN201910571021 A CN 201910571021A CN 112147681 A CN112147681 A CN 112147681A
- Authority
- CN
- China
- Prior art keywords
- inversion
- low
- matrix
- constructing
- natural logarithm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
- G01V1/28—Processing seismic data, e.g. analysis, for interpretation, for correction
- G01V1/30—Analysis
- G01V1/306—Analysis for determining physical properties of the subsurface, e.g. impedance, porosity or attenuation profiles
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V2210/00—Details of seismic processing or analysis
- G01V2210/60—Analysis
- G01V2210/62—Physical property of subsurface
- G01V2210/624—Reservoir parameters
- G01V2210/6242—Elastic parameters, e.g. Young, Lamé or Poisson
Abstract
A prestack inversion method and system based on gamma _ Zoeppritz equation are disclosed. The method can comprise the following steps: step 1: constructing a low-frequency soft constraint term; step 2: constructing a seismic data fitting difference term; and step 3: constructing a reflection coefficient sparse constraint term; and 4, step 4: constructing an inversion target function according to the seismic data fitting difference term, the low-frequency soft constraint term and the reflection coefficient sparse constraint term; and 5: and obtaining an elastic parameter inversion result according to the inversion target function. According to the method, an accurate relation between an angle reflection coefficient and a three-elastic-parameter natural logarithm is established by deducing a gamma _ Zoeppritz equation, a first-order difference matrix and a reflection coefficient sparse constraint term and a low-frequency soft constraint term represented by a three-variable Cauchy distribution are introduced to construct an inversion target function, and finally a high-resolution and transversely continuous inversion result is obtained by utilizing a generalized linear inversion mode.
Description
Technical Field
The invention relates to the field of oil and gas exploration, in particular to a prestack inversion method and system based on a gamma _ Zoeppritz equation.
Background
The prestack seismic data contains abundant offset information, and the change of amplitude along with offset/incidence angle reveals lithology change of underground medium and fluid composition change in pores. Therefore, a plurality of rock elastic parameters can be extracted from the angle part stacked seismic data by utilizing the prestack AVA synchronous inversion, wherein the longitudinal and transverse wave velocity is more sensitive to the change of reservoir lithology and fluid in pores and is a 'hydrocarbon indicator' preferred by seismic interpreters.
However, conventional prestack elastic parameter inversion needs to invert longitudinal wave velocity \ longitudinal wave impedance, transverse wave velocity \ transverse wave impedance first, and then convert the longitudinal wave velocity and the transverse wave impedance into a longitudinal-transverse wave velocity ratio, and indirect conversion of elastic parameters causes error accumulation, so that inversion accuracy of the longitudinal-transverse wave velocity ratio is affected. In addition, the conventional prestack elastic parameter inversion is mostly based on an approximate formula of a Zoeppritz equation, and the conditions of medium and small angle hypothesis, slow change of elastic parameters in the vertical direction and the like need to be met, so that the application range and the inversion accuracy of the prestack elastic parameter inversion are further limited.
The prior art has conducted intensive research on the inversion problem of prestack seismic. The prestack AVA inversion can be traced back to the weighted stacking method proposed by Smith at the earliest, the method belongs to band-limited inversion, the band-limited effect of wavelets is not considered, and the inversion result is still the elastic parameter reflectivity. Introducing Bayes theory into pre-stack seismic inversion, and providing an analytic solution of mean and variance of model parameter posterior distribution by assuming that likelihood function of seismic data and prior distribution of model parameters are both compliant with multivariate Gaussian distribution, and indicating that solution of deterministic inversion is expected of model parameter posterior distribution; and the solution of random inversion can be realized by sampling from the posterior probability through MCMC and other technologies. The vertical resolution of inversion can be further improved by providing the 'long tail' distribution relative to the Gaussian distribution, the correlation of three parameters is removed by means of the decorrelation technology of the parameter covariance matrix, the unsuitability of pre-stack inversion is improved, and pre-stack binomial and trinomial inversion is respectively realized aiming at Lp norm distribution, univariate Cauchy distribution and Huber distribution in the long tail distribution. By introducing a rock physics empirical formula as a constraint, the stability of pre-stack inversion is further improved, and a pre-stack three-parameter synchronous inversion algorithm based on an angle gather is realized, wherein the algorithm is a core algorithm of a pre-stack inversion module of HRS software. Bayesian pre-stack inversion is further developed, and the robustness of an inversion result is improved by introducing rock physical relationship constraint and point constraint. By introducing multivariate skewed Gaussian distribution as prior constraint of three-parameter natural logarithm, the precision of Bayesian three-parameter synchronous inversion is further improved; the three-parameter reflectivity obeys the three-variable Cauchy distribution by considering the cross correlation among the three parameters under the condition of geological background, and the precision of pre-stack inversion is further improved relative to the single-variable Cauchy distribution; generalized linear inversion was developed for the exact Zoeppritz equation, however this method still requires assuming the upper layer compressional velocity ratio as a smooth background compressional velocity ratio.
The longitudinal and transverse wave velocity ratio is taken as an important parameter in lithology and hydrocarbon-bearing property prediction, and is usually obtained by indirect calculation by utilizing an elastic parameter inversion result, which inevitably causes an accumulated error; in addition, the precision and the application range of the Zoeppritz linear approximation formula used in the prestack elastic parameter inversion in the above research are limited to a certain extent. Therefore, it is necessary to develop a prestack inversion method and system based on γ _ Zoeppritz equation.
The information disclosed in this background section is only for enhancement of understanding of the general background of the invention and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art already known to a person skilled in the art.
Disclosure of Invention
The invention provides a prestack inversion method and a prestack inversion system based on a gamma _ Zoeppritz equation, which can establish an accurate relation between an angle reflection coefficient and a three-elastic parameter natural logarithm by deducing the gamma _ Zoeppritz equation, introduce a first-order difference matrix and a reflection coefficient sparse constraint term and a low-frequency soft constraint term represented by a three-variable Cauchy distribution, construct an inversion target function, and finally obtain a high-resolution and transversely continuous inversion result by utilizing a generalized linear inversion mode.
According to one aspect of the invention, a method for prestack inversion based on the gamma _ Zoeppritz equation is provided. The method may include: step 1: constructing a low-frequency soft constraint term; step 2: constructing a seismic data fitting difference term; and step 3: constructing a reflection coefficient sparse constraint term; and 4, step 4: constructing an inversion target function according to the seismic data fitting difference term, the low-frequency soft constraint term and the reflection coefficient sparse constraint term; and 5: and obtaining an elastic parameter inversion result according to the inversion target function.
Preferably, the step 1 specifically includes: step 11: acquiring a longitudinal wave velocity model, a longitudinal and transverse wave velocity ratio model and a density model; step 12: analyzing the seismic data frequency spectrum, determining low pass frequency and low cut-off frequency, and further calculating the low-frequency trend of the natural logarithm of the three elastic parameters; step 13: and constructing a low-pass filtering operator, and constructing a low-frequency soft constraint term together with the low-frequency trend of the three-elasticity parameter natural logarithm.
Preferably, the step 2 specifically includes: step 21: acquiring angle superposition data; step 22: reading angle wavelets and constructing an angle wavelet convolution matrix; step 23: constructing an inversion initial model according to the low-frequency trend of the three-elasticity-parameter natural logarithm; step 24: constructing a gamma _ Zoeppritz equation according to the accurate Zoeppritz equation; step 25: according to the gamma _ Zoeppritz equation, the inversion initial model and the angle wavelet convolution matrix, and in combination with a Taylor first-order expansion formula, constructing an AVO forward equation set based on the gamma _ Zoeppritz equation; step 26: and constructing a seismic data fitting difference term according to the AVO forward equation set based on the gamma _ Zoeppritz equation and the angle stacking data.
Preferably, the γ _ Zoeppritz equation in step 24 is:
wherein the content of the first and second substances,
respectively representing the natural logarithm of the longitudinal wave velocity, the longitudinal wave velocity ratio and the density of the upper medium,
respectively showing the longitudinal wave velocity and the longitudinal and transverse wave velocity ratio of the lower mediumAnd the natural logarithm of the density, alpha1、α2、β1、β2Respectively representing the incident angle of longitudinal wave, the transmission angle of longitudinal wave, the reflection angle of converted wave and the transmission angle of converted wave, Rpp、Rps、Tpp、TpsThe longitudinal wave reflection coefficient, converted wave reflection coefficient, longitudinal wave transmission coefficient, and converted wave transmission coefficient are respectively expressed.
Preferably, in step 25, the AVO forward equation based on the γ _ Zoeppritz equation is as follows:
H＝Gm (2)
wherein H ═ Δ d + Gm0Δ d represents the difference between the measured seismic data and the synthetic seismic data, m0Representing an inverse initial model; m represents an inversion parameter;
W(θk) Representing an angle of incidence of thetakAngle wavelet convolution matrix of (L)vp(θk),Lγ(θk),Lρ(θk) Respectively, expressed by an incident angle of thetakThe first partial derivatives of the longitudinal wave reflection coefficient to the longitudinal wave velocity natural logarithm, the longitudinal wave velocity ratio natural logarithm and the density natural logarithm form a double diagonal matrix, wherein k is 1,2, … M,
wherein the content of the first and second substances,
represents the first partial derivative of the longitudinal wave reflection coefficient of the jth sampling point to the natural logarithm of the longitudinal wave velocity of the jth sampling point,
represents the first partial derivative of the longitudinal wave reflection coefficient of the jth sampling point to the natural logarithm of the longitudinal-transverse wave velocity ratio of the jth sampling point,
and (3) representing the first partial derivative of the longitudinal wave reflection coefficient of the jth sampling point to the natural logarithm of the density of the jth sampling point, wherein j is 1,2 and … N.
Preferably, the step 3 specifically includes: step 31: calculating a scale matrix by using a maximum expectation algorithm according to the logging data; step 32: and constructing a reflection coefficient sparse constraint term by combining a first-order difference matrix and the tri-variable Cauchy distribution according to the scale matrix.
Preferably, in step 4, the inversion objective function is:
wherein, (H-Gm)T(H-Gm) is the seismic data fitting difference,
is a sparse constraint term of the reflection coefficient, (omega m-m)T)T(Ωm-mT) The low-frequency soft constraint term is represented by lambda, the reflection coefficient sparse constraint weight is represented by kappa, the low-frequency soft constraint weight is represented by T, and the matrix transposition is represented by T; wherein the content of the first and second substances,
d is a first-order difference matrix and is a second-order difference matrix,
ifor a 3 row 3N column matrix:
wherein [ ] A]xyThe x-th row and the y-th column of the matrix are represented,
is a scale matrix;
where L is a low-pass filter matrix and,
wherein the content of the first and second substances,
the natural logarithm low-frequency trend of longitudinal wave velocity, the natural logarithm low-frequency trend of longitudinal and transverse wave velocity ratio and the natural logarithm low-frequency trend of density are respectively represented.
Preferably, the step 5 specifically includes: step 51: solving an inversion target function by using an improved iterative reweighted least square algorithm to obtain an inversion result of the three-elastic-parameter natural logarithm; step 52: and performing indexing on the inversion result of the three-elastic-parameter natural logarithm to obtain a three-elastic-parameter inversion result.
Preferably, in step 51, the improved iterative reweighted least squares algorithm is: step 511: low frequency trend m of three elastic parameter natural logarithmpAs an inverse initial model m0(ii) a Step 512: according to m0Calculating an AVO forward modeling matrix G, a matrix H and a non-uniformity weighting matrix Q; step 513: solving by using Cholesky decomposition and upper and lower triangular matrix decomposition: (G)TG+λQ+κΩTΩ)m＝GTH+κΩTmpObtaining an inversion result m of the iteration; step 514: taking m as an initial model m of the next iteration0(ii) a Step 515: repeating the step 512 to the step 514 until the specified iteration times are reached, and outputting an inversion result m; wherein the non-uniformity weighting matrix Q is:
according to another aspect of the present invention, a prestack inversion system based on γ _ Zoeppritz equation is provided, wherein the system comprises: a memory storing computer-executable instructions; a processor executing computer executable instructions in the memory to perform the steps of: step 1: constructing a low-frequency soft constraint term; step 2: constructing a seismic data fitting difference term; and step 3: constructing a reflection coefficient sparse constraint term; and 4, step 4: constructing an inversion target function according to the seismic data fitting difference term, the low-frequency soft constraint term and the reflection coefficient sparse constraint term; and 5: and obtaining an elastic parameter inversion result according to the inversion target function.
Preferably, the step 1 specifically includes: step 11: acquiring a longitudinal wave velocity model, a longitudinal and transverse wave velocity ratio model and a density model; step 12: analyzing the seismic data frequency spectrum, determining low pass frequency and low cut-off frequency, and further calculating the low-frequency trend of the natural logarithm of the three elastic parameters; step 13: and constructing a low-pass filtering operator, and constructing a low-frequency soft constraint term together with the low-frequency trend of the three-elasticity parameter natural logarithm.
Preferably, the step 2 specifically includes: step 21: acquiring angle superposition data; step 22: reading angle wavelets and constructing an angle wavelet convolution matrix; step 23: constructing an inversion initial model according to the low-frequency trend of the three-elasticity-parameter natural logarithm; step 24: constructing a gamma _ Zoeppritz equation according to the accurate Zoeppritz equation; step 25: according to the gamma _ Zoeppritz equation, the inversion initial model and the angle wavelet convolution matrix, and in combination with a Taylor first-order expansion formula, constructing an AVO forward equation set based on the gamma _ Zoeppritz equation; step 26: and constructing a seismic data fitting difference term according to the AVO forward equation set based on the gamma _ Zoeppritz equation and the angle stacking data.
Preferably, the γ _ Zoeppritz equation in step 24 is:
wherein the content of the first and second substances,
respectively representing the natural logarithm of the longitudinal wave velocity, the longitudinal wave velocity ratio and the density of the upper medium,
respectively representing the natural logarithm of the longitudinal wave velocity, the longitudinal-transverse wave velocity ratio and the density of the underlying medium, alpha1、α2、β1、β2Respectively representing the incident angle of longitudinal wave, the transmission angle of longitudinal wave, the reflection angle of converted wave and the transmission angle of converted wave, Rpp、Rps、Tpp、TpsThe longitudinal wave reflection coefficient, converted wave reflection coefficient, longitudinal wave transmission coefficient, and converted wave transmission coefficient are respectively expressed.
Preferably, in step 25, the AVO forward equation based on the γ _ Zoeppritz equation is as follows:
H＝Gm (2)
wherein H ═ Δ d + Gm0Δ d represents the difference between the measured seismic data and the synthetic seismic data, m0Representing an inverse initial model; m represents an inversion parameter;
W(θk) Representing an angle of incidence of thetakAngle wavelet convolution matrix of (L)vp(θk),Lγ(θk),Lρ(θk) Respectively, expressed by an incident angle of thetakThe first partial derivatives of the longitudinal wave reflection coefficient to the longitudinal wave velocity natural logarithm, the longitudinal wave velocity ratio natural logarithm and the density natural logarithm form a double diagonal matrix, wherein k is 1,2, … M,
wherein the content of the first and second substances,
represents the first partial derivative of the longitudinal wave reflection coefficient of the jth sampling point to the natural logarithm of the longitudinal wave velocity of the jth sampling point,
represents the first partial derivative of the longitudinal wave reflection coefficient of the jth sampling point to the natural logarithm of the longitudinal-transverse wave velocity ratio of the jth sampling point,
and (3) representing the first partial derivative of the longitudinal wave reflection coefficient of the jth sampling point to the natural logarithm of the density of the jth sampling point, wherein j is 1,2 and … N.
Preferably, the step 3 specifically includes: step 31: calculating a scale matrix by using a maximum expectation algorithm according to the logging data; step 32: and constructing a reflection coefficient sparse constraint term by combining a first-order difference matrix and the tri-variable Cauchy distribution according to the scale matrix.
Preferably, in step 4, the inversion objective function is:
wherein, (H-Gm)T(H-Gm) is the seismic data fitting difference,
is a sparse constraint term of the reflection coefficient, (omega m-m)T)T(Ωm-mT) Is composed ofA low-frequency soft constraint term, wherein lambda represents a reflection coefficient sparse constraint weight, kappa represents a low-frequency soft constraint weight, and T represents a matrix transposition; wherein the content of the first and second substances,
d is a first-order difference matrix and is a second-order difference matrix,
ifor a 3 row 3N column matrix:
wherein [ ] A]xyThe x-th row and the y-th column of the matrix are represented,
is a scale matrix;
where L is a low-pass filter matrix and,
wherein the content of the first and second substances,
the natural logarithm low-frequency trend of longitudinal wave velocity, the natural logarithm low-frequency trend of longitudinal and transverse wave velocity ratio and the natural logarithm low-frequency trend of density are respectively represented.
Preferably, the step 5 specifically includes: step 51: solving an inversion target function by using an improved iterative reweighted least square algorithm to obtain an inversion result of the three-elastic-parameter natural logarithm; step 52: and performing indexing on the inversion result of the three-elastic-parameter natural logarithm to obtain a three-elastic-parameter inversion result.
Preferably, in step 51, the improved iterative reweighted least squares algorithm is: step 511: low frequency trend m of three elastic parameter natural logarithmpAs an inverse initial modelm0(ii) a Step 512: according to m0Calculating an AVO forward modeling matrix G, a matrix H and a non-uniformity weighting matrix Q; step 513: solving by using Cholesky decomposition and upper and lower triangular matrix decomposition: (G)TG+λQ+κΩTΩ)m＝GTH+κΩTmpObtaining an inversion result m of the iteration; step 514: taking m as an initial model m of the next iteration0(ii) a Step 515: repeating the step 512 to the step 514 until the specified iteration times are reached, and outputting an inversion result m; wherein the non-uniformity weighting matrix Q is:
the method and apparatus of the present invention have other features and advantages which will be apparent from or are set forth in detail in the accompanying drawings and the following detailed description, which are incorporated herein, and which together serve to explain certain principles of the invention.
Drawings
The above and other objects, features and advantages of the present invention will become more apparent by describing in more detail exemplary embodiments thereof with reference to the attached drawings, in which like reference numerals generally represent like parts.
FIG. 1 shows a flow chart of the steps of a method of prestack inversion based on the γ _ Zoeppritz equation in accordance with the present invention.
Fig. 2a, 2b, 2c show schematic diagrams of angle overlay data with central angles of 8 degrees, 18 degrees, 28 degrees, respectively, according to an embodiment of the invention.
Fig. 3 shows a schematic diagram of angular wavelets with central angles of 8 degrees, 18 degrees, 28 degrees, respectively, according to fig. 2a, 2b, 2 c.
Fig. 4a, 4b, and 4c are schematic diagrams illustrating a compressional velocity inversion profile, a compressional velocity ratio inversion profile, and a density inversion profile, respectively, according to an embodiment of the invention.
Fig. 5a, 5b and 5c are schematic diagrams respectively illustrating a compressional-compressional velocity ratio inversion result, a compressional velocity inversion result and a density inversion result of a well bypass according to an embodiment of the invention.
Detailed Description
The invention will be described in more detail below with reference to the accompanying drawings. While the preferred embodiments of the present invention are shown in the drawings, it should be understood that the present invention may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.
FIG. 1 shows a flow chart of the steps of a method of prestack inversion based on the γ _ Zoeppritz equation in accordance with the present invention.
In this embodiment, the method for prestack inversion based on γ _ Zoeppritz equation according to the present invention may include: step 1: constructing a low-frequency soft constraint term; step 2: constructing a seismic data fitting difference term; and step 3: constructing a reflection coefficient sparse constraint term; and 4, step 4: constructing an inversion target function according to the seismic data fitting difference term, the low-frequency soft constraint term and the reflection coefficient sparse constraint term; and 5: and obtaining an elastic parameter inversion result according to the inversion target function.
In one example, step 1 specifically includes: step 11: acquiring a longitudinal wave velocity model, a longitudinal and transverse wave velocity ratio model and a density model; step 12: analyzing the seismic data frequency spectrum, determining low pass frequency and low cut-off frequency, and further calculating the low-frequency trend of the natural logarithm of the three elastic parameters; step 13: and constructing a low-pass filtering operator, and constructing a low-frequency soft constraint term together with the low-frequency trend of the three-elasticity parameter natural logarithm.
In one example, step 2 specifically includes: step 21: acquiring angle superposition data; step 22: reading angle wavelets and constructing an angle wavelet convolution matrix; step 23: constructing an inversion initial model according to the low-frequency trend of the three-elasticity-parameter natural logarithm; step 24: constructing a gamma _ Zoeppritz equation according to the accurate Zoeppritz equation; step 25: according to a gamma _ Zoeppritz equation, an inversion initial model and an angle wavelet convolution matrix, and in combination with a Taylor first-order expansion formula, an AVO forward equation set based on the gamma _ Zoeppritz equation is constructed; step 26: and constructing a seismic data fitting difference term according to the AVO forward equation set based on the gamma _ Zoeppritz equation and the angle stacking data.
In one example, the γ _ Zoeppritz equation in step 24 is:
wherein the content of the first and second substances,
respectively representing the natural logarithm of the longitudinal wave velocity, the longitudinal wave velocity ratio and the density of the upper medium,
respectively representing the natural logarithm of the longitudinal wave velocity, the longitudinal-transverse wave velocity ratio and the density of the underlying medium, alpha1、α2、β1、β2Respectively representing the incident angle of longitudinal wave, the transmission angle of longitudinal wave, the reflection angle of converted wave and the transmission angle of converted wave, Rpp、Rps、Tpp、TpsThe longitudinal wave reflection coefficient, converted wave reflection coefficient, longitudinal wave transmission coefficient, and converted wave transmission coefficient are respectively expressed.
In one example, in step 25, the AVO forward equation set based on the γ _ Zoeppritz equation is:
H＝Gm (2)
wherein H ═ Δ d + Gm0Δ d represents the difference between the measured seismic data and the synthetic seismic data, m0Representing an inverse initial model; m represents an inversion parameter;
W(θk) Representing an angle of incidence of thetakAngle wavelet convolution matrix of (L)vp(θk),Lγ(θk),Lρ(θk) Respectively, expressed by an incident angle of thetakThe first partial derivatives of the longitudinal wave reflection coefficient to the longitudinal wave velocity natural logarithm, the longitudinal wave velocity ratio natural logarithm and the density natural logarithm form a double diagonal matrix, wherein k is 1,2, … M,
wherein the content of the first and second substances,
represents the first partial derivative of the longitudinal wave reflection coefficient of the jth sampling point to the natural logarithm of the longitudinal wave velocity of the jth sampling point,
represents the first partial derivative of the longitudinal wave reflection coefficient of the jth sampling point to the natural logarithm of the longitudinal-transverse wave velocity ratio of the jth sampling point,
and a first-order partial derivative of the longitudinal wave reflection coefficient of the jth sampling point to the density natural logarithm of the jth sampling point is represented, j is 1,2 and … N, the first-order partial derivative is obtained by slightly perturbing the inversion initial model, then obtaining the perturbed longitudinal wave reflection coefficient by means of a gamma _ Zoepprit equation, and finally calculating by means of first-order central difference.
In one example, step 3 specifically includes: step 31: calculating a scale matrix by using a maximum expectation algorithm according to the logging data; step 32: and constructing a reflection coefficient sparse constraint term by combining a first-order difference matrix and the three-variable Cauchy distribution according to the scale matrix.
In one example, in step 4, the inverse objective function is:
wherein, (H-Gm)T(H-Gm) is the seismic data fitting difference,
is a sparse constraint term of the reflection coefficient, (omega m-m)T)T(Ωm-mT) The low-frequency soft constraint term is represented by lambda, the reflection coefficient sparse constraint weight is represented by kappa, the low-frequency soft constraint weight is represented by T, and the matrix transposition is represented by T; wherein the content of the first and second substances,
d is a first-order difference matrix and is a second-order difference matrix,
ifor a 3 row 3N column matrix:
wherein [ ] A]xyThe x-th row and the y-th column of the matrix are represented,
is a scale matrix;
where L is a low-pass filter matrix and,
wherein the content of the first and second substances,
respectively representing the natural logarithm low-frequency trend of longitudinal wave velocity and the natural logarithm low-frequency trend of longitudinal and transverse wave velocity ratioTrend and density natural log low frequency trend.
In one example, step 5 specifically includes: step 51: solving an inversion target function by using an improved iterative reweighted least square algorithm to obtain an inversion result of the three-elastic-parameter natural logarithm; step 52: and indexing the inversion result of the natural logarithm of the three elastic parameters to obtain the inversion result of the three elastic parameters.
In one example, in step 51, the improved iterative reweighted least squares algorithm is: step 511: low frequency trend m of three elastic parameter natural logarithmpAs an inverse initial model m0(ii) a Step 512: according to m0Calculating an AVO forward modeling matrix G, a matrix H and a non-uniformity weighting matrix Q; step 513: solving by using Cholesky decomposition and upper and lower triangular matrix decomposition: (G)TG+λQ+κΩTΩ)m＝GTH+κΩTmpObtaining an inversion result m of the iteration; step 514: taking m as an initial model m of the next iteration0(ii) a Step 515: repeating the step 512 to the step 514 until the specified iteration times are reached, and outputting an inversion result m; wherein the non-uniformity weighting matrix Q is:
specifically, the prestack inversion method based on the gamma _ Zoeppritz equation according to the invention can comprise the following steps:
step 1: constructing a low-frequency soft constraint term, which specifically comprises the following steps:
step 11: obtaining a longitudinal wave velocity model VpA Mod, a longitudinal and transverse wave velocity ratio model gamma _ Mod and a density model rho _ Mod;
step 12: analysing the seismic data spectrum to determine a low pass frequency flow_passAnd low cut-off frequency flow_cutAnd further calculating the low-frequency trend of the natural logarithm of the three elastic parameters, wherein the low-frequency trend of the natural logarithm of the longitudinal wave velocity is formula (5):
the low frequency trend of the natural logarithm of the longitudinal-transverse wave velocity ratio is formula (6):
the low frequency trend of the natural logarithm of the density is formula (7):
wherein lowpass _ Filter represents low-pass filtering;
step 13: constructing a low-pass filtering operator, and constructing a low-frequency soft constraint term together with the low-frequency trend of the three-elasticity parameter natural logarithm; the function of the low-pass filter operator L is to express the low-pass filtering process as matrix operation, so as to construct a low-frequency soft constraint term, and add the low-frequency soft constraint term into an inversion objective function, wherein the expression of the low-pass filter operator L is as follows:
in the formula (2), Λ is a diagonal matrix, and diagonal elements of the diagonal matrix consist of low-pass filter window functions; fDA DFT-transformed matrix is represented and,
representation matrix FDIs the inverse of, i.e. the inverse DFT transform matrix, where FDThe mathematical expression of (a) is:
wherein xi ═ e-2πj/NJ represents a unit imaginary number, and N represents the number of reflection coefficients.
Step 2: constructing a seismic data fitting difference term, which specifically comprises the following steps:
step 21: obtaining the angle superposition data as formula (10):
wherein d (θ)k) Where k is 1,2, … M denotes the incident angle θkAngle overlay data of (a);
step 22: reading M angular wavelets w (theta)k) K is 1,2, … M, an angular wavelet convolution matrix W (θ)k),k＝1,2,…M；
Step 23: constructing an inversion initial model according to the low-frequency trend of the three-elastic-parameter natural logarithm, and inverting the initial model m during the first inversion iteration0＝mT；
Step 24: constructing a gamma _ Zoeppritz equation as a formula (1) according to the accurate Zoeppritz equation; because the accurate Zoeppritz equation is used for inverting three elastic parameters of longitudinal wave velocity, longitudinal and transverse wave velocity ratio and density, the accurate Zoeppritz equation needs to be recombined into functions of a longitudinal wave velocity natural logarithm, a longitudinal and transverse wave velocity ratio natural logarithm and a density natural logarithm, namely a gamma-Zoeppritz equation; natural logarithm is selected instead of elastic parameters, so that a reflection coefficient sparse constraint term is introduced conveniently in the follow-up process;
step 25: according to the gamma _ Zoeppritz equation, the inversion initial model and the convolution matrix of the angle wavelets, the gamma _ Zoeppritz equation is subjected to linear approximation by using a Taylor first-order expansion, and the convolution theory is combined to obtain:
Δd＝GΔm (11)
where Δ d represents the difference between the angle stacking data and the forward data, and Δ m represents the disturbance amount of the inversion parameter:
Δm＝m-m0 (12)
wherein m represents inversion parameters, i.e. natural logarithm of compressional velocity, natural logarithm of velocity ratio of compressional and shear, and natural logarithm of density, m0An initial model representing inversion parameters; g represents a forward matrix formed by a Jacobian matrix and a wavelet convolution matrix;
considering that the prior distribution of the disturbance quantity Δ m of the inversion parameters is not convenient for statistics, the prior distribution is not easy to calculateThis does not facilitate direct inversion, requiring a conversion of Δ m to m. Therefore, Gm is added to both sides of the equal sign of formula (11)0That is, the AVO forward equation set based on the γ _ Zoeppritz equation is obtained as the formula (2), where H ═ Δ d + Gm0；
Step 26: and constructing a seismic data fitting difference term according to the AVO forward equation set based on the gamma _ Zoeppritz equation and the angle stacking data.
And step 3: constructing a reflection coefficient sparse constraint term, specifically comprising: step 31: calculating a scale matrix by using a maximum expectation algorithm according to the logging data; step 32: and constructing a reflection coefficient sparse constraint term by combining a first-order difference matrix and the three-variable Cauchy distribution according to the scale matrix.
And 4, step 4: and (4) constructing an inversion objective function as a formula (4) according to the seismic data fitting difference term, the low-frequency soft constraint term and the reflection coefficient sparse constraint term.
And 5: obtaining an elastic parameter inversion result according to an inversion target function, which specifically comprises the following steps: step 51: solving an inversion target function by using an improved iterative reweighted least square algorithm to obtain an inversion result of the three-elastic-parameter natural logarithm; step 52: according to the inversion result of the natural logarithm of the three-elasticity parameters, performing exponential transformation on the inversion result to obtain the inversion result of the three-elasticity parameters;
the improved iteration reweighted least square algorithm comprises the following steps: step 511: low frequency trend m of three elastic parameter natural logarithmpAs an inverse initial model m0(ii) a Step 512: according to m0Calculating an AVO forward modeling matrix G, a matrix H and a non-uniformity weighting matrix Q; step 513: solving by using Cholesky decomposition and upper and lower triangular matrix decomposition: (G)TG+λQ+κΩTΩ)m＝GTH+κΩTmpObtaining an inversion result m of the iteration; step 514: taking m as an initial model m of the next iteration0(ii) a Step 515: repeating the step 512 to the step 514 until the specified iteration times are reached, and outputting an inversion result m; wherein the non-uniformity weighting matrix Q is:
according to the method, an accurate relation between an angle reflection coefficient and a three-elastic parameter natural logarithm is established by deducing a gamma _ Zoeppritz equation, a first-order difference matrix and a reflection coefficient sparse constraint term and a low-frequency soft constraint term of a three-variable Cauchy distribution representation are introduced, an inversion target function is constructed, and finally a high-resolution and transversely continuous inversion result is obtained by utilizing a generalized linear inversion mode.
Application example
To facilitate understanding of the solution of the embodiments of the present invention and the effects thereof, a specific application example is given below. It will be understood by those skilled in the art that this example is merely for the purpose of facilitating an understanding of the present invention and that any specific details thereof are not intended to limit the invention in any way.
Fig. 2a, 2b, 2c show schematic diagrams of angle overlay data with central angles of 8 degrees, 18 degrees, 28 degrees, respectively, according to an embodiment of the invention.
Fig. 3 shows a schematic diagram of angular wavelets with central angles of 8 degrees, 18 degrees, 28 degrees, respectively, according to fig. 2a, 2b, 2 c.
The prestack inversion method based on the gamma-Zoeppritz equation comprises the following steps:
step 1: constructing a low-frequency soft constraint term, which specifically comprises the following steps: step 11: acquiring a longitudinal wave velocity model, a longitudinal and transverse wave velocity ratio model and a density model; step 12: analyzing the seismic data frequency spectrum, determining low pass frequency and low cut-off frequency, and further calculating the low-frequency trend of the natural logarithm of the three elastic parameters; step 13: and constructing a low-pass filtering operator, and constructing a low-frequency soft constraint term together with the low-frequency trend of the three-elasticity parameter natural logarithm.
Step 2: constructing a seismic data fitting difference term, which specifically comprises the following steps: step 21: obtaining angle overlay data, as shown in FIGS. 2a-2 c; step 22: reading the angle wavelets, and constructing an angle wavelet convolution matrix as shown in FIG. 3; step 23: constructing an inversion initial model according to the low-frequency trend of the three-elasticity-parameter natural logarithm; step 24: constructing a gamma _ Zoeppritz equation as a formula (1) according to the accurate Zoeppritz equation; step 25: according to a gamma _ Zoeppritz equation, an inversion initial model and an angle wavelet convolution matrix, and in combination with a Taylor first-order expansion formula, constructing an AVO forward equation set based on the gamma _ Zoeppritz equation as a formula (2); step 26: and constructing a seismic data fitting difference term according to the AVO forward equation set based on the gamma _ Zoeppritz equation and the angle stacking data.
And step 3: constructing a reflection coefficient sparse constraint term, specifically comprising: step 31: calculating a scale matrix by using a maximum expectation algorithm according to the logging data; step 32: and constructing a reflection coefficient sparse constraint term by combining a first-order difference matrix and the three-variable Cauchy distribution according to the scale matrix.
And 4, step 4: and (4) constructing an inversion objective function as a formula (4) according to the seismic data fitting difference term, the low-frequency soft constraint term and the reflection coefficient sparse constraint term.
And 5: obtaining an elastic parameter inversion result according to an inversion target function, which specifically comprises the following steps: step 51: solving an inversion target function by using an improved iterative reweighted least square algorithm to obtain an inversion result of the three-elastic-parameter natural logarithm; step 52: according to the inversion result of the natural logarithm of the three-elasticity parameters, performing exponential transformation on the inversion result to obtain the inversion result of the three-elasticity parameters; the improved iteration reweighted least square algorithm comprises the following steps: step 511: low frequency trend m of three elastic parameter natural logarithmpAs an inverse initial model m0(ii) a Step 512: according to m0Calculating an AVO forward modeling matrix G, a matrix H and a non-uniformity weighting matrix Q; step 513: solving by using Cholesky decomposition and upper and lower triangular matrix decomposition: (G)TG+λQ+κΩTΩ)m＝GTH+κΩTmpObtaining an inversion result m of the iteration; step 514: taking m as an initial model m of the next iteration0(ii) a Step 515: repeating the step 512 to the step 514 until the specified iteration times are reached, and outputting an inversion result m; wherein the non-uniformity weighting matrix Q is:
fig. 4a, 4b, and 4c are schematic diagrams illustrating a compressional velocity inversion profile, a compressional velocity ratio inversion profile, and a density inversion profile, respectively, according to an embodiment of the invention. It can be seen that the longitudinal-transverse wave velocity ratio profile is transversely continuous and has higher vertical resolution, and can show the transverse change of the reservoir physical properties.
Fig. 5a, 5b, and 5c respectively show schematic diagrams of a longitudinal-to-transverse wave velocity ratio inversion result, a longitudinal wave velocity inversion result, and a density inversion result of a well side channel according to an embodiment of the present invention, where a verification well is located at a CDP301, a thin solid line is an actual measurement well curve, a thick dotted line is a well side channel inversion result, and a longitudinal wave velocity inversion result, a longitudinal-to-transverse wave velocity ratio inversion result, and a density inversion result are sequentially shown from left to right, and it can be seen from the diagrams that the well side channel inversion result matches the actual measurement curve inversion result.
In conclusion, the invention establishes the accurate relation between the angle reflection coefficient and the three-elastic-parameter natural logarithm by deducing the gamma _ Zoeppritz equation, introduces the first-order difference matrix and the reflection coefficient sparse constraint term and the low-frequency soft constraint term represented by the three-variable Cauchy distribution, constructs the inversion target function, and finally obtains the high-resolution and transversely continuous inversion result by utilizing the generalized linear inversion mode.
It will be appreciated by persons skilled in the art that the above description of embodiments of the invention is intended only to illustrate the benefits of embodiments of the invention and is not intended to limit embodiments of the invention to any examples given.
According to an embodiment of the present invention, there is provided a prestack inversion system based on γ _ Zoeppritz equation, the system comprising: a memory storing computer-executable instructions; a processor executing computer executable instructions in the memory to perform the steps of: step 1: constructing a low-frequency soft constraint term; step 2: constructing a seismic data fitting difference term; and step 3: constructing a reflection coefficient sparse constraint term; and 4, step 4: constructing an inversion target function according to the seismic data fitting difference term, the low-frequency soft constraint term and the reflection coefficient sparse constraint term; and 5: and obtaining an elastic parameter inversion result according to the inversion target function.
In one example, step 1 specifically includes: step 11: acquiring a longitudinal wave velocity model, a longitudinal and transverse wave velocity ratio model and a density model; step 12: analyzing the seismic data frequency spectrum, determining low pass frequency and low cut-off frequency, and further calculating the low-frequency trend of the natural logarithm of the three elastic parameters; step 13: and constructing a low-pass filtering operator, and constructing a low-frequency soft constraint term together with the low-frequency trend of the three-elasticity parameter natural logarithm.
In one example, step 2 specifically includes: step 21: acquiring angle superposition data; step 22: reading angle wavelets and constructing an angle wavelet convolution matrix; step 23: constructing an inversion initial model according to the low-frequency trend of the three-elasticity-parameter natural logarithm; step 24: constructing a gamma _ Zoeppritz equation according to the accurate Zoeppritz equation; step 25: according to a gamma _ Zoeppritz equation, an inversion initial model and an angle wavelet convolution matrix, and in combination with a Taylor first-order expansion formula, an AVO forward equation set based on the gamma _ Zoeppritz equation is constructed; step 26: and constructing a seismic data fitting difference term according to the AVO forward equation set based on the gamma _ Zoeppritz equation and the angle stacking data.
In one example, the γ _ Zoeppritz equation in step 24 is:
wherein the content of the first and second substances,
respectively representing the natural logarithm of the longitudinal wave velocity, the longitudinal wave velocity ratio and the density of the upper medium,
respectively representing the natural logarithm of the longitudinal wave velocity, the longitudinal-transverse wave velocity ratio and the density of the underlying medium, alpha1、α2、β1、β2Respectively representing the incident angle of longitudinal wave, the transmission angle of longitudinal wave, the reflection angle of converted wave and the transmission angle of converted wave, Rpp、Rps、Tpp、TpsThe longitudinal wave reflection coefficient, converted wave reflection coefficient, longitudinal wave transmission coefficient, and converted wave transmission coefficient are respectively expressed.
In one example, in step 25, the AVO forward equation set based on the γ _ Zoeppritz equation is:
H＝Gm (2)
wherein H ═ Δ d + Gm0Δ d represents the difference between the measured seismic data and the synthetic seismic data, m0Representing an inverse initial model; m represents an inversion parameter;
W(θk) Representing an angle of incidence of thetakAngle wavelet convolution matrix of (L)vp(θk),Lγ(θk),Lρ(θk) Respectively, expressed by an incident angle of thetakThe first partial derivatives of the longitudinal wave reflection coefficient to the longitudinal wave velocity natural logarithm, the longitudinal wave velocity ratio natural logarithm and the density natural logarithm form a double diagonal matrix, wherein k is 1,2, … M,
wherein the content of the first and second substances,
represents the first partial derivative of the longitudinal wave reflection coefficient of the jth sampling point to the natural logarithm of the longitudinal wave velocity of the jth sampling point,
represents the first partial derivative of the longitudinal wave reflection coefficient of the jth sampling point to the natural logarithm of the longitudinal-transverse wave velocity ratio of the jth sampling point,
and (3) representing the first partial derivative of the longitudinal wave reflection coefficient of the jth sampling point to the natural logarithm of the density of the jth sampling point, wherein j is 1,2 and … N.
In one example, step 3 specifically includes: step 31: calculating a scale matrix by using a maximum expectation algorithm according to the logging data; step 32: and constructing a reflection coefficient sparse constraint term by combining a first-order difference matrix and the three-variable Cauchy distribution according to the scale matrix.
In one example, in step 4, the inverse objective function is:
wherein, (H-Gm)T(H-Gm) is the seismic data fitting difference,
is a sparse constraint term of the reflection coefficient, (omega m-m)T)T(Ωm-mT) The low-frequency soft constraint term is represented by lambda, the reflection coefficient sparse constraint weight is represented by kappa, the low-frequency soft constraint weight is represented by T, and the matrix transposition is represented by T; wherein the content of the first and second substances,
d is a first-order difference matrix and is a second-order difference matrix,
ifor a 3 row 3N column matrix:
wherein [ ] A]xyThe x-th row and the y-th column of the matrix are represented,
is a scale matrix;
wherein L is the low pass filter momentThe number of the arrays is determined,
wherein the content of the first and second substances,
the natural logarithm low-frequency trend of longitudinal wave velocity, the natural logarithm low-frequency trend of longitudinal and transverse wave velocity ratio and the natural logarithm low-frequency trend of density are respectively represented.
In one example, step 5 specifically includes: step 51: solving an inversion target function by using an improved iterative reweighted least square algorithm to obtain an inversion result of the three-elastic-parameter natural logarithm; step 52: and indexing the inversion result of the natural logarithm of the three elastic parameters to obtain the inversion result of the three elastic parameters.
In one example, in step 51, the improved iterative reweighted least squares algorithm is: step 511: low frequency trend m of three elastic parameter natural logarithmpAs an inverse initial model m0(ii) a Step 512: according to m0Calculating an AVO forward modeling matrix G, a matrix H and a non-uniformity weighting matrix Q; step 513: solving by using Cholesky decomposition and upper and lower triangular matrix decomposition: (G)TG+λQ+κΩTΩ)m＝GTH+κΩTmpObtaining an inversion result m of the iteration; step 514: taking m as an initial model m of the next iteration0(ii) a Step 515: repeating the step 512 to the step 514 until the specified iteration times are reached, and outputting an inversion result m; wherein the non-uniformity weighting matrix Q is:
the system establishes an accurate relation between an angle reflection coefficient and a three-elastic parameter natural logarithm by deducing a gamma _ Zoeppritz equation, introduces a first-order difference matrix and a reflection coefficient sparse constraint term and a low-frequency soft constraint term represented by a three-variable Cauchy distribution, constructs an inversion target function, and finally obtains a high-resolution and transversely continuous inversion result by utilizing a generalized linear inversion mode.
It will be appreciated by persons skilled in the art that the above description of embodiments of the invention is intended only to illustrate the benefits of embodiments of the invention and is not intended to limit embodiments of the invention to any examples given.
Having described embodiments of the present invention, the foregoing description is intended to be exemplary, not exhaustive, and not limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments.
Claims (10)
1. A prestack inversion method based on a gamma _ Zoeppritz equation is characterized by comprising the following steps:
step 1: constructing a low-frequency soft constraint term;
step 2: constructing a seismic data fitting difference term;
and step 3: constructing a reflection coefficient sparse constraint term;
and 4, step 4: constructing an inversion target function according to the seismic data fitting difference term, the low-frequency soft constraint term and the reflection coefficient sparse constraint term;
and 5: and obtaining an elastic parameter inversion result according to the inversion target function.
2. The method for prestack inversion based on γ _ Zoeppritz equation as claimed in claim 1, wherein said step 1 specifically comprises:
step 11: acquiring a longitudinal wave velocity model, a longitudinal and transverse wave velocity ratio model and a density model;
step 12: analyzing the seismic data frequency spectrum, determining low pass frequency and low cut-off frequency, and further calculating the low-frequency trend of the natural logarithm of the three elastic parameters;
step 13: and constructing a low-pass filtering operator, and constructing a low-frequency soft constraint term together with the low-frequency trend of the three-elasticity parameter natural logarithm.
3. The method for prestack inversion based on γ _ Zoeppritz equation as claimed in claim 2, wherein said step 2 specifically comprises:
step 21: acquiring angle superposition data;
step 22: reading angle wavelets and constructing an angle wavelet convolution matrix;
step 23: constructing an inversion initial model according to the low-frequency trend of the three-elasticity-parameter natural logarithm;
step 24: constructing a gamma _ Zoeppritz equation according to the accurate Zoeppritz equation;
step 25: according to the gamma _ Zoeppritz equation, the inversion initial model and the angle wavelet convolution matrix, and in combination with a Taylor first-order expansion formula, constructing an AVO forward equation set based on the gamma _ Zoeppritz equation;
step 26: and constructing a seismic data fitting difference term according to the AVO forward equation set based on the gamma _ Zoeppritz equation and the angle stacking data.
4. The method of prestack inversion based on γ _ Zoeppritz equation of claim 3, wherein the γ _ Zoeppritz equation in step 24 is:
wherein the content of the first and second substances,
respectively representing the natural logarithm of the longitudinal wave velocity, the longitudinal wave velocity ratio and the density of the upper medium,
respectively representing the natural logarithm of the longitudinal wave velocity, the longitudinal-transverse wave velocity ratio and the density of the underlying medium, alpha1、α2、β1、β2Respectively representing the incident angle of longitudinal wave, the transmission angle of longitudinal wave, the reflection angle of converted wave and the transmission angle of converted wave, Rpp、Rps、Tpp、TpsThe longitudinal wave reflection coefficient, converted wave reflection coefficient, longitudinal wave transmission coefficient, and converted wave transmission coefficient are respectively expressed.
5. The method of pre-stack inversion based on γ _ Zoeppritz equation of claim 3, wherein in step 25, the set of AVO forward equations based on γ _ Zoeppritz equation is:
H＝Gm (2)
wherein H ═ Δ d + Gm0Δ d represents the difference between the measured seismic data and the synthetic seismic data, m0Representing an inverse initial model; m represents an inversion parameter;
W(θk) Representing an angle of incidence of thetakAngle wavelet convolution matrix of (L)vp(θk),Lγ(θk),Lρ(θk) Respectively, expressed by an incident angle of thetakThe first partial derivatives of the longitudinal wave reflection coefficient to the longitudinal wave velocity natural logarithm, the longitudinal wave velocity ratio natural logarithm and the density natural logarithm form a double diagonal matrix, wherein k is 1,2, … M,
wherein the content of the first and second substances,
represents the first partial derivative of the longitudinal wave reflection coefficient of the jth sampling point to the natural logarithm of the longitudinal wave velocity of the jth sampling point,
represents the first partial derivative of the longitudinal wave reflection coefficient of the jth sampling point to the natural logarithm of the longitudinal-transverse wave velocity ratio of the jth sampling point,
and (3) representing the first partial derivative of the longitudinal wave reflection coefficient of the jth sampling point to the natural logarithm of the density of the jth sampling point, wherein j is 1,2 and … N.
6. The method for prestack inversion based on γ _ Zoeppritz equation as claimed in claim 1, wherein said step 3 specifically comprises:
step 31: calculating a scale matrix by using a maximum expectation algorithm according to the logging data;
step 32: and constructing a reflection coefficient sparse constraint term by combining a first-order difference matrix and the tri-variable Cauchy distribution according to the scale matrix.
7. The method of pre-stack inversion based on γ _ Zoeppritz equation of claim 1, wherein in step 4, the inversion objective function is:
wherein, (H-Gm)T(H-Gm) is the seismic data fitting difference,
is a sparse constraint term of the reflection coefficient, (omega m-m)T)T(Ωm-mT) The low-frequency soft constraint term is represented by lambda, the reflection coefficient sparse constraint weight is represented by kappa, the low-frequency soft constraint weight is represented by T, and the matrix transposition is represented by T; wherein the content of the first and second substances,
d is a first-order difference matrix and is a second-order difference matrix,
ifor a 3 row 3N column matrix:
wherein [ ] A]xyThe x-th row and the y-th column of the matrix are represented,
is a scale matrix;
where L is a low-pass filter matrix and,
wherein the content of the first and second substances,
the natural logarithm low-frequency trend of longitudinal wave velocity, the natural logarithm low-frequency trend of longitudinal and transverse wave velocity ratio and the natural logarithm low-frequency trend of density are respectively represented.
8. The method for prestack inversion based on γ _ Zoeppritz equation as claimed in claim 1, wherein said step 5 specifically comprises:
step 51: solving an inversion target function by using an improved iterative reweighted least square algorithm to obtain an inversion result of the three-elastic-parameter natural logarithm;
step 52: and performing indexing on the inversion result of the three-elastic-parameter natural logarithm to obtain a three-elastic-parameter inversion result.
9. The method of pre-stack inversion based on γ _ Zoeppritz equation of claim 8, wherein in step 51, the modified iterative reweighed least squares algorithm is:
step 511: low frequency trend m of three elastic parameter natural logarithmpAs an inverse initial model m0；
Step 512: according to m0Calculating an AVO forward modeling matrix G, a matrix H and a non-uniformity weighting matrix Q;
step 513: solving by using Cholesky decomposition and upper and lower triangular matrix decomposition: (G)TG+λQ+κΩTΩ)m＝GTH+κΩTmpObtaining an inversion result m of the iteration;
step 514: taking m as an initial model m of the next iteration0；
Step 515: repeating the step 512 to the step 514 until the specified iteration times are reached, and outputting an inversion result m;
wherein the non-uniformity weighting matrix Q is:
10. a system for prestack inversion based on the γ _ Zoeppritz equation, the system comprising:
a memory storing computer-executable instructions;
a processor executing computer executable instructions in the memory to perform the steps of:
step 1: constructing a low-frequency soft constraint term;
step 2: constructing a seismic data fitting difference term;
and step 3: constructing a reflection coefficient sparse constraint term;
and 4, step 4: constructing an inversion target function according to the seismic data fitting difference term, the low-frequency soft constraint term and the reflection coefficient sparse constraint term;
and 5: and obtaining an elastic parameter inversion result according to the inversion target function.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910571021.1A CN112147681A (en) | 2019-06-28 | 2019-06-28 | Pre-stack inversion method and system based on gamma _ Zoeppritz equation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910571021.1A CN112147681A (en) | 2019-06-28 | 2019-06-28 | Pre-stack inversion method and system based on gamma _ Zoeppritz equation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112147681A true CN112147681A (en) | 2020-12-29 |
Family
ID=73869226
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910571021.1A Pending CN112147681A (en) | 2019-06-28 | 2019-06-28 | Pre-stack inversion method and system based on gamma _ Zoeppritz equation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112147681A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113640871A (en) * | 2021-08-10 | 2021-11-12 | 成都理工大学 | Seismic wave impedance inversion method based on heavily-weighted L1 norm sparse constraint |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2141515A1 (en) * | 2008-07-03 | 2010-01-06 | Ifp | Method for the joint inversion of seismic data represented on different time scales |
| CN102692645A (en) * | 2012-06-01 | 2012-09-26 | 中国石油集团川庆钻探工程有限公司地球物理勘探公司 | Method for performing joint inversion on P-wave and S-wave velocity ratio of reservoir by utilizing P-wave and converted wave data |
| CN109212599A (en) * | 2017-06-30 | 2019-01-15 | 中国石油化工股份有限公司 | A kind of prestack Simultaneous Retrieving method of seismic data |
-
2019
- 2019-06-28 CN CN201910571021.1A patent/CN112147681A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2141515A1 (en) * | 2008-07-03 | 2010-01-06 | Ifp | Method for the joint inversion of seismic data represented on different time scales |
| CN102692645A (en) * | 2012-06-01 | 2012-09-26 | 中国石油集团川庆钻探工程有限公司地球物理勘探公司 | Method for performing joint inversion on P-wave and S-wave velocity ratio of reservoir by utilizing P-wave and converted wave data |
| CN109212599A (en) * | 2017-06-30 | 2019-01-15 | 中国石油化工股份有限公司 | A kind of prestack Simultaneous Retrieving method of seismic data |
Non-Patent Citations (4)
| Title |
|---|
| FANG YUAN ET AL.: "Generalized linear joint PP–PS inversion based on two constraints", 《APPLIED GEOPHYSICS》 * |
| 张丰麒等: "联合PP、PS波叠前AVA反演方法", 《物探与化探》 * |
| 张丰麒等: "贝叶斯三参数低频软约束同步反演", 《石油地球物理勘探》 * |
| 方圆等: "基于YPD-Zoeppritz方程的杨氏模量和泊松比直接反演方法", 《石油物探》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113640871A (en) * | 2021-08-10 | 2021-11-12 | 成都理工大学 | Seismic wave impedance inversion method based on heavily-weighted L1 norm sparse constraint |
| CN113640871B (en) * | 2021-08-10 | 2023-09-01 | 成都理工大学 | Seismic wave impedance inversion method based on re-weighted L1 norm sparse constraint |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA3043310C (en) | Method for estimating petrophysical properties for single or multiple scenarios from several spectrally variable seismic and full wavefield inversion products | |
| CN102466816B (en) | Inversion method for stratum elasticity constant parameter of pre-stack seismic data | |
| EP3028071B1 (en) | Method and device for the generation and application of anisotropic elastic parameters in horizontal transverse isotropic (hti) media | |
| KR101548976B1 (en) | Estimation of soil properties using waveforms of seismic surface waves | |
| US6901333B2 (en) | Method and device for the generation and application of anisotropic elastic parameters | |
| EP1746443B1 (en) | Method of estimating elastic parameters and rock composition of underground formations using seismic data | |
| US8379482B1 (en) | Using seismic attributes for data alignment and seismic inversion in joint PP/PS seismic analysis | |
| Pérez et al. | High-resolution prestack seismic inversion using a hybrid FISTA least-squares strategy | |
| US20040186667A1 (en) | Source-independent full waveform inversion of seismic data | |
| EP2160633B1 (en) | Creating an absorption parameter model | |
| US9348049B2 (en) | Simultaneous joint estimation of the P-P and P-S residual statics | |
| US20050273266A1 (en) | Seismic event correlation and Vp-Vs estimation | |
| CN111025387B (en) | Pre-stack earthquake multi-parameter inversion method for shale reservoir | |
| CN103954992B (en) | A kind of the Method of Deconvolution and device | |
| CN101201409B (en) | Method for revising earthquake data phase | |
| CN110895348B (en) | Method, system and storage medium for extracting low-frequency information of seismic elastic impedance | |
| EP3273274A1 (en) | Device and method for estimating pre-stack wavelet model from seismic gathers | |
| CN114924314A (en) | Seismic shear transverse wave inversion model construction method, porosity prediction method and apparatus | |
| CN113552624B (en) | Porosity prediction method and device | |
| CN112147681A (en) | Pre-stack inversion method and system based on gamma _ Zoeppritz equation | |
| CN112147683B (en) | Pre-stack sparse layer inversion method and system based on rock physical relationship constraint | |
| US6778907B1 (en) | Method for estimation of propagation paths for seismic signals | |
| CN110988991B (en) | Elastic parameter inversion method, device and system | |
| CN115327625A (en) | Reservoir lithology identification method | |
| CN116381793B (en) | Pre-stack inversion method and device for structure TV regularized joint inter-channel difference constraint |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201229 |