CN116027406B - Multi-channel simultaneous inversion identification method, device and medium for improving inversion resolution - Google Patents
Multi-channel simultaneous inversion identification method, device and medium for improving inversion resolution Download PDFInfo
- Publication number
- CN116027406B CN116027406B CN202310091703.9A CN202310091703A CN116027406B CN 116027406 B CN116027406 B CN 116027406B CN 202310091703 A CN202310091703 A CN 202310091703A CN 116027406 B CN116027406 B CN 116027406B
- Authority
- CN
- China
- Prior art keywords
- inversion
- representing
- seismic
- frequency domain
- noise
- 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.)
- Active
Links
Landscapes
- Geophysics And Detection Of Objects (AREA)
Abstract
The application discloses a multi-channel simultaneous inversion identification method, a multi-channel simultaneous inversion identification device and a multi-channel simultaneous inversion medium for improving inversion resolution. The time domain seismic inversion method is superior to the frequency domain seismic inversion method in anti-noise, but the frequency domain seismic inversion method is superior to the time domain seismic inversion method in resolution. The application combines the advantages of the seismic inversion method in the time domain and the frequency domain, and improves the resolution of the inversion result on the premise of ensuring noise resistance. Aiming at the problem that the traditional channel-by-channel inversion does not consider transverse information, a multi-channel simultaneous inversion method is introduced, so that the correlation between adjacent seismic channels is enhanced, and the transverse continuity of inversion results is improved. According to the background of the seismic data, the purposes of high noise resistance and high resolution can be achieved by adjusting the weight coefficients of the time domain and the frequency domain.
Description
Technical Field
The application relates to the technical field of seismic surveying, in particular to a multichannel simultaneous inversion identification method, device and medium for improving inversion resolution.
Background
In seismic exploration, the problem of seismic resolution has led to the development of seismic exploration technology. Resolution has a tremendous impact on seismic acquisition, processing and interpretation. Resolution and signal-to-noise ratio are two independent concepts, and are also important factors that affect high resolution and high precision surveys, respectively.
Typically, seismic inversion can only be performed in the frequency or time domain. Compared with a time domain inversion method, the frequency domain inversion method can improve the resolution of seismic inversion. But this is limited to the case where no noise or noise influence is small. When the random noise energy is too large, the noise immunity of the frequency domain seismic inversion method is poor, and the inversion result is quite unstable. Compared with frequency domain seismic inversion, the time domain seismic inversion method is more rapid in development, and the solution space of the parameters to be inverted can be effectively reduced by utilizing the information of the low-frequency model and regularization constraint. In time domain seismic inversion, the inversion method based on the convolution model can convert the seismic inversion problem into a linear optimization problem, and a relatively stable solution is obtained. Meanwhile, the method has low sensitivity to noise and high operation speed, so that the method is widely applied to geological structure detection and oil reservoir production tasks. However, the resolution of the inversion results depends on the signal-to-noise ratio of the seismic data, which limits the effectiveness of time domain seismic inversion.
In addition, conventional seismic inversion is realized through channel-by-channel inversion, so that correlation between adjacent seismic channels can be ignored to a great extent, and the problems of excessively low transverse resolution, small suppression capability for random noise and the like are caused.
Disclosure of Invention
The present application has been made to solve the above-mentioned problems occurring in the prior art. Therefore, a multi-channel simultaneous inversion identification method, device and medium for improving inversion resolution are needed, and in seismic inversion, inversion methods of different transformation domains have different characteristics. The time domain seismic inversion method is superior to the frequency domain seismic inversion method in anti-noise, but the frequency domain seismic inversion method is superior to the time domain seismic inversion method in resolution. The mixed seismic inversion method (the method provided by the application) combines the advantages of the seismic inversion method in the time domain and the frequency domain, and improves the resolution of the inversion result on the premise of guaranteeing noise resistance.
According to a first aspect of the present application, there is provided a multi-channel simultaneous inversion identification method for improving inversion resolution, the method comprising:
based on the convolution model, the seismic record is represented by the following formula (1):
（1）
in the method, in the process of the application,representing the seismic data and,representing a sequence of seismic reflection coefficients,the noise is represented by a characteristic of the noise,representing a wavelet matrix;
wherein the seismic reflection coefficients in the sequence of seismic reflection coefficients are represented by impedances:
（2）
in the method, in the process of the application,R i representing the first in the sequence of seismic reflection coefficientsiThe reflection coefficient of the individual earthquake,Z i+1 andZ i respectively represent the firsti+1And (b)iThe number of impedances, ln, represents the logarithmic calculation,Drepresenting a differential matrix;
is provided withL=lnZRepresenting the logarithm of the impedance, equation (1) is changed to:
（1）
a convolution model in the frequency domain is obtained by fourier transformation:
（4）
in the method, in the process of the application,representing the frequency spectrum of the seismic data,a spectrum representing a sequence of seismic reflection coefficients,the spectrum of the noise is represented and,representing the spectrum of the wavelet;
based on the formula (4), forward equations of the frequency domain and the time domain are obtained, as shown in the formula (5) and the formula (6):
（5）
（6）
in the method, in the process of the application,the wavelet kernel matrices are each in the time domain,、the frequency domain seismic records and the positive calculus submatrices,representing random noise;
based on the equation (5) and the equation (6), an objective function of the mixed domain is determined as shown in the equation (7):
（7）
in the method, in the process of the application,the weight coefficients representing the initial model are represented,representing a logarithmic form of the initial model;
introducing longitudinal and transverse differential operators into an objective function of the mixed domain, and realizing multi-channel simultaneous inversion:
（8）
in the method, in the process of the application,the weight coefficients representing the difference operator are represented,the L1 norm is represented by the expression,、respectively representing a transverse differential operator and a longitudinal differential operator;
the matrix operation in the L1 norm is replaced by Lagrangian multipliers, resulting in a constrained objective function as follows:
（9）
in the method, in the process of the application,、representing lagrangian multipliers;
introduction of a dual term、Converting the constrained objective function into an unconstrained objective function to obtain:
（10）
in the method, in the process of the application,a weight coefficient representing the dual term, which determines the degree of association between the Lagrangian multiplier and the difference operator;
based on a soft threshold shrinkage method, decomposing the formula (10) into a plurality of single constraint problem solutions:
（11）
（12）
（13）
（14）
（15）
according to the solving result, inverting to obtain impedance。
According to a second aspect of the present application, there is provided a multi-channel simultaneous inversion identification apparatus for improving inversion resolution, the apparatus comprising a processor configured to:
based on the convolution model, the seismic record is represented by the following formula (1):
（1）
in the method, in the process of the application,representing the seismic data and,representing a sequence of seismic reflection coefficients,the noise is represented by a characteristic of the noise,representing a wavelet matrix;
wherein the seismic reflection coefficients in the sequence of seismic reflection coefficients are represented by impedances:
（2）
in the method, in the process of the application,R i representing the first in the sequence of seismic reflection coefficientsiThe reflection coefficient of the individual earthquake,Z i+1 andZ i respectively represent the firsti+1And (b)iThe number of impedances, ln, represents the logarithmic calculation,Drepresenting a differential matrix;
is provided withL=lnZRepresenting the logarithm of the impedance, equation (1) is changed to:
（3）
a convolution model in the frequency domain is obtained by fourier transformation:
（4）
in the method, in the process of the application,representing the frequency spectrum of the seismic data,a spectrum representing a sequence of seismic reflection coefficients,the spectrum of the noise is represented and,representing the spectrum of the wavelet;
based on the formula (4), forward equations of the frequency domain and the time domain are obtained, as shown in the formula (5) and the formula (6):
（5）
（6）
in the method, in the process of the application,the wavelet kernel matrices are each in the time domain,、the frequency domain seismic records and the positive calculus submatrices,representing random noise;
based on the equation (5) and the equation (6), an objective function of the mixed domain is determined as shown in the equation (7):
（7）
in the method, in the process of the application,the weight coefficients representing the initial model are represented,representing a logarithmic form of the initial model;
introducing longitudinal and transverse differential operators into an objective function of the mixed domain, and realizing multi-channel simultaneous inversion:
（8）
in the method, in the process of the application,the weight coefficients representing the difference operator are represented,the L1 norm is represented by the expression,、respectively representing a transverse differential operator and a longitudinal differential operator;
the matrix operation in the L1 norm is replaced by Lagrangian multipliers, resulting in a constrained objective function as follows:
（9）
in the method, in the process of the application,、representing lagrangian multipliers;
introduction of a dual term、Converting the constrained objective function into an unconstrained objective function to obtain:
（10）
in the method, in the process of the application,a weight coefficient representing the dual term, which determines the space between the Lagrangian multiplier and the difference operatorIs a degree of association of (a);
based on a soft threshold shrinkage method, decomposing the formula (10) into a plurality of single constraint problem solutions:
（11）
（12）
（13）
（14）
（15）
according to the solving result, inverting to obtain impedance。
According to a third aspect of the present application, there is provided a multi-channel simultaneous inversion identification system for improving inversion resolution, the system comprising: a memory for storing a computer program; a processor for executing the computer program to implement the method as described above.
According to a fourth aspect of the application, there is provided a non-transitory computer readable storage medium storing instructions which, when executed by a processor, perform the method as described above.
The multi-channel simultaneous inversion identification method, the multi-channel simultaneous inversion identification device and the medium for improving inversion resolution according to the various schemes have at least the following technical effects:
the application can greatly enhance the recognition rate of the sensor to the hydrogen, reduce the influence of the interference gas components and improve the reliability of the sensor.
Drawings
In the drawings, which are not necessarily drawn to scale, like numerals may describe similar components in different views. The same reference numerals with letter suffixes or different letter suffixes may represent different instances of similar components. The accompanying drawings illustrate various embodiments by way of example in general and not by way of limitation, and together with the description and claims serve to explain the inventive embodiments. Wherever possible, the same reference numbers will be used throughout the drawings to refer to the same or like parts. Such embodiments are illustrative and not intended to be exhaustive or exclusive of the present apparatus or method.
FIG. 1 shows a schematic diagram of a time domain convolution model according to an embodiment of the present disclosure.
Fig. 2 shows a schematic diagram of a frequency domain convolution model according to an embodiment of the application.
FIG. 3 illustrates synthetic seismic record data in accordance with an embodiment of the application: (a) 10% noise; (b) 30% noise; (c) 50% noise.
Fig. 4 shows different noise frequency domain inversion results according to an embodiment of the application: (a) noiseless; (b) the noise is 10%; (c) the noise is 30%; (d) the noise was 50%.
FIG. 5 shows different noise time domain inversion results according to an embodiment of the application: (a) noiseless; (b) the noise is 10%; (c) the noise is 30%; (d) the noise was 50%.
FIG. 6 shows different noise mixture domain inversion results according to an embodiment of the application: (a) noiseless; (b) the noise is 10%; (c) the noise is 30%; (d) the noise was 50%.
FIG. 7 illustrates acoustic impedance in accordance with an embodiment of the present application: (a) a real model; (b) an initial model.
FIG. 8 shows the impedance inversion results for different methods according to embodiments of the application: (a) time domain single pass inversion; (b) time domain multi-channel inversion; (C) mixed domain single pass inversion; (d) mixed domain multi-channel inversion.
Fig. 9 shows absolute errors according to an embodiment of the application: (a) time domain single pass inversion; (B) mixed domain single pass inversion; (c) time domain multi-channel inversion; (d) mixed domain multi-channel inversion.
FIG. 10 shows a graph comparing signal-to-noise ratios for single-channel inversion with simultaneous inversion of multiple channels in accordance with an embodiment of the present application.
Detailed Description
The present application will be described in detail below with reference to the drawings and detailed description to enable those skilled in the art to better understand the technical scheme of the present application. Embodiments of the present application will be described in further detail below with reference to the drawings and specific examples, but not by way of limitation. The order in which the steps are described herein by way of example should not be construed as limiting if there is no necessity for a relationship between each other, and it should be understood by those skilled in the art that the steps may be sequentially modified without disrupting the logic of each other so that the overall process is not realized.
The embodiment of the application provides a multichannel simultaneous inversion identification method for improving inversion resolution, which can achieve the purposes of high noise resistance and high resolution by adjusting weight coefficients of a time domain and a frequency domain according to the background of seismic data. As in fig. 1 and 2, schematic diagrams of a time domain convolution model and a frequency domain convolution model are shown, respectively. Based on the time domain convolution model and the frequency domain convolution model, the method comprises the following steps:
(1) According to the convolution model proposed by Robinson, the seismic record can be expressed as the product of the formation reflection coefficient and the wavelet matrix:
（1）
in the method, in the process of the application,representing the seismic data and,representing a sequence of seismic reflection coefficients,the noise is represented by a characteristic of the noise,representing a wavelet matrix;
also, the reflection coefficient may be expressed in terms of impedance:
（2）
in the method, in the process of the application,R i representing the first in the sequence of seismic reflection coefficientsiThe reflection coefficient of the individual earthquake,Z i+1 andZ i respectively represent the firsti+1And (b)iThe number of impedances, ln, represents the logarithmic calculation,Drepresenting a differential matrix.
Is provided withL=lnZ. Thus, formula (1) can be written as:
（3）
(2) The Robinson convolution model is also applicable to the frequency domain. The convolution model in the frequency domain can be obtained by fourier transformation:
（4）
in the method, in the process of the application,representing the frequency spectrum of the seismic data,a spectrum representing a sequence of seismic reflection coefficients,the spectrum of the noise is represented and,representing the spectrum of the wavelet.
(3) Forward equations in the frequency and time domains can be derived as in equations (5) and (6):
（5）
（6）
in the method, in the process of the application,the wavelet kernel matrices are each in the time domain,、the frequency domain seismic records and the positive calculus submatrices,representing random noise.
(4) Thus, the objective function of the hybrid domain inversion can be expressed in the form of expression (7).Andthe weight coefficients of the time domain fidelity item and the frequency domain fidelity item control the proportion of the time domain fidelity item and the frequency domain fidelity item in the inversion process. In addition, the initial model constraint inversion is introduced to achieve the effect of improving inversion stability;
（7）
in the method, in the process of the application,the weight coefficients representing the initial model are represented,representing the logarithmic form of the initial model.
(5) Introducing longitudinal and transverse differential operators into an objective function to realize simultaneous inversion of multiple channels;
（8）
in the method, in the process of the application,the weight coefficients representing the difference operator are represented,the L1 norm is represented by the expression,、representing a lateral differential operator and a longitudinal differential operator, respectively.
(6) The matrix operation in the L1 norm is replaced by Lagrangian multipliers, resulting in a constrained objective function as follows:
（9）
in the method, in the process of the application,、representing the lagrangian multiplier.
(7) Next, a dual item is introduced、Converting a constrained objective function into an unconstrained objective function to obtain the followingResults:
（10）
in the method, in the process of the application,a weight coefficient representing the dual term, which determines the degree of association between the Lagrangian multiplier and the difference operator;
(8) Decomposing the equation (10) into a plurality of single constraint problem solutions by using an ADMM algorithm soft threshold shrinkage algorithm pair type:
（11）
（12）
（13）
（14）
（15）
(9) Finally, inverting to obtain impedance。
Application example:
one-dimensional experiment:
and extracting a piece of seismic trace data from the Marmousi2 model, and verifying the noise immunity and feasibility of the mixed domain inversion method. 10%, 30% and 50% gaussian white noise were added to the synthetic seismic record of fig. 3, respectively. The black line is the true value of the synthetic seismic record and the dashed line is the synthetic seismic record after noise addition. Fig. 4, 5, and 6 show the frequency domain, time domain, and mixed domain impedance inversion results, respectively, for different noise contexts. The black line is the true value, the dotted line is the initial low frequency model, and the dashed line is the inverse value. The inversion results of the three methods can well restore the real situation of the impedance without noise interference (fig. 4a, 5a, 6 a). As the noise increases, we can get the following information from fig. 4, 5 and 6 (marked with a dashed oval): although the time domain inversion result can keep better stability, the accuracy of the formation characterization gradually decreases. For the frequency domain inversion results, although the resolution of the formation is high, the inversion results highlight unstable phenomena as the noise increases. However, the mixed domain inversion results inherit both the high stability of the time domain and the high resolution of the frequency domain.
Two-dimensional experiment:
the stability and transverse continuity of the mixed domain multi-channel simultaneous inversion method are further verified through a two-dimensional experiment. Fig. 7a is a real model of acoustic impedance and fig. 7b is an initial model obtained by gaussian low pass filtering. To fully confirm the conclusion that the mixed domain inversion method is superior to the time domain inversion method, 30% of Gaussian white noise is added to the two-dimensional model for inversion, and the inversion result is shown in FIG. 8. In the presence of noise, the inversion effect of the frequency domain is poor, so we do not give the inversion result of the frequency domain here. Under the influence of noise, single-channel inversion suffers from severe vertical streak due to the lack of correlation between adjacent channels. The simultaneous inversion of multiple channels can well solve the problem, and the transverse continuity of inversion is improved. Fig. 9 shows the absolute error of the inversion result, and it can be seen that the multi-channel simultaneous inversion accuracy is high. The results show that the hybrid domain inversion method is superior to the time domain inversion method even in the case of noise interference. Fig. 10 shows a signal-to-noise ratio (SNR) comparison of the inversion results. The result shows that the mixed domain multi-channel simultaneous inversion method has higher accuracy and reliability.
The embodiment of the application also provides a multi-channel simultaneous inversion identification device for improving inversion resolution, which comprises a processor, wherein the processor is configured to:
based on the convolution model, the seismic record is represented by the following formula (1):
（1）
in the method, in the process of the application,representing the seismic data and,representing a sequence of seismic reflection coefficients,the noise is represented by a characteristic of the noise,representing a wavelet matrix;
wherein the seismic reflection coefficients in the sequence of seismic reflection coefficients are represented by impedances:
（2）
in the method, in the process of the application,R i representing the first in the sequence of seismic reflection coefficientsiThe reflection coefficient of the individual earthquake,Z i+1 andZ i respectively represent the firsti+1And (b)iThe number of impedances, ln, represents the logarithmic calculation,Drepresenting a differential matrix;
is provided withL=lnZRepresenting the logarithm of the impedance, equation (1) is changed to:
（1）
a convolution model in the frequency domain is obtained by fourier transformation:
（4）
in the method, in the process of the application,representing the frequency spectrum of the seismic data,a spectrum representing a sequence of seismic reflection coefficients,the spectrum of the noise is represented and,representing the spectrum of the wavelet;
based on the formula (4), forward equations of the frequency domain and the time domain are obtained, as shown in the formula (5) and the formula (6):
（5）
（6）
in the method, in the process of the application,the wavelet kernel matrices are each in the time domain,、the frequency domain seismic records and the positive calculus submatrices,representing random noise;
based on the equation (5) and the equation (6), an objective function of the mixed domain is determined as shown in the equation (7):
（7）
in the method, in the process of the application,the weight coefficients representing the initial model are represented,representing a logarithmic form of the initial model;
introducing longitudinal and transverse differential operators into an objective function of the mixed domain, and realizing multi-channel simultaneous inversion:
（8）
in the method, in the process of the application,the weight coefficients representing the difference operator are represented,the L1 norm is represented by the expression,、respectively representing a transverse differential operator and a longitudinal differential operator;
the matrix operation in the L1 norm is replaced by Lagrangian multipliers, resulting in a constrained objective function as follows:
（9）
in the method, in the process of the application,、representing lagrangian multipliers;
introduction ofDual items、Converting the constrained objective function into an unconstrained objective function to obtain:
（10）
in the method, in the process of the application,a weight coefficient representing the dual term, which determines the degree of association between the Lagrangian multiplier and the difference operator;
based on a soft threshold shrinkage method, decomposing the formula (10) into a plurality of single constraint problem solutions:
（11）
（12）
（13）
（14）
（15）
according to the solving result, inverting to obtain impedance。
The multi-channel simultaneous inversion identification device for improving inversion resolution provided by the embodiment of the application belongs to the same technical conception as the method explained before, and has the technical effects basically consistent, and is not repeated here.
The embodiment of the application also provides a multi-channel simultaneous inversion identification system for improving inversion resolution, which comprises the following steps:
a memory for storing a computer program;
a processor for executing the computer program to implement the method of any of the embodiments of the application.
Embodiments of the present application also provide a non-transitory computer readable medium storing instructions which, when executed by a processor, perform a method according to any of the embodiments of the present application.
Furthermore, although exemplary embodiments have been described herein, the scope thereof includes any and all embodiments having equivalent elements, modifications, omissions, combinations (e.g., of the various embodiments across), adaptations or alterations as pertains to the present application. The elements in the claims are to be construed broadly based on the language employed in the claims and are not limited to examples described in the present specification or during the practice of the application, which examples are to be construed as non-exclusive. It is intended, therefore, that the specification and examples be considered as exemplary only, with a true scope and spirit being indicated by the following claims and their full scope of equivalents.
The above description is intended to be illustrative and not restrictive. For example, the above-described examples (or one or more aspects thereof) may be used in combination with each other. For example, other embodiments may be used by those of ordinary skill in the art upon reading the above description. In addition, in the above detailed description, various features may be grouped together to streamline the application. This is not to be interpreted as an intention that the features of the claimed application are essential to any of the claims. Rather, inventive subject matter may lie in less than all features of a particular inventive embodiment. Thus, the following claims are hereby incorporated into the detailed description as examples or embodiments, with each claim standing on its own as a separate embodiment, and it is contemplated that these embodiments may be combined with one another in various combinations or permutations. The scope of the application should be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled.
Claims (4)
1. A multi-channel simultaneous inversion identification method for improving inversion resolution, the method comprising:
based on the convolution model, the seismic record is represented by the following formula (1):
S＝WR+N (1)
wherein S represents seismic data, R represents a seismic reflection coefficient sequence, N represents noise, and W represents a wavelet matrix;
wherein the seismic reflection coefficients in the sequence of seismic reflection coefficients are represented by impedances:
wherein R is i Representing the ith seismic reflection coefficient, Z, in the sequence of seismic reflection coefficients i+1 And Z i Respectively representing the (i+1) th impedance and the (i) th impedance, ln (·) represents logarithmic calculation, and D represents a differential matrix;
let l=ln Z denote the logarithm of the impedance, change equation (1) to:
S＝WDL+N (3)
a convolution model in the frequency domain is obtained by fourier transformation:
S(ω)＝W(ω)R(ω)+N(ω) (4)
where S (ω) represents the spectrum of the seismic data, R (ω) represents the spectrum of the sequence of seismic reflection coefficients, N (ω) represents the spectrum of the noise, and W (ω) represents the spectrum of the wavelet;
based on the formula (4), forward equations of the frequency domain and the time domain are obtained, as shown in the formula (5) and the formula (6):
S＝GL+N (5)
S'＝G'L+N F (6)
wherein G=WD is wavelet kernel matrix of time domain, S 'and G' are seismic record and positive operation submatrix of frequency domain, N F Representing random noise;
based on the equation (5) and the equation (6), an objective function of the mixed domain is determined as shown in the equation (7):
wherein, beta represents the weight coefficient of the initial model, L 0 Representing the logarithmic form of the initial model, alpha 1 Weight coefficient, alpha, representing time domain fidelity term 2 A weight coefficient representing a frequency domain fidelity term;
introducing longitudinal and transverse differential operators into an objective function of the mixed domain, and realizing multi-channel simultaneous inversion:
where, γ represents the weight coefficient of the difference operator,represents L1 norm, D x 、D y Respectively representing a transverse differential operator and a longitudinal differential operator;
the matrix operation in the L1 norm is replaced by Lagrangian multipliers, resulting in a constrained objective function as follows:
wherein d x 、d y Representing lagrangian multipliers;
introduction of the dual term C x 、C y Converting the constrained objective function into an unconstrained objective function to obtain:
wherein λ represents a weight coefficient of the dual term, which determines a degree of association between the lagrangian multiplier and the difference operator;
based on a soft threshold shrinkage method, decomposing the formula (10) into a plurality of single constraint problem solutions:
from the solution, the inversion yields the impedance z=exp (L).
2. A multi-channel simultaneous inversion identification apparatus for improving inversion resolution, the apparatus comprising a processor configured to:
based on the convolution model, the seismic record is represented by the following formula (1):
S＝WR+N (1)
wherein S represents seismic data, R represents a seismic reflection coefficient sequence, N represents noise, and W represents a wavelet matrix;
wherein the seismic reflection coefficients in the sequence of seismic reflection coefficients are represented by impedances:
wherein R is i Representing the ith seismic reflection coefficient, Z, in the sequence of seismic reflection coefficients i+1 And Z i Respectively representing the (i+1) th impedance and the (i) th impedance, ln represents logarithmic calculation, and D represents a differential matrix;
let l=ln Z denote the logarithm of the impedance, change equation (1) to:
S＝WDL+N (3)
a convolution model in the frequency domain is obtained by fourier transformation:
S(ω)＝W(ω)R(ω)+N(ω) (4)
where S (ω) represents the spectrum of the seismic data, R (ω) represents the spectrum of the sequence of seismic reflection coefficients, N (ω) represents the spectrum of the noise, and W (ω) represents the spectrum of the wavelet;
based on the formula (4), forward equations of the frequency domain and the time domain are obtained, as shown in the formula (5) and the formula (6):
S＝GL+N (5)
S'＝G'L+N F (6)
wherein G=WD is wavelet kernel matrix of time domain, S 'and G' are seismic record and positive operation submatrix of frequency domain, N F Representing random noise;
based on the equation (5) and the equation (6), an objective function of the mixed domain is determined as shown in the equation (7):
wherein, beta represents the weight coefficient of the initial model, L 0 Representing an initial modulusLogarithmic form of form alpha 1 Weight coefficient, alpha, representing time domain fidelity term 2 A weight coefficient representing a frequency domain fidelity term;
introducing longitudinal and transverse differential operators into an objective function of the mixed domain, and realizing multi-channel simultaneous inversion:
where, γ represents the weight coefficient of the difference operator,represents L1 norm, D x 、D y Respectively representing a transverse differential operator and a longitudinal differential operator;
the matrix operation in the L1 norm is replaced by Lagrangian multipliers, resulting in a constrained objective function as follows:
wherein d x 、d y Representing lagrangian multipliers;
introduction of the dual term C x 、C y Converting the constrained objective function into an unconstrained objective function to obtain:
wherein λ represents a weight coefficient of the dual term, which determines a degree of association between the lagrangian multiplier and the difference operator;
based on a soft threshold shrinkage method, decomposing the formula (10) into a plurality of single constraint problem solutions:
from the solution, the inversion yields the impedance z=exp (L).
3. A multichannel simultaneous inversion identification system for improving inversion resolution is characterized in that: the system comprises:
a memory for storing a computer program;
a processor for executing the computer program to implement the method of claim 1.
4. A non-transitory computer readable storage medium storing instructions which, when executed by a processor, perform the method of claim 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310091703.9A CN116027406B (en) | 2023-02-10 | 2023-02-10 | Multi-channel simultaneous inversion identification method, device and medium for improving inversion resolution |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310091703.9A CN116027406B (en) | 2023-02-10 | 2023-02-10 | Multi-channel simultaneous inversion identification method, device and medium for improving inversion resolution |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116027406A CN116027406A (en) | 2023-04-28 |
| CN116027406B true CN116027406B (en) | 2023-08-18 |
Family
ID=86075821
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310091703.9A Active CN116027406B (en) | 2023-02-10 | 2023-02-10 | Multi-channel simultaneous inversion identification method, device and medium for improving inversion resolution |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116027406B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016075550A1 (en) * | 2014-11-14 | 2016-05-19 | Cgg Services Sa | Device and method for weighted sparse inversion for seismic processing |
| CN107589448A (en) * | 2017-07-13 | 2018-01-16 | 西安交通大学 | A kind of multitrace seismogram reflection coefficient sequence Simultaneous Inversion method |
| WO2018201114A1 (en) * | 2017-04-28 | 2018-11-01 | Pioneer Natural Resources Usa, Inc. | High resolution seismic data derived from pre-stack inversion and machine learning |
| CN111025395A (en) * | 2020-01-06 | 2020-04-17 | 中国石油化工股份有限公司 | Method for improving thin inter-layer resolution based on compressed sensing Gaussian frequency domain |
| CN111856568A (en) * | 2019-04-30 | 2020-10-30 | 中国石油天然气股份有限公司 | MWV model-based frequency domain multi-channel reflection coefficient joint inversion method and system |
| CN112649857A (en) * | 2019-10-11 | 2021-04-13 | 中国石油化工股份有限公司 | Fluid factor direct inversion method and system based on pre-stack seismic data |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112596107B (en) * | 2020-12-11 | 2021-09-28 | 中国科学院地质与地球物理研究所 | Elastic parameter inversion method and device |
-
2023
- 2023-02-10 CN CN202310091703.9A patent/CN116027406B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016075550A1 (en) * | 2014-11-14 | 2016-05-19 | Cgg Services Sa | Device and method for weighted sparse inversion for seismic processing |
| WO2018201114A1 (en) * | 2017-04-28 | 2018-11-01 | Pioneer Natural Resources Usa, Inc. | High resolution seismic data derived from pre-stack inversion and machine learning |
| CN107589448A (en) * | 2017-07-13 | 2018-01-16 | 西安交通大学 | A kind of multitrace seismogram reflection coefficient sequence Simultaneous Inversion method |
| CN111856568A (en) * | 2019-04-30 | 2020-10-30 | 中国石油天然气股份有限公司 | MWV model-based frequency domain multi-channel reflection coefficient joint inversion method and system |
| CN112649857A (en) * | 2019-10-11 | 2021-04-13 | 中国石油化工股份有限公司 | Fluid factor direct inversion method and system based on pre-stack seismic data |
| CN111025395A (en) * | 2020-01-06 | 2020-04-17 | 中国石油化工股份有限公司 | Method for improving thin inter-layer resolution based on compressed sensing Gaussian frequency domain |
Non-Patent Citations (1)
| Title |
|---|
| 井间地震时频域联合速度和衰减系数同步层析反演;刘浩杰 等;《物探化探计算技术》;第41卷(第2期);158-165 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116027406A (en) | 2023-04-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wu et al. | White noise attenuation of seismic trace by integrating variational mode decomposition with convolutional neural network | |
| CN113687414B (en) | Data-augmentation-based seismic interbed multiple suppression method for convolutional neural network | |
| CN105116442A (en) | Lithologic oil-gas reservoir weak-reflection seismic signal reconstruction method | |
| CN107179550B (en) | A kind of seismic signal zero phase deconvolution method of data-driven | |
| CN112394414B (en) | Two-step seismic diffraction wave field prestack separation method | |
| CN110687597B (en) | Wave impedance inversion method based on joint dictionary | |
| Saad et al. | Unsupervised deep learning for single-channel earthquake data denoising and its applications in event detection and fully automatic location | |
| CN113077386A (en) | Seismic data high-resolution processing method based on dictionary learning and sparse representation | |
| CN115146667A (en) | Multi-scale seismic noise suppression method based on curvelet transform and multi-branch deep self-coding | |
| Wang et al. | Denoising with weak signal preservation by group-sparsity transform learning | |
| CN112526599A (en) | Wavelet phase estimation method and system based on weighted L1 norm sparsity criterion | |
| Yuan et al. | Attenuation of linear noise based on denoising convolutional neural network with asymmetric convolution blocks | |
| CN110146923A (en) | A kind of efficient high accuracy depth domain methods of seismic wavelet extraction | |
| CN116027406B (en) | Multi-channel simultaneous inversion identification method, device and medium for improving inversion resolution | |
| CN117574062A (en) | Small loop transient electromagnetic signal denoising method based on VMD-DNN model | |
| Jiang et al. | Seismic wavefield information extraction method based on adaptive local singular value decomposition | |
| CN115267911B (en) | Model and data driving deep learning-based earthquake multiple suppression method | |
| CN109857979A (en) | A kind of log approximating method based on wavelet analysis and BP neural network | |
| CN114185095B (en) | Method for suppressing multiple waves of three-dimensional plane wave domain seismic data | |
| CN115453629A (en) | Waveform correction method and device based on time-frequency domain adaptive filtering | |
| CN111856559B (en) | Multi-channel seismic spectrum inversion method and system based on sparse Bayes learning theory | |
| CN112363217A (en) | Random noise suppression method and system for seismic data | |
| CN115685340A (en) | Method and device for improving seismic characterization precision of thin-layer structure | |
| CN115236744A (en) | Method for strong shielding stripping and weak signal energy compensation of igneous rock | |
| CN111352159B (en) | Nuclear norm and generalized total variation joint constraint seismic random noise suppression method |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |