A target function determination method and device for full waveform inversion

Abstract

The application provides a target function determination method and device for full waveform inversion, which comprises the following steps: performing nonlinear transformation on obtained seismic observation data and seismic synthetic data; constructing a distance matrix according to the seismic synthetic data, the seismic observation data and the seismic trace data after nonlinear transformation; determining an optimal permutation vector from the distance matrix; determining a single-channel data deviation corresponding to each seismic trace by combining the optimal permutation vector with a pre-set deviation calculation expression; and summing all the single-channel data deviations to obtain a target function for full waveform inversion. The optimal transmission under nonlinear transformation is used to obtain a relatively convex target function, thereby solving the period jump problem in full waveform inversion.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of land or marine natural source seismic and artificial source seismic data processing, and particularly relates to a target function determination method and device for full waveform inversion. BACKGROUND

[0002] Full waveform inversion is the most powerful imaging technique so far. Full waveform inversion relies on solving wave equation to simulate wave field propagation in the subsurface, then comparing the difference between synthetic seismogram and observed data, transmitting data error to form gradient, updating model to reduce data error, using new model as initial model and repeating iteration until the algorithm converges to the optimal value. Generally, full waveform inversion uses least square waveform error, which not only fits the amplitude of seismic signal, but also fits the phase of seismic signal; therefore, theoretically, full waveform inversion can distinguish size up to half wavelength.

[0003] But full waveform inversion needs an accurate initial model when solving subsurface medium parameters. Because the target function is based on waveform comparison, when the difference between initial value and observed data event is large, periodic jump phenomenon is easy to occur, leading to inaccurate inversion result. In actual production, it is not easy to provide an accurate initial model. SUMMARY

[0004] The present application provides a target function determination method and device for full waveform inversion, which is used to obtain a relatively convex target function by using optimized transmission under nonlinear transformation.

[0005] In the first aspect, the present application provides a target function determination method for full waveform inversion, comprising:

[0006] Performing nonlinear transformation on obtained seismic observation data and seismic synthetic data;

[0007] Constructing a distance matrix according to the seismic synthetic data, the seismic observation data, and the seismic trace data after nonlinear transformation;

[0008] Determining a best permutation vector from the distance matrix;

[0009] Determining a single-channel data deviation corresponding to each seismic trace by combining the best permutation vector through a pre-set deviation calculation expression;

[0010] Summing all the single-channel data deviations to obtain a target function for full waveform inversion.

[0011] Optionally, constructing a distance matrix according to the seismic synthetic data, the seismic observation data, and the seismic trace time after nonlinear transformation, comprises:

[0012] normalizing time difference between synthetic single trace data of all the seismic synthetic data and observed single trace data corresponding to the seismic observed data, to obtain time difference normalized values;

[0013] determining amplitude difference of the seismic trace data after the nonlinear transformation, and normalizing the amplitude difference to obtain amplitude difference normalized values;

[0014] obtaining matrix elements of the distance matrix based on the time difference normalized values and the amplitude difference normalized values, and constructing the distance matrix based on the matrix elements.

[0015] Optionally, determining an optimal permutation vector from the distance matrix, comprising:

[0016] sorting all the matrix elements of the distance matrix by a bid ranking algorithm to obtain a sorting result;

[0017] determining the optimal permutation vector according to the sorting result.

[0018] Optionally, the bias calculation expression is specifically:

[0019] h=∑ ij c ij m ij ;

[0020] wherein, h is a bias, c ij is a distance matrix, and m ij is a permutation matrix.

[0021] In a second aspect, the present application provides a device for determining a target function of full waveform inversion, comprising:

[0022] a transformation module configured to perform nonlinear transformation on obtained seismic observed data and seismic synthetic data;

[0023] a matrix construction module configured to construct a distance matrix according to the seismic synthetic data, the seismic observed data, and seismic trace data after nonlinear transformation;

[0024] an optimal permutation vector determination module configured to determine an optimal permutation vector from the distance matrix;

[0025] a single trace data bias determination module configured to determine single trace data bias corresponding to each seismic trace by a pre-set bias calculation expression combined with the optimal permutation vector;

[0026] a target function determination module configured to sum all the single trace data bias to obtain a target function of full waveform inversion.

[0027] Optionally, the matrix construction module comprises:

[0028] a time difference normalization value determination sub-module configured to normalize a time difference between a synthetic single-channel data of all the seismic synthetic data and an observed single-channel data corresponding to the seismic observed data, to obtain a time difference normalization value;

[0029] an amplitude difference normalization value determination sub-module configured to determine an amplitude difference of the seismic trace data after the nonlinear transformation and normalize the amplitude difference to obtain an amplitude difference normalization value;

[0030] a construction sub-module configured to obtain a matrix element of the distance matrix based on the time difference normalization value and the amplitude difference normalization value, and construct the distance matrix based on the matrix element.

[0031] Optionally, the optimal permutation vector determination module comprises:

[0032] a sorting result determination sub-module configured to sort all the matrix elements of the distance matrix by a bid ranking algorithm to obtain a sorting result;

[0033] an optimal permutation vector determination sub-module configured to determine the optimal permutation vector according to the sorting result. Optionally, the bias calculation expression is specifically:

[0034] h = ∑ ij c ij m ij ;

[0035] wherein h is a bias, c ij is a distance matrix, and m ij is a permutation matrix.

[0036] In a third aspect, the present application provides an electronic device comprising a processor and a memory, wherein the memory stores computer readable instructions, and when the computer readable instructions are executed by the processor, the steps in the method provided in the first aspect are executed.

[0037] In a fourth aspect, the present application provides a storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps in the method provided in the first aspect are executed.

[0038] From the above technical solutions, it can be seen that the present application has the following advantages:

[0039] The application provides a target function determination method and device for full waveform inversion, and the method comprises the following steps: performing nonlinear transformation on obtained seismic observation data and seismic synthetic data; constructing a distance matrix according to the seismic synthetic data, the seismic observation data and the seismic trace data after nonlinear transformation; determining an optimal permutation vector from the distance matrix; determining a single-channel data deviation corresponding to each seismic trace by combining the optimal permutation vector through a pre-set deviation calculation expression; and summing all the single-channel data deviations to obtain a target function for full waveform inversion. The optimal transmission under nonlinear transformation is used to obtain a relatively convex target function, thereby solving the period jump problem in full waveform inversion. BRIEF DESCRIPTION OF DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0041] Figure 1 A flow chart of a first embodiment of the target function determination method for full waveform inversion of the present application;

[0042] Figure 2 A flow chart of a second embodiment of the target function determination method for full waveform inversion of the present application;

[0043] Figure 3 A flow chart of a third embodiment of the target function determination method for full waveform inversion of the present application;

[0044] Figure 4 A flow chart of a fourth embodiment of the target function determination method for full waveform inversion of the present application;

[0045] Figure 5 A schematic diagram of seismic observation data;

[0046] Figure 6 A schematic diagram of seismic synthetic data;

[0047] Figure 7 A schematic diagram of a real model;

[0048] Figure 8 A schematic diagram of an initial model;

[0049] Figure 9 A comparison diagram of initial data;

[0050] Figure 10 A schematic diagram of inversion results;

[0051] Figure 11 a final synthesis data schematic diagram;

[0052] Figure 12 a final data comparison schematic diagram;

[0053] Figure 13 a structural block diagram of an embodiment of a target function determination device for full waveform inversion of the application. DETAILED DESCRIPTION

[0054] The embodiment of the application provides a target function determination method and device for full waveform inversion, which is used for obtaining a relatively convex target function by using optimized transmission under nonlinear transformation.

[0055] In order to make the application purposes, features and advantages of the application more obvious and easy to understand, the technical solutions in the embodiments of the application will be clearly and completely described below in combination with the drawings in the embodiments of the application. Obviously, the following described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the application.

[0056] Embodiment one, please refer to Figure 1 , Figure 1 a flow step diagram of the embodiment one of the target function determination method for full waveform inversion of the application, comprising:

[0057] Step S101, performing nonlinear transformation on obtained seismic observation data and seismic synthesis data;

[0058] Step S102, constructing a distance matrix according to the seismic synthesis data, the seismic observation data and the seismic trace data after nonlinear transformation;

[0059] Step S103, determining an optimal permutation vector from the distance matrix;

[0060] Step S104, determining a single-channel data deviation corresponding to each seismic trace by combining the optimal permutation vector through a pre-set deviation calculation expression;

[0061] Step S105, summing all the single-channel data deviations to obtain a target function for full waveform inversion.

[0062] The application provides a target function determination method for full waveform inversion, comprising the following steps: performing nonlinear transformation on obtained seismic observation data and seismic synthetic data; constructing a distance matrix according to the seismic synthetic data, the seismic observation data and the seismic trace data after nonlinear transformation; determining an optimal permutation vector from the distance matrix; determining a single-channel data deviation corresponding to each seismic trace by combining the optimal permutation vector with a preset deviation calculation expression; and summing all the single-channel data deviations to obtain a target function for full waveform inversion. The optimal transmission under nonlinear transformation is used to obtain a relatively convex target function, thereby solving the period jump problem in full waveform inversion.

[0063] Embodiment two, please refer to Figure 2 , Figure 2 is a flow chart of a target function determination method for waveform inversion according to the application, comprising the following steps:

[0064] Step S201, performing nonlinear transformation on obtained seismic observation data and seismic synthetic data;

[0065] Step S202, normalizing a time difference between synthetic single-channel data of all the seismic synthetic data and observed single-channel data corresponding to the seismic observation data to obtain a time difference normalized value;

[0066] Step S203, determining an amplitude difference of the seismic trace data after nonlinear transformation, and performing normalization processing on the amplitude difference to obtain an amplitude difference normalized value;

[0067] Step S204, obtaining matrix elements of the distance matrix based on the time difference normalized value and the amplitude difference normalized value, and constructing the distance matrix based on the matrix elements;

[0068] Step S205, determining an optimal permutation vector from the distance matrix;

[0069] Step S206, determining a single-channel data deviation corresponding to each seismic trace by combining the optimal permutation vector with a preset deviation calculation expression;

[0070] Step S207, summing all the single-channel data deviations to obtain a target function for full waveform inversion.

[0071] Embodiment three, please refer to Figure 3 , Figure 3 is a flow chart of a target function determination method for full waveform inversion according to the application, comprising the following steps:

[0072] Step S301, performing nonlinear transformation on obtained seismic observation data and seismic synthetic data;

[0073] Step S302, constructing a distance matrix according to the seismic synthetic data, the seismic observation data, and the seismic trace data after nonlinear transformation;

[0074] Step S303, sorting all the matrix elements of the distance matrix through a bid ranking algorithm to obtain a sorting result;

[0075] Step S304, determining the optimal permutation vector according to the sorting result;

[0076] Step S305, determining the single-channel data deviation corresponding to each seismic trace through a pre-set deviation calculation expression combined with the optimal permutation vector;

[0077] Step S306, summing all the single-channel data deviations to obtain a target function of full waveform inversion.

[0078] Embodiment four, please refer to Figure 4 , Figure 4 is a flow chart of a method for determining a target function of full waveform inversion according to the present application, comprising:

[0079] Step S401, performing nonlinear transformation on the obtained seismic observation data and seismic synthetic data;

[0080] It should be noted that full waveform inversion is to compare the relationship between seismic synthetic data and seismic observation data, and in actual observation records, the seismic observation data and the seismic synthetic data are discrete.

[0081] In the embodiment of the present application, it is assumed that the observation single-channel data of the seismic observation data at t j is d j , and the observation single-channel data of the seismic synthetic data at T i is u i . Nonlinear transformation can effectively balance the direct amplitude difference of different seismic events, for example, for land data, application of TANH can effectively enhance the strength of body waves, which is important for improving the contribution of surface waves to obtain deep structure.

[0082] Step S402, normalizing the time difference between the synthetic single-channel data of all the seismic synthetic data and the observation single-channel data corresponding to the seismic observation data to obtain a time difference normalized value;

[0083] Step S403, determining the amplitude difference of the seismic trace data after nonlinear transformation, and performing normalization processing on the amplitude difference to obtain an amplitude difference normalized value;

[0084] Step S404, obtaining a matrix element of the distance matrix based on the time difference normalized value and the amplitude difference normalized value, and constructing the distance matrix based on the matrix element;

[0085] In the embodiment of the application, the seismic trace after the nonlinear transformation is denoted as

[0086] wherein α and β are two constants, which prevent the situation of exceeding the expression range of the computer from occurring in data calculation, and then a distance matrix c can be constructed as follows ij As shown below:

[0087]

[0088] wherein T is the total observation actual, A is the maximum value of the observation data, and η is a user input parameter.

[0089] Step S405, sorting all the matrix elements of the distance matrix through a bid ranking algorithm to obtain a sorting result.

[0090] Step S406, determining the best permutation vector according to the sorting result.

[0091] Step S407, determining the single trace data deviation corresponding to each seismic trace through a pre-set deviation calculation expression combined with the best permutation vector.

[0092] Specifically, the deviation calculation expression is specifically as follows:

[0093] h = ∑ ij c ij m ij ;

[0094] wherein h is the deviation, c ij is the distance matrix, m ij is the permutation matrix, which is obtained by row permutation and column permutation of the unit matrix, and the permutation matrix is equivalent to the permutation vector; or in other words, the permutation matrix can be represented by a vector: j = σ i .

[0095] In the embodiment of the application, after the distance matrix is constructed, the best permutation vector is quickly obtained through the bid ranking algorithm, and then the data deviation corresponding to each seismic trace is determined through the pre-set deviation calculation expression combined with the best permutation vector.

[0096] Step S408, summing all the single trace data deviations to obtain the objective function of the full waveform inversion.

[0097] In the embodiment of the present application, all the single-channel data deviations are added together to obtain the objective function required by the whole waveform inversion.

[0098] Please refer to Figure 5 and Figure 6 , Figure 5 the seismic observation data generated in the real model of Figure 7 , i.e., the seismic shot gather, Figure 6 the seismic synthetic data generated in the initial model of Figure 8 , and the comparison between Figure 5 and Figure 6 obtains the initial data comparison diagram shown in Figure 9 , wherein the letter O represents the observation data, and the letter S represents the synthetic data. As can be seen, the data difference is large, and the period jump phenomenon is serious. The inversion result obtained by the objective function determination method provided in the embodiment of the present application is shown in Figure 10 , because the objective function is relatively convex, the velocity structure can be well inverted, and the final synthetic data diagram and the final data comparison diagram obtained on this basis are shown in Figure 11 and Figure 12 , respectively. As can be seen, the whole seismic signal is fitted.

[0099] The present application provides a method for determining an objective function of waveform inversion. The distance matrix is constructed to determine the optimal permutation vector, and then the single-channel data deviation corresponding to each seismic channel is determined, and then the objective function of the waveform inversion is obtained. The optimal transmission under the nonlinear transformation is used to obtain a relatively convex objective function, and the period jump problem in the waveform inversion is solved.

[0100] In the fifth embodiment, please refer to Figure 13 , Figure 13 is a structural block diagram of an embodiment of a device for determining an objective function of waveform inversion, and comprises:

[0101] The transformation module 501 is configured to perform nonlinear transformation on the obtained seismic observation data and seismic synthetic data.

[0102] The matrix construction module 502 is configured to construct a distance matrix according to the seismic synthetic data, the seismic observation data, and the seismic channel data after the nonlinear transformation.

[0103] The optimal permutation vector determination module 503 is configured to determine an optimal permutation vector from the distance matrix.

[0104] The single-channel data deviation determination module 504 is configured to determine the single-channel data deviation corresponding to each seismic channel by combining the optimal permutation vector and a pre-set deviation calculation expression.

[0105] The target function determination module 505 is configured to sum all the single-channel data deviations to obtain a target function of full waveform inversion.

[0106] In an optional embodiment, the matrix construction module 502 comprises:

[0107] The time difference normalization value determination submodule is configured to normalize time differences between synthetic single-channel data of all the seismic synthetic data and observed single-channel data corresponding to the seismic observation data to obtain time difference normalization values.

[0108] The amplitude difference normalization value determination submodule is configured to determine amplitude differences of the seismic trace data after the nonlinear transformation and normalize the amplitude differences to obtain amplitude difference normalization values.

[0109] The construction submodule is configured to obtain matrix elements of the distance matrix based on the time difference normalization values and the amplitude difference normalization values and construct the distance matrix based on the matrix elements.

[0110] In an optional embodiment, the optimal permutation vector determination module 503 comprises:

[0111] The sorting result determination submodule is configured to sort all the matrix elements of the distance matrix by using a bid ranking algorithm to obtain a sorting result.

[0112] The optimal permutation vector determination submodule is configured to determine the optimal permutation vector according to the sorting result. In an optional embodiment, the deviation calculation expression is specifically as follows:

[0113] h = ∑ ij c ij m ij ;

[0114] wherein h is a deviation, c ij is an element in the i-th row and the j-th column of the distance matrix, and m ij is an element in the i-th row and the j-th column of the corresponding permutation vector.

[0115] Embodiment six, the embodiments of the present application also provide an electronic device comprising a memory and a processor, the memory stores a computer program, and the computer program is executed by the processor to make the processor execute the steps of a target function determination method for full waveform inversion, comprising:

[0116] Performing nonlinear transformation on the obtained seismic observation data and seismic synthetic data.

[0117] Constructing a distance matrix according to the seismic synthetic data, the seismic observation data, and the seismic trace data after the nonlinear transformation.

[0118] determining an optimal permutation vector from the distance matrix;

[0119] determining a single-channel data bias corresponding to each seismic trace by a preset bias calculation expression combined with the optimal permutation vector;

[0120] summing all the single-channel data biases to obtain a target function of full waveform inversion.

[0121] In an optional embodiment, a distance matrix is constructed according to the seismic synthetic data, the seismic observation data, and the seismic trace time after nonlinear transformation, comprising:

[0122] normalizing a time difference between synthetic single-channel data of all the seismic synthetic data and observed single-channel data corresponding to the seismic observation data to obtain a time difference normalized value;

[0123] determining an amplitude difference of the seismic trace data after nonlinear transformation, and performing normalization processing on the amplitude difference to obtain an amplitude difference normalized value;

[0124] obtaining matrix elements of the distance matrix based on the time difference normalized value and the amplitude difference normalized value, and constructing the distance matrix based on the matrix elements.

[0125] In an optional embodiment, the optimal permutation vector is determined from the distance matrix, comprising:

[0126] sorting all the matrix elements of the distance matrix by a bid ranking algorithm to obtain a sorting result;

[0127] determining the optimal permutation vector according to the sorting result.

[0128] In an optional embodiment, the bias calculation expression is specifically:

[0129] h = ∑ ij c ij m ij ;

[0130] wherein h is a bias, c ij is an element in the i-th row and the j-th column of the distance matrix, and m ij is an element in the i-th row and the j-th column of the corresponding permutation vector.

[0131] Embodiment seven, the embodiments of the present application further provide a computer storage medium having a computer program stored thereon, the computer program is executed by the processor to realize the steps of a target function determination method of full waveform inversion, comprising:

[0132] performing nonlinear transformation on the obtained seismic observation data and seismic synthetic data;

[0133] constructing a distance matrix according to the seismic synthetic data, the seismic observation data, and the seismic trace data after nonlinear transformation;

[0134] determining an optimal permutation vector from the distance matrix;

[0135] determining a single-channel data deviation corresponding to each seismic trace according to a preset deviation calculation expression and the optimal permutation vector;

[0136] summing all the single-channel data deviations to obtain a target function of full waveform inversion.

[0137] In an optional embodiment, the distance matrix is constructed according to the seismic synthetic data, the seismic observation data, and the seismic trace time after nonlinear transformation, and includes:

[0138] normalizing a time difference between synthetic single-channel data of all the seismic synthetic data and observed single-channel data corresponding to the seismic observation data to obtain a time difference normalized value;

[0139] determining an amplitude difference of the seismic trace data after nonlinear transformation, and performing normalization processing on the amplitude difference to obtain an amplitude difference normalized value;

[0140] obtaining matrix elements of the distance matrix based on the time difference normalized value and the amplitude difference normalized value, and constructing the distance matrix based on the matrix elements.

[0141] In an optional embodiment, the optimal permutation vector is determined from the distance matrix, and includes:

[0142] sorting all the matrix elements of the distance matrix by a bid ranking algorithm to obtain a sorting result;

[0143] determining the optimal permutation vector according to the sorting result.

[0144] In an optional embodiment, the deviation calculation expression is specifically:

[0145] h = ∑ ij c ij m ij ;

[0146] wherein h is a deviation, c ij is an element in the i th row and the j th column of the distance matrix, and m ij is an element in the i th row and the j th column of the corresponding permutation vector.

[0147] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the foregoing method embodiments, which will not be repeated here.

[0148] In several embodiments provided in the present application, it should be understood that the disclosed methods, devices, electronic devices and storage media can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of 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, and can be electrical, mechanical or other forms.

[0149] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0150] In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of software functional unit.

[0151] When the integrated unit is realized in the form of software functional 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 present application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a readable storage medium, including a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the present application. The foregoing readable storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk, and various storage program codes.

[0152] The above-described embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit the present application; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those skilled in the art that the technical solutions recorded in the foregoing embodiments can still be modified, or some technical features can be replaced by equivalent replacements; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for determining the objective function through full waveform inversion, characterized in that, include: Nonlinear transformations are performed on the acquired seismic observation data and seismic composite data; A distance matrix is ​​constructed based on the synthetic seismic data, the seismic observation data, and the seismic trace data after nonlinear transformation; Determine the optimal permutation vector from the distance matrix; By using a pre-defined deviation calculation expression and combining it with the optimal permutation vector, the single-channel data deviation corresponding to each seismic trace is determined. Summing all the single-channel data deviations yields the objective function for full waveform inversion.

2. The method for determining the objective function of full waveform inversion according to claim 1, characterized in that, Based on the synthetic seismic data, the seismic observation data, and the nonlinearly transformed seismic trace time, a distance matrix is ​​constructed, including: Normalize the time difference between the synthetic single-channel data of all the aforementioned seismic synthetic data and the observation single-channel data of the corresponding seismic observation data to obtain the time difference normalization value; The amplitude difference of the seismic trace data after nonlinear transformation is determined, and the amplitude difference is normalized to obtain the normalized value of amplitude difference. Based on the normalized time difference value and the normalized amplitude difference value, the matrix elements of the distance matrix are obtained, and the distance matrix is ​​constructed based on the matrix elements.

3. The method for determining the objective function of full waveform inversion according to claim 2, characterized in that, Determining the optimal permutation vector from the distance matrix includes: The distance matrix is ​​sorted using a bidding ranking algorithm to obtain the sorting result; Based on the sorting results, the optimal permutation vector is determined.

4. The method for determining the objective function of full waveform inversion according to claim 1, characterized in that, The specific expression for calculating the deviation is as follows: ； in, For deviation, It is a distance matrix. Let be the permutation matrix.

5. A device for determining the objective function through full waveform inversion, characterized in that, include: The transformation module is used to perform nonlinear transformations on the acquired seismic observation data and seismic synthetic data; The matrix construction module is used to construct a distance matrix based on the synthetic seismic data, the seismic observation data, and the nonlinearly transformed seismic trace data. The optimal permutation vector determination module is used to determine the optimal permutation vector from the distance matrix; The single-channel data deviation determination module is used to determine the single-channel data deviation corresponding to each seismic channel by combining the pre-set deviation calculation expression with the optimal permutation vector. The objective function determination module is used to sum all the single-channel data deviations to obtain the objective function for full waveform inversion.

6. The apparatus for determining the objective function of full waveform inversion according to claim 5, characterized in that, The matrix construction module includes: The time difference normalization value determination submodule is used to normalize the time difference between the synthetic single-channel data of all the said seismic synthetic data and the observation single-channel data of the corresponding seismic observation data to obtain the time difference normalization value. The amplitude difference normalization value determination submodule is used to determine the amplitude difference of the seismic trace data after nonlinear transformation, and to normalize the amplitude difference to obtain the amplitude difference normalization value. A submodule is constructed to obtain the matrix elements of the distance matrix based on the normalized value of the time difference and the normalized value of the amplitude difference, and to construct the distance matrix based on the matrix elements.

7. The apparatus for determining the objective function of full waveform inversion according to claim 6, characterized in that, The optimal permutation vector determination module includes: The sorting result determination submodule is used to sort all the matrix elements of the distance matrix using a bidding ranking algorithm to obtain the sorting result; The optimal permutation vector determination submodule is used to determine the optimal permutation vector based on the sorting result.

8. The method for determining the objective function of full waveform inversion according to claim 5, characterized in that, The specific expression for calculating the deviation is as follows: ； in, For deviation, It is a distance matrix. Let be the permutation matrix.

9. An electronic device, characterized in that, It includes a processor and a memory, the memory storing computer-readable instructions that, when executed by the processor, perform the method as described in any one of claims 1-4.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it performs the method as described in any one of claims 1-4.

Images