Seismic inversion method, electronic device and storage medium

Abstract

Embodiments of the present application provide a seismic inversion method, an electronic device and a storage medium, and relate to the field of seismic interpretation. The method comprises: performing well-seismic calibration processing according to logging data and time-domain seismic records to obtain time-domain logging data; obtaining frequency-domain seismic records according to the time-domain seismic records; extracting horizon information from the time-domain seismic records and generating an elastic impedance initial model according to the time-domain logging data and the horizon information; constructing a forward operator to synthesize pre-stack time-domain seismic records and pre-stack frequency-domain seismic records according to the elastic impedance initial model and the time-domain seismic records; performing multi-objective gradient descent solving processing to obtain a Pareto solution set; performing sorting decision processing to obtain a target solution and outputting the target solution as a pre-stack elastic impedance inversion result. By constructing an inversion basis and combining time-frequency domain data to perform multi-objective gradient descent solving and sorting, the problem of single domain control is avoided, the robustness of the inversion result is ensured, and accurate reservoir feature description under deep high noise is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of seismic interpretation, and in particular to a seismic inversion method, an electronic device and a storage medium. BACKGROUND

[0002] With the continuous growth of global oil and gas resource demand and the progress of exploration technology, deep oil and gas reservoirs have gradually become the focus of exploration and development. However, the complex deep geological structure, low signal-to-noise ratio of seismic data and other problems seriously affect the accuracy of reservoir characterization.

[0003] The prior art is based on pre-stack elastic impedance inversion, which describes reservoir properties by extending the Aki-Richards equation combined with multi-angle seismic information. Mixed time-frequency domain inversion further integrates time domain and frequency domain data, uses two-norm to construct the objective function, and optimizes the inversion result by gradient descent algorithm. The prior art relies on multi-domain data cooperation, but the inversion process still mainly relies on traditional optimization algorithm.

[0004] Because the two-norm objective function is sensitive to the order of magnitude difference between time domain and frequency domain data, the time domain data are concentrated in the reflection layer arrival time (amplitude value is large), which leads to the inversion result excessively relying on time domain information; at the same time, the gradient descent algorithm does not effectively balance the contribution of multi-domain data, which weakens the frequency domain noise resistance advantage. Finally, the inversion result is dominated by a single domain, and it is difficult to achieve high precision and high resolution in high noise deep reservoirs. Therefore, the existing mixed time-frequency domain inversion method has the defect of insufficient accuracy of reservoir characterization in deep high noise environment, which restricts the efficiency and accuracy of deep oil and gas reservoir exploration and development. SUMMARY

[0005] The embodiments of the present application provide a seismic inversion method, an electronic device and a storage medium to achieve accurate reservoir characterization in deep high noise environment.

[0006] In a first aspect, the embodiments of the present application provide a seismic inversion method, comprising:

[0007] Obtaining well logging data and time domain seismic records, and performing well-to-seismic calibration processing according to the well logging data to obtain time domain well logging data;

[0008] Performing data conversion processing according to the time domain seismic records to obtain frequency domain seismic records;

[0009] Extracting horizon information from the time domain seismic records, and generating an elastic impedance initial model according to the time domain well logging data and the horizon information;

[0010] According to the elastic impedance initial model and the time domain seismic records, constructing a forward operator;

[0011] synthesizing pre-stack time domain seismic records according to the seismic wavelet, the elastic impedance initial model and the forward operator;

[0012] synthesizing pre-stack frequency domain seismic records according to the seismic wavelet, the elastic initial impedance initial model and the forward operator;

[0013] performing multi-objective gradient descent solving processing according to the pre-stack time domain seismic records, the pre-stack frequency domain seismic records, the forward operator, the elastic impedance initial model and a pre-built initial objective function, to obtain a Pareto solution set;

[0014] performing sorting decision processing according to the Pareto solution set, to obtain a target solution, and outputting the target solution as a pre-stack elastic impedance inversion result.

[0015] In a possible implementation, the well-seismic calibration processing according to the logging data to obtain time domain logging data comprises:

[0016] performing data extraction processing according to the logging data to obtain P-wave velocity, S-wave velocity and density curves;

[0017] performing conversion processing according to the P-wave velocity, S-wave velocity and density curves according to a time-depth relationship conversion curve to obtain time domain logging data.

[0018] In a possible implementation, the calculation expression of the forward operator constructed according to the elastic impedance initial model and the time domain seismic records is:

[0019]

[0020] In the formula, represents the time domain seismic records, represents the forward operator, represents the elastic impedance initial model, represents random noise.

[0021] In a possible implementation, the pre-built initial objective function is:

[0022]

[0023] In the formula, represents the time domain objective function, represents the frequency domain objective function, represents the elastic impedance initial model, , , represents the time domain seismic records, represents the frequency domain seismic records, the frequency domain seismic records being obtained by performing Fourier transform on the time domain seismic records,​​ denotes a time-domain forward operator, denotes a frequency-domain forward operator, denotes the cross-correlation of the two data in the brackets, denotes a balancing factor between the time domain and the frequency domain, denotes an initial model constraint.

[0024] In a possible implementation, the Pareto solution set includes a plurality of non-dominated solutions and a target function value of each non-dominated solution on a corresponding target function;

[0025] Accordingly, the sorting decision processing according to the Pareto solution set to obtain a target solution includes:

[0026] constructing a solution set matrix according to all non-dominated solutions and all target function values;

[0027] determining a probability matrix and information entropy according to the solution set matrix;

[0028] determining a difference degree and a weight reflected by the entropy according to the probability matrix and the information entropy;

[0029] determining a first weighted distance and a second weighted distance of each non-dominated solution according to the solution set matrix, the difference degree and the weight reflected by the entropy;

[0030] calculating a sorting score of each non-dominated solution by a non-dominated solution distance method according to the first weighted distance and the second weighted distance of each non-dominated solution;

[0031] determining a non-dominated solution as a target solution from all non-dominated solutions according to the sorting scores of all non-dominated solutions.

[0032] In a possible implementation, the determining of the first weighted distance and the second weighted distance of each non-dominated solution according to the solution set matrix, the difference degree and the weight reflected by the entropy includes:

[0033] determining an ideal solution and a negative ideal solution of each target function according to a plurality of target function values in the solution set matrix;

[0034] determining the first weighted distance of each non-dominated solution according to the solution set matrix and the ideal solution of each target function, and determining the second weighted distance of each non-dominated solution according to the solution set matrix, the negative ideal solution of each target function, the difference degree and the weight reflected by the entropy.

[0035] In a possible implementation, the calculation formula of the determining of the first weighted distance of each non-dominated solution according to the solution set matrix and the ideal solution of each target function, and the determining of the second weighted distance of each non-dominated solution according to the solution set matrix, the negative ideal solution of each target function, the difference degree and the weight reflected by the entropy is:

[0036]

[0037] wherein, denotes the first weighted distance, denotes the second weighted distance, = 2, denotes the objective function value after the range normalization processing, denotes the weight of the entropy reflection, denotes the ideal solution, denotes the negative ideal solution.

[0038] In a possible implementation, the non-dominated solution is determined from all non-dominated solutions as the target solution according to the ranking scores of all non-dominated solutions, and the method comprises the following steps:

[0039] The non-dominated solutions are ranked according to the ranking scores of all non-dominated solutions to obtain a non-dominated solution sequence;

[0040] A non-dominated solution with the largest ranking score is obtained from the non-dominated solution sequence as the target solution.

[0041] In a second aspect, an embodiment of the present application provides a seismic inversion device, comprising:

[0042] An acquisition module is configured to acquire well logging data and time domain seismic records, and perform well-seismic calibration processing according to the well logging data to obtain time domain well logging data;

[0043] A conversion module is configured to perform data conversion processing according to the time domain seismic records to obtain frequency domain seismic records;

[0044] A model establishing module is configured to extract horizon information from the time domain seismic records, and generate an elastic impedance initial model according to the time domain well logging data and the horizon information;

[0045] A construction module is configured to construct a forward operator according to the elastic impedance initial model and the time domain seismic records;

[0046] A generation module is configured to synthesize pre-stack time domain seismic records according to a seismic wavelet, the elastic impedance initial model and the forward operator;

[0047] The generation module is further configured to synthesize pre-stack frequency domain seismic records according to a seismic wavelet, the elastic impedance initial model and the forward operator;

[0048] an analysis module, configured to perform a multi-objective gradient descent solving process according to the pre-stack time domain seismic record, the pre-stack frequency domain seismic record, the forward operator, the initial elastic impedance model and a pre-built initial objective function, to obtain a Pareto solution set;

[0049] a processing module, configured to perform a sorting decision process according to the Pareto solution set, to obtain a target solution, and output the target solution as a pre-stack elastic impedance inversion result.

[0050] In a possible implementation, the conversion module is specifically configured to:

[0051] perform data extraction processing according to the logging data, to obtain a P-wave velocity curve, a S-wave velocity curve and a density curve;

[0052] perform conversion processing according to the P-wave velocity curve, the S-wave velocity curve and the density curve according to a time-depth relationship conversion curve, to obtain time domain logging data.

[0053] In a possible implementation, the calculation expression of the forward operator constructed by the construction module is:

[0054]

[0055] in the formula, represents a time domain seismic record, represents a forward operator, represents an initial elastic impedance model, represents random noise.

[0056] In a possible implementation, the pre-built initial objective function in the analysis module is:

[0057]

[0058] in the formula, represents a time domain objective function, represents a frequency domain objective function, represents an initial elastic impedance model, , , represents a time domain seismic record, represents a frequency domain seismic record, which is obtained by performing Fourier transform on the time domain seismic record, represents a time domain forward operator, represents a frequency domain forward operator, represents mutual correlation of two data in parentheses, represents a balance factor between the time domain and the frequency domain, represents initial model constraint.

[0059] ​​In a possible implementation, the processing module is specifically configured to:

[0060] constructing a solution set matrix according to all non-dominated solutions and all objective function values;

[0061] determining a probability matrix and information entropy according to the solution set matrix;

[0062] determining an entropy-reflecting difference degree and a weight according to the probability matrix and the information entropy;

[0063] determining a first weighted distance and a second weighted distance of each non-dominated solution according to the solution set matrix, the entropy-reflecting difference degree and the weight;

[0064] calculating a ranking score of the non-dominated solution by using a non-dominated solution distance method according to the first weighted distance and the second weighted distance of each non-dominated solution;

[0065] determining a non-dominated solution as a target solution from all non-dominated solutions according to the ranking scores of all non-dominated solutions.

[0066] In a possible implementation, the processing module is further configured to:

[0067] determining an ideal solution and a negative ideal solution of each objective function according to the objective function values in the solution set matrix;

[0068] determining a first weighted distance of each non-dominated solution according to the solution set matrix and the ideal solution of each objective function, and determining a second weighted distance of each non-dominated solution according to the solution set matrix, the negative ideal solution of each objective function, the entropy-reflecting difference degree and the weight.

[0069] In a possible implementation, the calculation formula of the processing module for determining the second weighted distance of each non-dominated solution is as follows:

[0070]

[0071] wherein, the first weighted distance is denoted by d1, the second weighted distance is denoted by d2, = 2, the objective function value after range normalization processing is denoted by xij, the weight of the entropy reflection is denoted by w, the ideal solution is denoted by z+, the negative ideal solution is denoted by z-.

[0072] In a possible implementation, the processing module is further configured to:

[0073] performing ranking processing on all non-dominated solutions according to the ranking scores of all non-dominated solutions to obtain a non-dominated solution sequence.

[0074] A non-dominated solution with the maximum ranking score is obtained from the non-dominated solution sequence according to the ranking score matching as a target solution.

[0075] In a third aspect, an electronic device is provided, comprising: a memory, a processor;

[0076] The memory stores computer-executable instructions.

[0077] The processor executes the computer-executable instructions stored in the memory, so that the processor executes the first aspect and / or various possible implementation manners of the first aspect.

[0078] In a fourth aspect, a computer-readable storage medium is provided, and the computer-readable storage medium stores computer-executable instructions. When the computer-executable instructions are executed by a processor, the computer-executable instructions are used to implement the first aspect and / or various possible implementation manners of the first aspect.

[0079] In a fifth aspect, a computer program product is provided, comprising a computer program. When the computer program is executed by a processor, the computer program implements the first aspect and / or various possible implementation manners of the first aspect.

[0080] The seismic inversion method, the electronic device, and the storage medium provided in the embodiments of the present application can obtain well logging data and a time domain seismic record, perform well-to-seismic calibration processing according to the well logging data to obtain time domain well logging data, perform data conversion processing according to the time domain seismic record to obtain a frequency domain seismic record, extract horizon information from the time domain seismic record, and generate an elastic impedance initial model according to the time domain well logging data and the horizon information. The forward operator is constructed according to the elastic impedance initial model and the time domain seismic record. The prestack time domain seismic record is synthesized according to the seismic wavelet, the elastic impedance initial model, and the forward operator. The prestack frequency domain seismic record is synthesized according to the seismic wavelet, the elastic initial impedance initial model, and the forward operator. The multi-objective gradient descent solving processing is performed according to the prestack time domain seismic record, the prestack frequency domain seismic record, the forward operator, the elastic impedance initial model, and a pre-built initial objective function to obtain a Pareto solution set. The sorting decision processing is performed according to the Pareto solution set to obtain a target solution, and the target solution is output as a prestack elastic impedance inversion result. The well logging data is obtained and the well-to-seismic calibration and data conversion are performed to fully integrate the well logging data and the seismic data. The elastic impedance initial model and the forward operator are generated, and the inversion basis is constructed based on the physical principle. The multi-objective gradient descent solving and the sorting decision are performed in combination with the time-frequency domain data, the single-domain control problem is effectively avoided, the precision, the resolution, and the reliability of the inversion result are improved, the deep oil and gas reservoir characteristics can be more accurately described, and the robustness of the final inversion result is ensured while the accurate reservoir characteristic description is realized in the deep high-noise environment. Attached Figure Description

[0081] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0082] Figure 1 A flowchart illustrating the seismic inversion method provided in this application;

[0083] Figure 2 A schematic diagram of the elastic impedance model provided in this application;

[0084] Figure 3 A schematic diagram of time-domain seismic records provided for this application;

[0085] Figure 4 A schematic diagram of frequency domain seismic records provided for this application;

[0086] Figure 5 A schematic diagram of the initial model of elastic impedance provided in this application;

[0087] Figure 6 A schematic diagram of the elastic impedance inversion results provided in this application;

[0088] Figure 7 This is a schematic diagram of the weighted elastic impedance inversion results using existing technology.

[0089] Figure 8 A schematic diagram of the seismic inversion device provided in this application;

[0090] Figure 9 A schematic diagram of the structure of the electronic device provided in this application.

[0091] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0092] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0093] First, let me explain the terms used in this application:

[0094] Elastic Impedance (EI): A parameter that integrates rock P-wave velocity, S-wave velocity, density, and other elastic parameters, considering the effect of seismic wave incidence angle, to quantitatively represent the relationship between rock elastic characteristics and seismic response, and to more accurately correlate subsurface rock physical properties with seismic data, aiding in reservoir identification.

[0095] Well-to-seismic calibration: By matching well logs (such as acoustic and density) with seismic reflection characteristics, a corresponding relationship between well depth domain and seismic time domain is established, allowing well high-resolution information to be integrated into seismic data, providing a basis for subsequent inversion.

[0096] Forward operator (G): Based on seismic wave propagation principles, a mathematical mapping from subsurface elastic impedance model (m) to seismic record (d) is constructed to simulate seismic wave propagation in the subsurface and generate principle seismic records, which is the core link connecting the model and actual observation data.

[0097] Pareto Set: In multi-objective optimization, a set of non-dominated solutions. Any solution in the set is not completely dominated by other solutions in all objectives (such as time domain and frequency domain inversion errors), reflecting the optimal trade-off between different objectives and providing multiple decision options.

[0098] Entropy Weight TOPSIS: A decision-making algorithm that combines entropy weight method and TOPSIS. The entropy weight method determines the target weight based on data dispersion, and TOPSIS calculates the distance between solutions and ideal / negative ideal solutions to sort and select the best from the multi-objective solution set, achieving objective and unbiased decision-making.

[0099] Gradient Descent: An iterative optimization algorithm that calculates the gradient of the objective function and adjusts the model parameters (such as elastic impedance) in the negative direction of the gradient to minimize the objective function (such as time-frequency domain inversion error), achieving model optimization, which is a common method for solving inversion.

[0100] Time domain / frequency domain seismic record: Time domain records seismic wave amplitude changes with time as the axis, reflecting the timing characteristics of seismic wave propagation; frequency domain exhibits the distribution of seismic wave frequency components through Fourier transform, and the combination of the two can complement the noise resistance and resolution advantages.

[0101] Non-dominated solution: A solution in the Pareto set, which is not superior to other solutions in all objectives in multi-objective optimization, is a "better solution" under multi-objective trade-off, constitutes the Pareto front, and supports optimal solution selection.

[0102] In the prior art, a conventional mixed time-frequency domain seismic inversion method is used to construct a mismatch function by using a two-norm and solve a target function by using a gradient descent algorithm. Since the time domain data is of a much larger order of magnitude than the frequency domain data, and the two-norm target function further amplifies the order of magnitude difference, at the same time, the time domain energy is concentrated near the arrival time of the individual reflection layer to form a larger amplitude value, while the frequency domain energy is dispersed in different frequency components, and the amplitude value of each frequency point is relatively small. This causes the constructed target function to shift the optimization result to the time domain, and the inversion result is controlled by a single domain. In a deep high-noise environment, the advantages of the noise resistance of the time domain inversion and the high resolution of the frequency domain inversion result cannot be fully combined, which makes the reservoir feature description accuracy poor and difficult to meet the needs of fine geophysical data for deep oil and gas reservoir exploration and development.

[0103] The seismic inversion method provided in the present application fully integrates logging and seismic data by obtaining logging data and performing well-seismic calibration and data conversion; generates an elastic impedance initial model and a forward operator, and constructs an inversion basis based on physical principles; combines time-frequency domain data to solve and sort decisions by using a multi-target gradient descent, effectively avoids the problem of single domain control, improves the accuracy, resolution and reliability of the inversion result, can more accurately describe the reservoir characteristics of deep oil and gas reservoirs, and ensures the robustness of the final inversion result while achieving accurate reservoir feature description in a deep high-noise environment.

[0104] The technical solutions of the present application and how the technical solutions of the present application solve the above technical problems will be described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes can not be described again in some embodiments. The embodiments of the present application will be described below with reference to the accompanying drawings.

[0105] Figure 1 The flowchart of the seismic inversion method provided in the present application is shown in FIG. 1, and the flowchart of the seismic inversion method provided in the present application includes the following steps: Figure 1

[0106] Step S101: Obtain logging data and time domain seismic records, and perform well-seismic calibration processing according to the logging data to obtain time domain logging data.

[0107] ​Specifically, the well logging data contains the physical property information of the underground rock formation and can be obtained from the oil field; the well-to-seismic calibration is based on the principle of seismic wave propagation, and the corresponding relationship between the well logging depth and the seismic time is established by comparing the well logging curve and the seismic reflection characteristics, so as to convert the well logging data into time domain well logging data. By obtaining the time domain well logging data containing accurate well logging information, and eliminating the time-depth difference between the well logging data and the seismic data through well-to-seismic calibration, the two are aligned in the time domain, the characterization ability of the seismic data to the underground geological structure is enhanced, the basic data support for subsequent inversion is provided, and the accuracy of subsequent inversion is improved.

[0108] The time domain well logging data can be represented by a Robinson convolution model:

[0109]

[0110] wherein, is the time domain well logging data; is random noise; is a seismic wavelet; is a reflection coefficient; is an integral variable used to represent the time shift of the seismic wavelet;

[0111] The seismic wavelet is reflected by different reflection interfaces (corresponding to reflection coefficients ) in the underground, and then the random noise is added, and the time domain well logging data is formed in the time domain stacking, and the time domain well logging data can be forwardly obtained by using the model.

[0112] Step S102: data conversion processing is performed according to the time domain seismic record to obtain a frequency domain seismic record.

[0113] Specifically, the time domain seismic record is converted from the time dimension to the frequency dimension by Fourier transform, the seismic signal changing with time is decomposed into the superposition of different frequency components, and the frequency domain seismic record is obtained. The problem of insufficient resolution in the time domain is solved by extracting the frequency domain features, and the noise robustness is enhanced.

[0114] Step S103: horizon information is extracted from the time domain seismic record, and an elastic impedance initial model is generated according to the time domain well logging data and the horizon information.

[0115] Specifically, the reflection characteristics of the stratigraphic interface are identified from the time-domain seismic record to extract the horizon information; the elastic impedance initial model reflecting the distribution of the elastic properties of the underground rock formation is constructed based on the elastic impedance principle by combining the elastic parameters of the rock formation in the well logging data. The generated elastic impedance initial model is the basis for inversion, which can effectively reduce the blindness of inversion and improve the efficiency and accuracy of inversion. The process of generating the elastic impedance initial model can be realized by using the existing software. After inputting the time-domain well logging data and horizon information, the elastic impedance initial model is automatically generated, as shown in Fig. 8. Figure 2

[0116] Step S104: constructing a forward operator according to the elastic impedance initial model and the time-domain seismic record.

[0117] Specifically, based on the forward principle of seismic wave propagation, the elastic impedance initial model and the time-domain seismic record are considered to simulate the propagation process of seismic waves in the underground medium by constructing a forward operator. By establishing the mapping relationship from the underground medium model to the seismic record, it is convenient to adjust the model parameters for inversion by comparing the forward result with the actual record subsequently. The constructed forward operator can simulate the propagation of seismic waves and provide a principle calculation basis for inversion.

[0118] Step S105: synthesizing the pre-stack time-domain seismic record according to the seismic wavelet, the elastic impedance initial model and the forward operator.

[0119] Specifically, by combining the forward operator with the elastic impedance initial model, the pre-stack time-domain seismic record is simulated and generated in the time domain according to the propagation law of seismic waves, and the principle time-domain seismic record based on the initial model is obtained. As shown in Fig. 8, the time-domain seismic records at different formation angles (8° (a), 16° (b), 24° (c)) with a signal-to-noise ratio of 5 are shown in the figure. In this embodiment, the ri ker wavelet can be used as the seismic wavelet when simulating the input model. The actual data can be extracted as the seismic wavelet by extracting the statistical wavelet in the seismic data. Figure 3

[0120] Step S106: synthesizing the pre-stack frequency-domain seismic record according to the seismic wavelet, the elastic initial impedance initial model and the forward operator.

[0121] Specifically, by using the forward operator and the elastic initial impedance initial model in the frequency domain, the principle frequency-domain seismic record based on the initial model is obtained. By comparing with the actual frequency-domain record, the initial model is analyzed from the frequency dimension, and more dimensional data support is provided for inversion. As shown in Fig. 9, the frequency-domain seismic records at different formation angles (8° (a), 16° (b), 24° (c)) with a signal-to-noise ratio of 5 are shown in the figure. Figure 4 ​​As shown in the figure, the frequency domain seismic records at different stratigraphic angles (8°(a), 16°(b), 24°(c)) with a signal-to-noise ratio of 5 are displayed. In this embodiment, the Riker wavelet can be used as the seismic wavelet when creating the simulation input model. Actual data can be obtained by extracting statistical wavelets from the seismic data.

[0122] Step S107: Perform multi-objective gradient descent solution processing based on pre-stack time-domain seismic records, pre-stack frequency-domain seismic records, forward modeling operators, elastic impedance initial model, and pre-built initial objective function to obtain the Pareto solution set.

[0123] Specifically, based on the principles of multi-objective optimization and the gradient descent algorithm, and using pre-stack time-domain seismic records, pre-stack frequency-domain seismic records, forward modeling operators, an initial elastic impedance model, and a pre-established initial objective function as a foundation, the Pareto solution set is obtained by adjusting the model parameters to minimize the objective function value. The Pareto solution set contains multiple optimal solutions under different objectives, providing multiple options for subsequent decision-making and improving the reliability of the inversion results.

[0124] The process derives the gradients of the objective function in the time and frequency domains using the chain rule:

[0125]

[0126]

[0127] Gradient vectors of the two objective functions: and By defining a search direction, both objective functions can be reduced simultaneously. However, the two gradient directions may be inconsistent or even conflicting. Therefore, a common search direction is constructed by combining gradients:

[0128] Normalize the gradient:

[0129]

[0130] Constructing the gradient matrix:

[0131]

[0132] The goal is to find a weight vector. This makes the combination direction Minimize. Equilibrium is achieved by defining the following quadratic programming problem:

[0133]

[0134] The objective function can be to minimize the norm of the combined direction, i.e. to find a unit direction that is closest to the combined contribution of the two gradients in the least squares sense. The constraints ensure that the weights are positive and sum to one. This can be solved using a quadratic programming algorithm (e.g. interior point method). The optimal weights are obtained Post-combined direction:

[0135]

[0136] The model is then updated using gradient descent:

[0137]

[0138] where is the step size, and a decaying step size is used to accelerate convergence to the optimal solution:

[0139]

[0140] The elastic impedance model obtained from each iteration of the low-frequency elastic impedance initial model and its corresponding objective function value are analyzed, the time-domain objective function value the frequency-domain objective function value, and attempts to be added to the Pareto set. The Pareto set is maintained by non-dominated sorting and crowding distance, ensuring that the size of the solution set does not exceed the preset value (e.g. 50).

[0141] The Pareto front is constructed by the following dominance criteria:

[0142] Let there be two candidate solutions and :

[0143]

[0144] i.e. is not worse than in all objectives, and is better in at least one objective. Therefore, the solution dominates , and should be removed. The Pareto front only retains the set of solutions that are not dominated by any other solution. The diversity of solutions is maintained by crowding distance sorting.

[0145] Step S108: Sort and decision processing according to the Pareto solution set, obtain the target solution, and output the target solution as the pre-stack elastic impedance inversion result.

[0146] Specifically, by comprehensively analyzing and ranking each solution in the Pareto solution set, a target solution that is more consistent with the actual geological conditions of the underground is obtained, and an accurate prestack elastic impedance inversion result is output, thereby providing reliable geological data for exploration and development of deep oil and gas reservoirs.

[0147] The seismic inversion method provided in the embodiments of the present application fully fuses logging and seismic data by acquiring logging data and performing well-seismic calibration and data conversion; generates an elastic impedance initial model and a forward operator, and constructs an inversion basis based on physical principles; and solves and ranks decisions in multiple target gradient descent based on time-frequency domain data, thereby effectively avoiding single domain control problems, improving the precision, resolution and reliability of the inversion result, and more accurately describing deep oil and gas reservoir characteristics while ensuring the robustness of the final inversion result and realizing accurate reservoir characteristic description in a deep high-noise environment.

[0148] The process of performing well-seismic calibration processing according to the logging data in the above embodiments to obtain time domain logging data is described in detail, and the specific implementation manner of the process includes the following steps.

[0149] Step a1: performing data extraction processing according to the logging data to obtain P-wave velocity, S-wave velocity and density curves.

[0150] Specifically, the logging data contains various physical information about the underground rock stratum, and through existing data extraction technology, the P-wave velocity, S-wave velocity and density curves are separated from the original logging data according to the corresponding relationship between different logging methods and the physical properties of the rock stratum, which reflect the elastic properties of the rock stratum and provide a data basis for subsequent calculation and conversion, thereby improving the accuracy and reliability of the subsequent steps and further ensuring the accuracy of well-seismic calibration.

[0151] Step a2: converting the P-wave velocity, S-wave velocity and density curves according to the time-depth relationship conversion curve to obtain time domain logging data.

[0152] Specifically, based on the time-depth conversion principle of seismic wave propagation, there is a corresponding relationship between propagation time and propagation depth when seismic waves propagate in underground media. By extracting the P-wave velocity, S-wave velocity and density curves, and combining the existing mathematical model of seismic wave propagation time and depth, the logging data in the depth domain is converted to the time domain to obtain the time domain logging data. The differences between logging depth and seismic time are eliminated, well-seismic matching is realized, the high-resolution information of logging can be integrated into the seismic data system, and basic data is provided for subsequent time domain-based seismic inversion.

[0153] The present embodiment describes the calculation expression of the forward operator constructed according to the elastic impedance initial model and the time domain seismic record in the above embodiments, and the expression is as follows:

[0154]

[0155] In the formula, denotes a time-domain seismic record, denotes a forward operator, denotes an elastic impedance initial model, denotes random noise.

[0156] Specifically, the calculation expression of the forward operator is obtained after the Robinson convolution model is discretized.

[0157] The forward operator can be expressed as:

[0158]

[0159]

[0160] wherein, is a seismic wavelet; : denotes the elastic impedance value of the i-th sampling point (or corresponding to a position under the ground). According to the expression of the forward operator, the time-domain forward operator can be obtained, and the frequency-domain forward operator can be obtained by converting the time-domain forward operator from the time dimension to the frequency dimension. The elastic impedance initial model m is used to describe the distribution of the elastic properties of the underground rock layer, and the forward operator

[0161] is a mapping relationship constructed by mathematical calculation based on the principle of seismic wave propagation, which converts the elastic property information of the underground rock layer into the response after the propagation of the seismic wave, i.e., the time-domain seismic record . The random noise simulates the inevitable interference factors in the actual exploration process, including environmental noise, instrument error, etc. In the process of using software to process these data, random noise is added to achieve the purpose of simulating noise resistance. The time-domain seismic record simulates the propagation of the seismic wave from the source in the underground medium containing rock layers with different elastic properties, and finally forms the seismic record received by the seismic instrument. In some optional embodiments, the pre-built initial objective function in the above embodiments is:

[0162]

[0163] In the formula,

[0164] denotes a time-domain objective function, denotes a frequency-domain objective function, denotes an elastic impedance initial model, , , , Represents time-domain earthquake records. This represents frequency domain seismic records, which are obtained by performing a Fourier transform on time domain seismic records. Represents the forward modeling operator in the time domain. This represents the frequency domain forward operator. This indicates that the two data points in parentheses are correlated. This represents the balance factor between the time domain and the frequency domain. This represents the initial model constraints.

[0165] The time-domain seismic record is transformed from the time dimension to the frequency dimension using Fourier transform, decomposing the time-varying seismic signal into a superposition of different frequency components to obtain the frequency-domain seismic record. Extracting frequency-domain features compensates for the insufficient resolution of the time domain and enhances noise robustness.

[0166] Specifically, frequency domain seismic records can be represented as:

[0167]

[0168] in, Represents angular frequency. For seismic wavelets The spectrum; Time-domain reflection coefficient sequence Spectrum:

[0169]

[0170] Discretize the expression to obtain its discrete form:

[0171]

[0172] Expand the complex matrix in the formula into expressions for its real and imaginary parts:

[0173]

[0174] The existing objective function for hybrid time-domain and frequency-domain seismic inversion is expressed as follows:

[0175]

[0176] Represents time-domain earthquake records. Represents the forward calculus operator in the time domain. Represents frequency domain seismic records. This represents the frequency domain forward operator. Used to control the contribution of time-domain and frequency-domain inversion results to the objective function. Represents the regularization term, existing conventional initial model constraints. .

[0177] But the existing objective function has the problem of inconsistency in the magnitude of the time domain and the frequency domain, and the mismatch function constructed by using the two-norm in the time domain and the frequency domain has the problem of difference in magnitude, which leads to the fact that the inversion result is often controlled and affected by a single domain, and the advantages of noise resistance of time domain inversion and high resolution of frequency domain inversion result cannot be fully combined, so the pre-built initial objective function constructed in this embodiment is:

[0178]

[0179] In the formula, The time domain objective function is represented by, The frequency domain objective function is represented by, The elastic impedance initial model is represented by, , , The time domain seismic record is represented by, The frequency domain seismic record is represented by, The time domain forward operator is represented by, The frequency domain forward operator is represented by, The two data in the parentheses are cross-correlated, The balance factor between the time domain and the frequency domain is represented by, The initial model constraint is represented by.

[0180] Specifically, by combining time-frequency domain cooperation and model constraint, the multi-solution problem of inversion can be effectively reduced, and the elastic impedance model is closer to the real underground medium (e.g., the deep reservoir boundary and the thin interbedded structure are more clearly depicted). It can adapt to deep complex geological conditions (high noise, tectonic fragmentation, etc.). Compared with the scene in the prior art where the inversion result is fuzzy and greatly disturbed by noise, a more reliable elastic impedance distribution can be output to assist in accurately identifying oil and gas reservoirs. The initial model constraint can reduce invalid iterations, the balance factor can simplify the difficulty of multi-domain adaptation and parameter adjustment, the inversion process is more efficient, the result stability is improved, the dependence on artificial experience is reduced, and thus the demand for industrial batch processing can be better met.

[0181] The process of ranking and decision-making according to the Pareto solution set in the above embodiment to obtain the target solution is described in detail. The Pareto solution set includes multiple non-dominated solutions and the objective function value of each non-dominated solution on the corresponding objective function. The specific implementation mode of the process includes the following steps:

[0182] Step c1, constructing a solution set matrix according to all non-dominated solutions and all objective function values.

[0183] Specifically, a standardized data structure is formed by arranging each non-dominated solution in the Pareto solution set and its corresponding objective function value (such as time domain error and frequency domain error) in matrix form. By transforming the multi-objective optimization results into a quantifiable mathematical matrix, input is provided for subsequent entropy weighting and TOPSIS algorithms, achieving a structured representation of the data and ensuring the consistency and comparability of different objective function values.

[0184] By combining the TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) algorithm with entropy weighting on the Pareto set, the importance of each objective function is automatically determined, achieving a decision-making process without human bias. Assume there are N non-dominated solutions in the final Pareto solution set, and the corresponding two objective function values ​​(time domain error and frequency domain error) form a matrix:

[0185]

[0186] in, For the first The solution is at the th solution. The value of the objective function The objective function is a dual objective function optimization in both the time and frequency domains.

[0187] Step c2: Determine the probability matrix and information entropy based on the solution set matrix.

[0188] Specifically, the objective function value is transformed into a probability distribution through a probability matrix, eliminating the influence of dimensional differences on the weights. Furthermore, the data dispersion of each objective function is quantified through information entropy, laying the foundation for objectively determining the weights, avoiding biases caused by subjective weighting, and making the subsequent evaluation of non-dominated solutions more consistent with the actual data distribution, thereby improving the scientific nature of decision-making.

[0189] Calculate the probability matrix With information entropy :

[0190]

[0191] in, It is a probability matrix. For information entropy, This represents the objective function value after range normalization.

[0192] Step c3: Based on the probability matrix and information entropy, determine the degree of difference and weight reflected by the entropy.

[0193] Specifically, the difference degree of each target is calculated according to the information entropy, and the difference degree = 1-information entropy. The greater the difference degree, the stronger the distinguishing ability of the target function index to the solution. Then, the weight of each target is obtained through normalization processing. By converting the information entropy into the difference degree which can be directly used for weighting, the subsequent analysis of non-dominated solutions can reasonably reflect the actual importance of each target, avoid the decision bias caused by improper subjective weight setting, and improve the accuracy of ranking.

[0194] The difference degree (i.e. target importance) reflected by entropy and weight:

[0195]

[0196] Wherein, is the difference degree, is the weight.

[0197] Step c4, according to the solution set matrix, the difference degree reflected by entropy and the weight, determining the first weighted distance and the second weighted distance of each non-dominated solution.

[0198] Specifically, the first weighted distance and the second weighted distance respectively correspond to the weighted distance of the non-dominated solution to the ideal point and the negative ideal point. The first weighted distance (distance to the ideal point) is the difference between the value of each target function of the non-dominated solution and the optimal value of the target function (ideal point, such as the minimum time domain error and the minimum frequency domain error), which is multiplied by the corresponding target weight and then accumulated. The second weighted distance (distance to the negative ideal point) is the weighted accumulation of the difference between the worst value of the target function (negative ideal point). By analyzing the target function weight and the target performance of the non-dominated solution through the weighted distance, the multi-objective optimization result is converted into a comparable distance index, which simplifies the distance size comparison of the complex multi-objective comparison and simplifies the decision-making basis.

[0199] Step c5, according to the first weighted distance and the second weighted distance of each non-dominated solution, the ranking score of the non-dominated solution is calculated by the optimal solution distance method.

[0200] Specifically, by using the TOPSIS algorithm, the first weighted distance (to the ideal solution) and the second weighted distance (to the negative ideal solution) of each solution are combined to calculate the comprehensive score, which combines the distance relationship of the positive and negative ideal solutions to realize the global ranking of the solution. The higher the score, the closer the solution is to the ideal state, which provides an objective basis for priority.

[0201] The TOPSIS score is defined as:

[0202]

[0203] Wherein The higher the score, the closer the solution is to the optimal solution. Finally, the solution with the highest score is selected as the optimal elastic impedance inversion result.

[0204] Step c6, according to the ranking score of all non-dominated solutions, determine a non-dominated solution from all non-dominated solutions as the target solution.

[0205] Specifically, the non-dominated solution with a higher ranking score in the multi-objective optimization Pareto solution set is closer to the ideal state (balance of each objective function optimal), and the final target solution is determined according to the size of the score. The target solution achieves a relatively optimal balance in the time domain, frequency domain and other multi-objectives, and provides accurate model results for seismic inversion and other applications.

[0206]

[0207] The optimal solution is The entropy weight-TOPSIS method not only considers the distribution information and difference between the objective functions, but also effectively realizes the identification of the target solution in the multi-objective optimization solution set through normalization, weighted distance measurement and similarity calculation.

[0208] The embodiment of the application realizes data-driven of the multi-objective solution set by constructing the solution set matrix standardized data, entropy weight method objective weighting and TOPSIS quantitative ranking, avoiding the limitations of relying on artificial experience in the prior art. And by measuring the importance of the difference and the weighted distance of the target, the performance of the solution is analyzed comprehensively, and the target solution balancing the contradiction between time domain noise resistance and frequency domain high resolution is screened out, effectively reducing the subjective bias and improving the scientificity and practicality of the inversion result of the deep complex reservoir.

[0209] The embodiment details the process of determining the first weighted distance and the second weighted distance of each non-dominated solution according to the solution set matrix, the difference reflected by entropy and the weight in the above embodiment. The specific implementation mode of the process includes the following steps:

[0210] Step d1, according to the multiple objective function values in the solution set matrix, determine the ideal solution and the negative ideal solution of each objective function.

[0211] Specifically, the ideal solution is the optimal value of each objective function (such as the minimum value in the minimization objective, the maximum value in the maximization objective), and the negative ideal solution is the worst value of each objective function (such as the maximum value in the minimization objective, the minimum value in the maximization objective). By traversing the objective function values in the solution set matrix, the optimal and worst benchmark points of each objective function are determined as the reference for distance calculation.

[0212] Step d2, according to the solution set matrix, the ideal solution of each objective function, determine the first weighted distance of each non-dominated solution, and according to the solution set matrix, the negative ideal solution of each objective function, the difference reflected by entropy and the weight, determine the second weighted distance of each non-dominated solution.

[0213] Specifically, the first weighted distance measures the degree to which the non-dominated solution approaches the optimal state, and the smaller the distance, the better the solution; the second weighted distance measures the degree to which the non-dominated solution is away from the worst state, and the greater the distance, the better the solution. By quantifying the pros and cons of the solution in two dimensions, and adjusting the weight through entropy difference, the bias caused by different degrees of dispersion of the objective function is avoided.

[0214] The embodiment of the present application defines the ideal and negative ideal state by extreme value to determine the optimization direction, and uses the entropy weight method to objectively assign weights and combine the Euclidean distance to quantify the performance difference, so as to convert the non-dominated solution in multi-objective optimization into a quantifiable numerical index for comparison, realize data-driven sorting of the multi-objective solution set, and make the sorting result more comprehensively reflect the situation of the non-dominated solution in the multi-objective space, thereby providing a more scientific and objective target solution screening method for pre-stack elastic impedance inversion.

[0215] In the above embodiment, the calculation formula for determining the first weighted distance of each non-dominated solution according to the solution set matrix, the ideal solution of each objective function, and the second weighted distance of each non-dominated solution according to the solution set matrix, the negative ideal solution of each objective function, the difference reflected by entropy and the weight is as follows:

[0216]

[0217] In the formula, denotes the first weighted distance, denotes the second weighted distance, =2, denotes the target function value after range normalization processing, denotes the weight reflected by entropy, denotes the ideal solution, denotes the negative ideal solution.

[0218] Specifically, the difference in time-frequency domain multi-objective is converted into an intuitive distance index for quantitative comparison. The entropy weight dynamically adjusts the weight to better adapt to deep high noise and strong multi-solution characteristics. In deep gravel reservoir inversion, the frequency domain weight sensitive to details can be strengthened, or the time domain noise resistance can be focused on, so that the distance calculation is more consistent with the actual geological constraints, providing a basis for subsequent sorting and avoiding the influence of subjective consciousness caused by manual weighting.

[0219] The embodiment of the present application details the process of determining a non-dominated solution as a target solution from all non-dominated solutions according to the sorting scores of all non-dominated solutions in the above embodiment. The specific implementation manner of the process includes the following steps:

[0220] Step e1, sorting processing is performed on all non-dominated solutions according to the sorting scores of all non-dominated solutions, to obtain a non-dominated solution sequence.

[0221] Specifically, based on the ranking score of all non-dominated solutions (the comprehensive score calculated by the superior-inferior solution distance method), the solution set is arranged in descending order from high to low, forming a non-dominated solution sequence. If there are solutions with the same score, further distinguish by the difference degree (i.e. weight) of the entropy weight method or the local comparison of the objective function value. Converting the multi-objective solution set into an ordered sequence of solutions clarifies the priority and provides a direct basis for subsequent decision-making.

[0222] Step e2, from the non-dominated solution sequence, a non-dominated solution with the maximum ranking score is matched as the target solution according to the ranking score.

[0223] Specifically, the solution with the maximum ranking score is selected from the ranked non-dominated solution sequence as the target solution. If multiple solutions have the same score and are in the first place, the solution with smaller difference degree or better noise resistance is selected. In the Pareto front, the solution with the best comprehensive performance is quickly selected to avoid the deviation caused by artificial intervention.

[0224] As shown in Figure 5 , Figure 6 and Figure 7 , in order to compare with the prior art, the same initial model and the same hyperparameters are used for illustration:

[0225] Figure 6 The inversion results of the pre-stack elastic impedance inversion method of the embodiment are shown in the figure, which are 8° (a), 16° (b), and 24° (c) under the condition of signal-to-noise ratio of 5. The inversion method has both noise resistance and high resolution, and can completely depict the shape of the underground geological structure even under the condition of high noise.

[0226] Figure 7 The mixed time-frequency domain inversion method of the prior art selects the weight distribution , , to control the degree of participation of the frequency domain seismic record in the inversion. Due to the influence of noise, the accuracy of the conventional frequency domain inversion result is limited by the quality of the seismic record, and the relationship between the noise resistance and the resolution of the inversion result cannot be reasonably balanced.

[0227] In order to quantitatively compare the accuracy of the single time domain elastic impedance inversion result, the single frequency domain elastic impedance inversion result, and the traditional mixed time-frequency domain elastic impedance inversion result of the present invention, the 8° elastic impedance inversion result of CDP=150 is selected for comparison and analysis by using the cross-correlation coefficient.

[0228] The correlation coefficient of vector and vector is defined as:

[0229]

[0230] wherein, is a vector is a vector is a covariance of vectors and are variances of vectors and respectively.

[0231] From the cross-correlation coefficient results, the correlation coefficient of the application is as high as 0.965, which is significantly better than other control methods, indicating that the inversion results are more consistent with the true elastic impedance model, which can ensure the robustness of the inversion results while realizing accurate reservoir characterization in deep high-noise environment.

[0232] The embodiment of the application sorts the non-dominated solutions based on the sorting score and selects the solution with the maximum score as the target solution, taking the quantitative index as the only basis, avoiding subjective bias caused by artificial experience, and ensuring the objectivity and scientificity of the target solution selection. The target solution selected from the Pareto solution set has noise resistance and high resolution characteristics, and provides a scientific and reliable elastic impedance inversion result for complex reservoir characterization and other geological scenes.

[0233] Figure 8 The structure of the seismic inversion device provided in the application is shown in the figure. Figure 8 As shown in the figure, the seismic inversion device 80 comprises:

[0234] The acquisition module 801 is configured to acquire logging data and time-domain seismic records, and perform well-seismic calibration processing according to the logging data to obtain time-domain logging data.

[0235] The conversion module 802 is configured to perform data conversion processing according to the time-domain seismic records to obtain frequency-domain seismic records.

[0236] The model establishment module 803 is configured to extract horizon information from the time-domain seismic records, and generate an elastic impedance initial model according to the time-domain logging data and the horizon information.

[0237] The construction module 804 is configured to construct a forward operator according to the elastic impedance initial model and the time-domain seismic records.

[0238] The generation module 805 is configured to generate pre-stack time-domain seismic records according to a seismic wavelet, the elastic impedance initial model and the forward operator.

[0239] The generation module 805 is further configured to generate pre-stack frequency-domain seismic records according to a seismic wavelet, the elastic initial impedance initial model and the forward operator.

[0240] The analysis module 806 is configured to perform multi-objective gradient descent solving processing according to the pre-stack time domain seismic record, the pre-stack frequency domain seismic record, the forward operator, the initial elastic impedance model and the pre-built initial objective function, to obtain a Pareto solution set;

[0241] The processing module 807 is configured to perform sorting decision processing according to the Pareto solution set, to obtain a target solution, and output the target solution as a pre-stack elastic impedance inversion result.

[0242] In a possible implementation, the conversion module 802 is specifically configured to:

[0243] perform data extraction processing according to the logging data, to obtain a P-wave velocity curve, a S-wave velocity curve and a density curve;

[0244] perform conversion processing on the P-wave velocity curve, the S-wave velocity curve and the density curve according to a time-depth relationship conversion curve, to obtain time domain logging data.

[0245] In a possible implementation, the calculation expression of the forward operator constructed by the construction module 804 is as follows:

[0246]

[0247] In the formula, represents a time domain seismic record, represents a forward operator, represents an initial elastic impedance model, represents a preset random noise.

[0248] In a possible implementation, the pre-built initial objective function in the analysis module is as follows:

[0249]

[0250] In the formula, represents a time domain objective function, represents a frequency domain objective function, represents an initial elastic impedance model, , , represents a time domain seismic record, represents a frequency domain seismic record, which is obtained by performing Fourier transform on the time domain seismic record, represents a time domain forward operator, represents a frequency domain forward operator, represents the cross-correlation of two data in the brackets, represents a balance factor between the time domain and the frequency domain, represents an initial model constraint.

[0251] ​​In a possible implementation, the processing module 807 is specifically configured to:

[0252] construct a solution set matrix according to all non-dominated solutions and all objective function values;

[0253] determine a probability matrix and an information entropy according to the solution set matrix;

[0254] determine a difference degree reflected by the entropy and a weight according to the probability matrix and the information entropy;

[0255] determine a first weighted distance and a second weighted distance of each non-dominated solution according to the solution set matrix, the difference degree reflected by the entropy, and the weight;

[0256] calculate a ranking score of the non-dominated solution by using the superior-inferior solution distance method according to the first weighted distance and the second weighted distance of each non-dominated solution;

[0257] determine a non-dominated solution as a target solution from all non-dominated solutions according to the ranking scores of all non-dominated solutions.

[0258] In a possible implementation, the processing module 807 is further configured to:

[0259] determine an ideal solution and a negative ideal solution of each objective function according to the multiple objective function values in the solution set matrix;

[0260] determine a first weighted distance of each non-dominated solution according to the solution set matrix and the ideal solution of each objective function, and determine a second weighted distance of each non-dominated solution according to the solution set matrix, the negative ideal solution of each objective function, the difference degree reflected by the entropy, and the weight.

[0261] In a possible implementation, the calculation formula of the processing module 807 for determining the second weighted distance of each non-dominated solution is as follows:

[0262]

[0263] wherein, the first weighted distance is denoted by d1, the second weighted distance is denoted by d2, = 2, the objective function value after the range normalization processing is denoted by x′i, the weight of the difference degree reflected by the entropy is denoted by w, the ideal solution is denoted by z*, the negative ideal solution is denoted by z*′.

[0264] In a possible implementation, the processing module 807 is further configured to:

[0265] perform a ranking processing on all non-dominated solutions according to the ranking scores of all non-dominated solutions, to obtain a non-dominated solution sequence.

[0266] A non-dominated solution with the maximum ranking score is obtained from the non-dominated solution sequence according to the ranking score, and is taken as the target solution.

[0267] The seismic inversion device provided in the embodiment can be used to execute the seismic inversion method described above, and has similar implementation principles and technical effects, which will not be described here again in the embodiment.

[0268] Figure 9 The hardware structure schematic diagram of the electronic device provided in the application is shown in FIG. 9, and the electronic device 90 includes at least one processor 901 and a memory 902. Figure 9 Optionally, the seismic inversion device 90 further includes a communication component 903. The processor 901, the memory 902 and the communication component 903 are connected through a bus 904.

[0269] In the specific implementation process, the at least one processor 901 executes the computer execution instructions stored in the memory 902, so that the at least one processor 901 executes the method as described above.

[0270] The specific implementation process of the processor 901 can refer to the method embodiments described above, and has similar implementation principles and technical effects, which will not be described here again in the embodiment.

[0271] In the above embodiments, it should be understood that the processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC) and the like. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor and the like. The steps of the method disclosed in the application can be directly embodied as hardware processor execution, or executed by a combination of hardware and software modules in the processor.

[0272] The memory can include a random access memory (RAM), and can also include a non-volatile memory (NVM), for example, at least one disk memory.

[0273] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, the bus in the drawings of the present application does not limit to only one bus or one type of bus.

[0274] The present application also provides a computer program product, comprising a computer program, which, when executed by a processor, implements the method described above.

[0275] The present application also provides a computer readable storage medium, which stores computer execution instructions, and when a processor executes the computer execution instructions, the method described above is implemented.

[0276] The readable storage medium described above can be realized by any type of volatile or non-volatile storage device or their combination, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general or special purpose computer.

[0277] An exemplary readable storage medium is coupled to the processor, so that the processor can read information from the readable storage medium, and can write information to the readable storage medium. Of course, the readable storage medium can also be an integral part of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist as discrete components in the device.

[0278] The division of units is only a logical functional division, and in actual implementation, there can be another division manner, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0279] The units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, i.e., may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0280] In addition, each functional unit in various embodiments of the application can be integrated into one processing unit, or each unit can exist physically, or two or more units can be integrated into one unit.

[0281] If the function is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the various embodiment methods of the application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0282] Those skilled in the art can understand that all or part of the steps of the above-mentioned method embodiments can be completed by program instruction related hardware. The aforementioned program can be stored in a computer readable storage medium. The program executes to perform the steps of the above-mentioned method embodiments; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk, and various media that can store program codes.

[0283] Finally, it should be noted that those skilled in the art, after considering the specification and practicing the application disclosed herein, will easily think of other embodiments of the application. The application is intended to cover any variations, uses, or adaptations of the application that follow the general principles of the application and include common knowledge or conventional techniques in the art that are not disclosed by the application, and is not limited to the precise structure described above and shown in the drawings, and various modifications and changes can be made without departing from the scope thereof. The scope of the application is only limited by the appended claims.

Claims

1. A seismic inversion method, characterized in that, include: Acquire well logging data and time-domain seismic records, and perform well-seismic calibration processing based on the well logging data to obtain time-domain well logging data; The time-domain seismic records are converted and processed to obtain frequency-domain seismic records. Stratigraphic information is extracted from the time-domain seismic records, and an initial elastic impedance model is generated based on the time-domain well logging data and the stratigraphic information. Based on the initial model of elastic impedance and time-domain seismic records, a forward modeling operator is constructed. Based on the seismic wavelet, the elastic impedance initial model, and the forward modeling operator, pre-stack time-domain seismic records are synthesized. Based on the seismic wavelet, the elastic initial impedance initial model, and the forward modeling operator, a pre-stack frequency domain seismic record is synthesized. Based on the pre-stack time-domain seismic records, the pre-stack frequency-domain seismic records, the forward modeling operator, the elastic impedance initial model, and the pre-established initial objective function, a multi-objective gradient descent solution is performed to obtain the Pareto solution set. The Pareto solution set is sorted and the target solution is obtained by performing sorting and decision processing. The target solution is then output as the pre-stack elastic impedance inversion result. The pre-built initial objective function is: In the formula, Represents the objective function in the time domain. Represents the objective function in the frequency domain. This represents the initial model of elastic impedance. , , Represents time-domain earthquake records. This represents frequency domain seismic records, which are obtained by performing a Fourier transform on time domain seismic records. Represents the forward calculus operator in the time domain. This represents the frequency domain forward operator. This indicates that the two data points in parentheses are correlated. This represents the balance factor between the time domain and the frequency domain. This represents the initial model constraints.

2. The method according to claim 1, characterized in that, The step of performing well-seismic calibration processing based on the well logging data to obtain time-domain well logging data includes: Data extraction and processing are performed on the well logging data to obtain P-wave velocity, S-wave velocity, and density curves. Based on the P-wave velocity, S-wave velocity, and density curves, the time-depth relationship conversion curve is used to convert and process the data to obtain time-domain logging data.

3. The method according to claim 1, characterized in that, The calculation expression for the forward modeling operator, constructed based on the initial elastic impedance model and time-domain seismic records, is as follows: In the formula, Represents time-domain earthquake records. Indicates the forward operand operator. This represents the initial model of elastic impedance. This represents random noise.

4. The method according to claim 1, characterized in that, The Pareto solution set includes multiple non-dominated solutions and the objective function value of each non-dominated solution on the corresponding objective function; Accordingly, the process of sorting and deciding based on the Pareto solution set to obtain the target solution includes: Construct a solution set matrix based on all non-dominated solutions and all objective function values; Based on the solution set matrix, determine the probability matrix and information entropy; Based on the probability matrix and information entropy, determine the degree of difference and weight reflected by the entropy; Based on the solution set matrix, the degree of difference reflected by the entropy, and the weights, determine the first weighted distance and the second weighted distance for each non-dominated solution; The ranking score of the non-dominated solution is calculated using the superior-inferior solution distance method based on the first weighted distance and the second weighted distance of each non-dominated solution. Based on the ranking scores of all non-dominated solutions, select one non-dominated solution as the target solution.

5. The method according to claim 4, characterized in that, The step of determining the first weighted distance and the second weighted distance for each non-dominated solution based on the solution set matrix, the difference reflected by the entropy, and the weights includes: Based on the multiple objective function values ​​in the solution set matrix, determine the ideal solution and negative ideal solution for each objective function; Based on the solution set matrix and the ideal solution of each objective function, determine the first weighted distance of each non-dominated solution, and based on the solution set matrix, the negative ideal solution of each objective function, the degree of difference reflected by entropy, and the weight, determine the second weighted distance of each non-dominated solution.

6. The method according to claim 4, characterized in that, The formula for determining the first weighted distance of each non-dominated solution based on the solution set matrix and the ideal solution of each objective function, and the formula for determining the second weighted distance of each non-dominated solution based on the solution set matrix, the negative ideal solution of each objective function, the difference reflected by entropy, and the weight, is as follows: In the formula, Indicates the first weighted distance. Indicates the second weighted distance. =2, This represents the objective function value after range normalization. The weights that represent entropy reflect. Represents the ideal solution. This represents the negative ideal solution.

7. The method according to claim 4, characterized in that, Based on the ranking scores of all non-dominated solutions, one non-dominated solution is selected as the target solution from all non-dominated solutions, including: Sort all non-dominated solutions according to their sorting scores to obtain a sequence of non-dominated solutions. The non-dominated solution with the largest sorted score is obtained from the non-dominated solution sequence and taken as the target solution.

8. An electronic device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-7.

Images