VTI medium Kirchhoff integral prestack depth migration imaging method based on PINN travel time calculation

Through the travel time calculation method based on PINN, a network of VTI medium equations was constructed, the depth domain velocity field was trained, and the travel time information of multiple sources of seismic points was obtained. Combined with the Kshchhoff integral offset formula, the VTI medium was solved. The problem of insufficient accuracy and efficiency in the integral offset process achieves efficient and accurate depth offset imaging.

CN118962796APending Publication Date: 2024-11-15CHINA NATIONAL OFFSHORE OIL (CHINA) CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411021778.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-29
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

In the prior art In the process of Kshchhoff integral offset, there are insufficient calculation accuracy and efficiency of VTI media when traveling, especially when calculating multi-source position, the difference approximation caused by the finite difference method limits the calculation efficiency and accuracy.

Method used

Using PINN-based travel time calculation method, by constructing a PINN network of VTI medium equation based on factorization, training the depth domain velocity field, obtaining travel time information of multiple sources of seismic points, and combining the Kshchhoff integral offset formula Achieve offset imaging.

Benefits of technology

The accuracy and efficiency of VTI medium Keshchhof's pre-stack depth offset imaging is improved, and it is suitable for imaging processing of complex structural areas, exceeding the accuracy and efficiency of conventional FSM methods.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118962796A_ABST
    Figure CN118962796A_ABST
Patent Text Reader

Abstract

The invention discloses a VTI medium Kirchhoff integral prestack depth migration imaging method based on PINN travel time calculation. The VTI medium Kirchhoff integral prestack depth migration imaging method comprises the following steps: constructing a PINN network based on a factorization VTI medium eikfunction equation as a PINN loss function; performing sampling training on the known VTI medium depth domain velocity field by using a PINN network to obtain a trained PINN network; solving multi-focus point travel time information of a VTI medium depth domain velocity field by using the trained PINN network; and realizing migration imaging of the VTI medium by using a Kirchhoff integral migration formula according to the travel time information and the seismic record of the multiple seismic source points. Therefore, the PINN-based travel time calculation method efficiently and accurately realizes travel time calculation of the depth domain velocity field of the VTI medium, and accurately realizes Kirchhoff integral prestack depth migration imaging of the VTI medium according to travel time information and seismic records.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to a VTI medium Kirchhoff integral prestack depth migration imaging method, device, equipment and medium based on PINN travel time calculation, and relates to the field of seismic data processing. Background Art

[0002] Seismic migration imaging is a key step in seismic exploration data processing, and it is also one of the difficulties in processing deep complex geological structure data. Kirchhoff integrated prestack depth migration imaging is widely used in industrial production because it is not limited by the observation system, has good target imaging effect and high computational efficiency. With the continuous improvement of exploration accuracy requirements, conventional isotropic Kirchhoff integrated migration can no longer meet the needs of actual production. Kirchhoff integrated migration based on vertical transverse isotropic media (VTI) has become a hot topic in current research.

[0003] Travel time calculation is the core step of Kirchhoff integral migration. At present, the calculation methods of seismic wave travel time include ray-based methods based on high-frequency ray approximation theory, wavefront construction methods based on Huygens principle, and numerical solution methods based on eikonal equations. Among them, the numerical solution method based on eikonal equations has the highest accuracy. There are mainly FMM (fast marching method) and FSM (fast scanning method) for VTI media travel time solutions. The two methods have similar accuracy, but FSM has higher computational efficiency. Therefore, FSM is currently the best method for VTI media travel time solution.

[0004] Although the FSM method has a good performance in solving the travel time of VTI media, its accuracy needs to be further improved due to the fact that the finite difference method itself uses differential approximation to replace differential. In addition, in the process of Kirchhoff integral migration, it is necessary to obtain the travel time of multiple source positions continuously, and the FSM method itself needs to consider the calculation order between grid points when solving, which excludes the possibility of improving the calculation efficiency of the FSM method through parallel calculation. Based on the influence of the above factors, the efficiency and accuracy of the calculation of multi-position source travel time in Kirchhoff integral migration need to be further improved. Summary of the invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, in view of the above problems, the purpose of the present invention is to provide a VTI medium Kirchhoff integral prestack depth migration imaging method, device, equipment and medium based on PINN travel time calculation to achieve more efficient and accurate VTI medium Kirchhoff integral prestack depth migration imaging.

[0006] In order to achieve the above-mentioned invention object, the technical solution adopted by the present invention is:

[0007] In a first aspect, the present invention provides a VTI medium Kirchhoff integral prestack depth migration imaging method based on PINN travel time calculation, the method comprising:

[0008] The PINN network is constructed based on the factorized VTI medium eikonal equation as part of the PINN loss function;

[0009] The constructed PINN network is used to sample and train the known VTI medium depth domain velocity field to obtain the trained PINN network;

[0010] The trained PINN network is used to obtain the multi-source travel time information of the velocity field in the depth domain of the VTI medium.

[0011] Based on the travel time information of multiple seismic source points in the depth domain velocity field of VTI media and seismic records, the Kirchhoff integral migration formula is used to realize the migration imaging of VTI media.

[0012] In some possible implementations, the PINN network is constructed based on the factorized VTI medium eikonal equation as part of the PINN loss function using the formula:

[0013]

[0014] In the formula, I represents the training set, N I represents the number of sampling points in the training set, (x s ,z s ) is the earthquake source coordinate, represents the spatial coordinates of the sample points in the training set, and L represents the predicted value of the unknown factor in the neural network training process. The residual of the factorized VTI medium eikonal equation is obtained. is the factor value given at the earthquake source.

[0015] In some possible implementations, the constructed PINN network is used to perform sampling training on the known VTI medium depth domain velocity field to obtain a trained PINN network. The specific process is as follows:

[0016] The constructed PINN network is used to sample the known VTI medium depth domain velocity field. A set proportion of grid points in the VTI medium depth domain velocity field are randomly selected as the training set for network training. The input parameters are the velocity grid point coordinates (x, z) and the source location coordinates (sx, sz). The output parameters are the unknown factor prediction values.

[0017] Some possible implementations of using the trained PINN network to obtain the multi-source travel time information of the VTI medium depth domain velocity field include:

[0018] After the deep neural network training is completed, the coordinates of the grid points in the area to be solved are used as the input parameters of the PINN network to obtain the unknown factor solutions of all grid points. The final travel time solution consists of the known factor part T0(x,z) calculated analytically and the unknown factor part predicted by the network Multiplying them gives:

[0019]

[0020] Among them, T(x,z) is the final travel time of the area to be solved.

[0021] In some possible implementations, based on the multi-source travel time information of the VTI medium depth domain velocity field and seismic records, the Kirchhoff integral migration formula is used to implement the migration imaging of the VTI medium using the formula:

[0022]

[0023] Among them, (S i ,G j ) represents the shot-check pair, IM(P) is the multi-shot stacking imaging result of the profile sample point, IG(i,P) is a data in the imaging gather, the migration result of the i-th shot record, P is the coordinate of the profile sample point, d i is the shot collection record, t is the time, i and j are the shot number and track number respectively, τ VTI (S i ,G j , P) is the travel time of the sample point in the VTI medium in the corresponding gun-detection pair, W M (S i ,G j ,P) is the weight factor.

[0024] In some possible implementations, W M (S i ,G j ,P) is expressed as:

[0025]

[0026] Where d is the vertical distance from the profile sample point to the observation horizontal plane, v(P) is the velocity of the profile sample point, r is the distance from the profile sample point to the detection point, and r0 is the distance from the profile sample point to the shot point.

[0027] In a second aspect, the present invention further provides a VTI medium Kirchhoff integral prestack depth migration imaging device based on PINN travel time calculation, the device comprising:

[0028] A network construction unit is configured to construct a PINN network based on the factorized VTI medium eikonal equation as a part of the PINN loss function;

[0029] A network training unit is configured to perform sampling training on a known VTI medium depth domain velocity field using the constructed PINN network to obtain a trained PINN network;

[0030] The travel time calculation unit is configured to obtain the travel time information of multiple earthquake source points of the velocity field in the depth domain of the VTI medium by using the trained PINN network;

[0031] The migration imaging unit is configured to realize the migration imaging of the VTI medium by using the Kirchhoff integral migration formula according to the multi-source point travel time information and seismic records of the VTI medium depth domain velocity field.

[0032] In some possible implementations, the PINN network is constructed based on the factorized VTI medium eikonal equation as part of the PINN loss function using the formula:

[0033]

[0034] In the formula, I represents the training set, N I represents the number of sampling points in the training set, (x s ,z s ) is the earthquake source coordinate, represents the spatial coordinates of the sample points in the training set, and L represents the predicted value of the unknown factor in the neural network training process. The residual of the factorized VTI medium eikonal equation is obtained. is the factor value given at the earthquake source.

[0035] In a third aspect, the present invention further provides an electronic device, comprising:

[0036] at least one processor; and

[0037] A memory communicatively connected to the processor; wherein,

[0038] The memory stores instructions executable by the processor, and the instructions are executed by the processor to enable the processor to perform any one of the methods described above.

[0039] In a fourth aspect, the present invention further provides a computer-readable storage medium storing one or more programs, wherein the one or more programs include computer instructions, and the computer instructions are used to enable a computer to execute any one of the methods described.

[0040] The present invention adopts the above technical solution, and has the following characteristics:

[0041] 1. The PINN-based traveltime calculation method of the present invention efficiently and accurately realizes the traveltime of the VTI medium depth domain velocity field, and finally accurately realizes the Kirchhoff integrated prestack depth migration imaging of the VTI medium based on the obtained traveltime information and known seismic records.

[0042] 2. The VTI medium Kirchhoff integral prestack depth migration based on PINN travel time calculation of the present invention can adapt to the imaging processing of complex structural areas, and has higher imaging accuracy and calculation efficiency than the Kirchhoff integral prestack depth migration based on conventional VTI fast scanning method travel time calculation. It is a more accurate and efficient VTI medium Kirchhoff integral prestack depth migration imaging method.

[0043] 3. The migration imaging process of the present invention is divided into four steps in total. The first step is to construct a PINN network by using the factorized VTI medium eikonal equation as part of the PINN loss function. The second step is to use PINN to sample and train the known VTI medium depth domain velocity field. The third step is to use the trained PINN network to obtain the travel time information of multiple earthquake source points. The fourth step is to use the Kirchhoff integral migration formula to realize the migration imaging of the VTI medium based on the obtained travel time information of multiple earthquake source points in the depth domain velocity field of the VTI medium in combination with known earthquake records. Therefore, the VTI medium travel time calculation based on the physical information driven neural network (PINN) of the present invention is compared with the conventional FSM method, which uses the automatic differentiation idea to solve the partial differential equation and has higher accuracy. In addition, the present invention uses the earthquake source position as the neural network input parameter and introduces a pre-training strategy based on the transfer learning idea, which greatly improves the efficiency of the VTI medium continuous multi-source position travel time calculation.

[0044] In summary, the present invention can be widely used in seismic data processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Throughout the accompanying drawings, the same reference numerals are used to represent the same components. In the accompanying drawings:

[0046] Figure 1 is a flow chart of a method according to an embodiment of the present invention;

[0047] Figure 2 A structure diagram of VTI medium travel time calculation based on PINN according to an embodiment of the present invention;

[0048] Figure 3 This is a PINN training flow chart of an embodiment of the present invention;

[0049] Figure 4 The actual data velocity model travel time isochrone based on VTIPINN and VTI FSM travel time calculation according to an embodiment of the present invention;

[0050] Figure 5 A single shot seismic record obtained after preprocessing actual data in an embodiment of the present invention;

[0051] Figure 6 The Kirchhoff integral migration imaging section of the actual data obtained based on the VTI FSM travel time according to the embodiment of the present invention;

[0052] Figure 7 The Kirchhoff integral migration imaging section of the actual data obtained based on the VTIPINN travel time according to the embodiment of the present invention;

[0053] Figure 8 A partial enlargement of the Kirchhoff integral migration imaging section of actual data based on different travel time calculation methods according to an embodiment of the present invention Figure 1 , where (a) is obtained based on VTI FSM travel time; (b) is obtained based on VTIPINN travel time;

[0054] Fig. 9 A partial enlargement of the Kirchhoff integral migration imaging section of actual data based on different travel time calculation methods according to an embodiment of the present invention Figure 2 , where (a) is obtained based on VTI FSM travel time; (b) is obtained based on VTIPINN travel time;

[0055] Fig.10 FIG. 4 is a structural diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0056] It should be understood that the terms used herein are only for the purpose of describing specific example embodiments and are not intended to be limiting. Unless the context clearly indicates otherwise, the singular forms "one", "an" and "said" as used herein may also be meant to include plural forms. The terms "include", "comprise", "contain", and "have" are inclusive, and therefore specify the existence of stated features, steps, operations, elements and / or parts, but do not exclude the existence or addition of one or more other features, steps, operations, elements, parts, and / or combinations thereof. The method steps, processes, and operations described herein are not interpreted as necessarily requiring them to be performed in the specific order described or illustrated, unless the execution order is clearly indicated. It should also be understood that additional or alternative steps may be used.

[0057] Although the terms first, second, third, etc. can be used in the text to describe multiple elements, components, regions, layers and / or sections, these elements, components, regions, layers and / or sections should not be limited by these terms. These terms can only be used to distinguish an element, component, region, layer or section from another region, layer or section. Unless the context clearly indicates, terms such as "first", "second" and other numerical terms do not imply order or sequence when used in the text. Therefore, the first element, component, region, layer or section discussed below can be referred to as the second element, component, region, layer or section without departing from the teaching of the example embodiments.

[0058] For ease of description, spatially relative terms may be used herein to describe the relationship of one element or feature relative to another element or feature as shown in the figures, such as "inside", "outside", "inner side", "outer side", "below", "above", etc. Such spatially relative terms are intended to include different orientations of the device in use or operation in addition to the orientation depicted in the figures.

[0059] The present invention provides a VTI medium Kirchhoff integral prestack depth migration imaging method, device, equipment and medium based on PINN travel time calculation, including: constructing a PINN network based on the factorized VTI medium eikonal equation as a part of the PINN loss function; using the constructed PINN network to sample and train the known VTI medium depth domain velocity field to obtain the trained PINN network; using the trained PINN network to obtain the multi-source point travel time information of the VTI medium depth domain velocity field; according to the multi-source point travel time information of the VTI medium depth domain velocity field and seismic records, using the Kirchhoff integral migration formula to realize the migration imaging of the VTI medium. Therefore, the present invention first uses PINN to calculate the multi-source point travel time of the known VTI medium depth domain velocity field, and then performs Kirchhoff integral prestack depth migration imaging according to the obtained travel time information and the known seismic records. The actual data processing results show that the VTI medium Kirchhoff integral migration based on PINN traveltime calculation can adapt to the imaging processing of complex structural areas, and has higher accuracy than the conventional VTI medium Kirchhoff integral migration imaging based on FSM traveltime calculation. It is an efficient and accurate VTI medium Kirchhoff integral prestack depth migration imaging method.

[0060] The exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0061] Embodiment 1: Figure 1 As shown, the VTI medium Kirchhoff integral prestack depth migration imaging method based on PINN travel time calculation provided in this embodiment uses PINN to obtain the travel time information of multiple earthquake source points of the known VTI medium depth domain velocity field, including:

[0062] S1. The factorized VTI medium eikonal equation is used as part of the PINN loss function to construct the PINN network.

[0063] In this embodiment, the VTI medium eikonal equation based on factorization is used as a part of the PINN loss function to construct the PINN network using formula (1). The construction formula is:

[0064]

[0065] In the formula, ξ represents the constructed PINN network, I represents the training set, and N I represents the number of sampling points in the training set, (x s ,z s ) is the source coordinate, (x*,z*) represents the spatial coordinates of the sample points in the training set, and L represents the predicted value of the unknown factor in the neural network training process. The residual of the VTI medium eikonal equation based on factorization is obtained. is the factor value given at the earthquake source, and its value is 1.

[0066] S2. Use PINN to sample and train the known VTI medium depth domain velocity field to obtain a trained PINN network.

[0067] In this embodiment, the constructed PINN network is used to sample the known VTI medium depth domain velocity field. 50% of the grid points in the VTI medium depth domain velocity field are randomly selected as the training set for network training. The input parameters are the velocity grid point coordinates (x, z) and the source position coordinates (x s ,z s ), the output parameter is the predicted value of the unknown factor

[0068] S3. Use the trained PINN network to obtain the travel time information of multiple earthquake source points.

[0069] In this embodiment, the trained PINN network is used to obtain the travel time information of multiple earthquake source points, and the velocity grid point coordinates (x, z) and the earthquake source position coordinates (x s ,z s ) as the input parameter of the PINN network, and the corresponding PINN network output, i.e. the unknown factor prediction value, is obtained The final travel time solution consists of the known factor part T0(x,z) calculated analytically and the unknown factor part predicted by the network Multiplying them gives:

[0070]

[0071] Among them, T(x,z) is the final travel time of the area to be solved. Therefore, the multi-source travel time of the VTI medium depth domain velocity field is calculated using the PINN method, thereby more efficiently obtaining more accurate travel time information of the VTI medium depth domain velocity field, thereby obtaining more accurate VTI medium Kirchhoff integrated prestack depth migration imaging results.

[0072] S4. Based on the obtained travel time information of multiple earthquake source points of the VTI medium depth domain velocity field and the known seismic records, the Kirchhoff integral migration formula is used to realize the migration imaging of the VTI medium.

[0073] In this embodiment, based on the obtained multi-source travel time information of the VTI medium depth domain velocity field and in combination with known seismic records, the Kirchhoff integral migration formula is used to realize the migration imaging of the VTI medium, which is realized by formula (2):

[0074]

[0075] Among them, (S i ,G j ) represents the shot-check pair, IM(P) is the multi-shot stacking imaging result of the profile sample point, IG(i,P) is a data in the imaging gather (the migration result of the i-th shot record), P is the coordinate of the profile sample point, d i is the shot collection record, t is the time, i and j are the shot number and track number respectively, I and J are the total number of shots and tracks, τ VTI (S i ,G j , P) is the travel time of the sample point in the VTI medium in the corresponding gun-detection pair, W M (S i ,G j ,P) is the weight factor, and its expression is as follows:

[0076]

[0077] Where d is the vertical distance from the profile point to the observation plane, v(P) is the velocity of the profile point, r is the distance from the profile point to the detector point, and r0 is the distance from the profile point to the shot point. The Kirchhoff integrated migration imaging profile of VTI medium based on PINN travel time calculation can be obtained by formula (2).

[0078] The following describes in detail the implementation process of the VTI medium Kirchhoff integral prestack depth migration imaging method based on PINN travel time calculation of the present invention through specific embodiments, and proposes a profile event identification method, including:

[0079] (1) Figure 2 As shown, the factorized VTI medium eikonal equation is used as a part of the PINN loss function to construct the PINN network.

[0080] (2) Based on the actual data of the L1 survey line in a certain work area, the inversion was performed to obtain the velocity field in the depth domain of the VTI medium. The total length of the survey line is 43.4 km. The data is collected by towline acquisition. There are 1456 shots in the source, and the data of each shot is received by 564 channels. The shot interval is 25 m, the channel interval is 12.5 m, the shot point depth is 12.5 m, the receiver point depth is 12.5 m, the minimum offset is 150 m, the time sampling interval is 4 ms, and the record length is 5 s. Figure 3 As shown, a randomly initialized neural network is first constructed, and then the neural network parameters are optimized based on minimizing the loss function. Finally, PINN is used to sample and train the depth domain velocity field of the VTI medium to obtain a trained PINN network.

[0081] (3) The trained PINN network is used to obtain the travel time information of multiple earthquake source points, where the travel time information of a certain earthquake source location is as follows: Figure 4 shown.

[0082] (4) Based on the obtained travel time information of multiple earthquake source points of the VTI medium depth domain velocity field, combined with known earthquake records.

[0083] In this embodiment, Figure 5 The single-shot seismic record shown in the figure uses the Kirchhoff integral migration formula to achieve migration imaging of VTI media. Figure 6 The figure shows the Kirchhoff integrated migration imaging profile of the actual data obtained based on the VTI FSM travel time. Figure 7 The following is the Kirchhoff integrated migration imaging profile based on the actual data obtained by VTIPINN travel time. Figure 8 , Fig. 9 They are Figure 6 , Figure 7 The black circle and black rectangular frame of the imaging section are enlarged. Figure 6 compared to, Figure 7 The imaging section shown has better continuity of the phase axis in the black circle and black rectangular frame; the fault structure in the black rectangular frame is more clearly depicted. Figure 8 and Fig. 9 At the point indicated by the black arrow, Figure 8 (b) and Fig. 9 (b) The continuity of the event axis is stronger than Figure 8 (a) and Fig. 9 (a) Easier to track; Fig. 9 The black circle in Fig. 9 (a) Compared with Fig. 9 The diffraction waves of (b) converge better, the wave group contrast is obvious, the breakpoint imaging is clear, and the fault is clear. This shows that the VTI medium travel time calculation method based on PINN is suitable for the processing of actual data, and the imaging accuracy is higher than the VTI medium Kirchhoff integral migration imaging based on conventional FSM travel time calculation. In addition, in the process of obtaining the travel time of continuous multiple source positions, the total time used by the VTI FSM method is 346 minutes, and the total training and calculation time of the VTIPINN method is 115 minutes. This shows that in the processing of actual data, the computational efficiency of the VTI medium travel time calculation method based on PINN is higher than that of the conventional FSM method. The above processing results achieve the ultimate goal of solving the technical problem, that is, verify the effectiveness of the present invention.

[0084] In summary, the traveltime calculation method based on PINN of the present invention efficiently and accurately realizes the traveltime of the velocity field in the depth domain of the VTI medium, and finally accurately realizes the Kirchhoff integral prestack depth migration imaging of the VTI medium based on the traveltime information obtained and the known seismic records. The Kirchhoff integral prestack depth migration of the VTI medium based on the PINN traveltime calculation can be adapted to the imaging processing of complex structural areas, and has higher imaging accuracy and calculation efficiency than the Kirchhoff integral prestack depth migration based on the conventional VTI fast scanning method traveltime calculation, and is a more accurate and efficient VTI medium Kirchhoff integral prestack depth migration imaging method.

[0085] Embodiment 2: The above-mentioned embodiment 1 provides a VTI medium Kirchhoff integral prestack depth migration imaging method based on PINN travel time calculation. Correspondingly, this embodiment provides a VTI medium Kirchhoff integral prestack depth migration imaging device based on PINN travel time calculation. The device provided in this embodiment can implement the VTI medium Kirchhoff integral prestack depth migration imaging method based on PINN travel time calculation of embodiment 1, and the device can be implemented by software, hardware or a combination of software and hardware. For the convenience of description, the description of this embodiment is divided into various units and described separately according to their functions. Of course, the functions of each unit can be implemented in the same or multiple software and / or hardware during implementation. For example, the device may include integrated or separate functional modules or functional units to perform the corresponding steps in each method of embodiment 1. Since the device of this embodiment is basically similar to the method embodiment, the description process of this embodiment is relatively simple, and the relevant parts can refer to the partial description of embodiment 1. The embodiment of the VTI medium Kirchhoff integral prestack depth migration imaging device based on PINN travel time calculation provided by the present invention is only schematic.

[0086] In this embodiment, the present invention further provides a VTI medium Kirchhoff integral prestack depth migration imaging device based on PINN travel time calculation, the device comprising:

[0087] A network construction unit is configured to construct a PINN network based on the factorized VTI medium eikonal equation as a part of the PINN loss function;

[0088] A network training unit is configured to perform sampling training on a known VTI medium depth domain velocity field using the constructed PINN network to obtain a trained PINN network;

[0089] The travel time calculation unit is configured to obtain the travel time information of multiple earthquake source points of the velocity field in the depth domain of the VTI medium by using the trained PINN network;

[0090] The migration imaging unit is configured to realize the migration imaging of the VTI medium by using the Kirchhoff integral migration formula according to the multi-source point travel time information and seismic records of the VTI medium depth domain velocity field.

[0091] Embodiment 3: This embodiment provides an electronic device corresponding to the VTI medium Kirchhoff-integrated prestack depth migration imaging method based on PINN travel time calculation provided in Embodiment 1. The electronic device may be an electronic device for a client, such as a mobile phone, a laptop computer, a tablet computer, a desktop computer, etc., to execute the method of Embodiment 1.

[0092] like Fig.10As shown, the electronic device includes a processor, a memory, a communication interface and a bus, and the processor, the memory and the communication interface are connected through the bus to complete mutual communication. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Component (EISA) bus, etc. The memory stores a computer program that can be run on the processor, and the processor executes the method of embodiment 1 when running the computer program. The implementation principle and technical effect are similar to those of embodiment 1, and will not be repeated here. Those skilled in the art can understand that Fig.10 The structure shown in the figure is only a block diagram of a part of the structure related to the scheme of the present application, and does not constitute a limitation on the computing device to which the scheme of the present application is applied. The specific computing device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0093] In a preferred embodiment, the logic instructions in the above-mentioned memory can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art, and the computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), optical disk and other media that can store program codes.

[0094] In a preferred embodiment, the processor may be a central processing unit (CPU), a digital signal processor (DSP) or other general-purpose processors of various types, which are not limited here.

[0095] Embodiment 4: This embodiment provides a computer-readable storage medium storing one or more programs, wherein the one or more programs include computer instructions. When the computer instructions are executed by a computer, the computer executes the method provided in the above-mentioned embodiment 1.

[0096] Embodiment 5: This embodiment provides a computer program product. The computer program product may include a computer program stored on a computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the method provided in the above-mentioned embodiment 1. Its implementation principle and technical effects are similar to those of embodiment 1 and will not be repeated here.

[0097] In a preferred embodiment, the computer-readable storage medium may be a tangible device that holds and stores instructions used by the instruction execution device, such as but not limited to an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any combination thereof. The computer-readable storage medium stores computer program instructions that cause the computer to execute the method provided in the first embodiment.

[0098] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (apparatus), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0099] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0100] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0101] Each embodiment in this specification is described in a progressive manner, and the same and similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In the description of this specification, the description of reference terms "a preferred embodiment", "further", "specifically", "in the present embodiment", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the embodiment of this specification. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples without contradiction.

[0102] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A VTI medium Kirchhoff integral prestack depth migration imaging method based on PINN travel time calculation, characterized in that: The method includes: The PINN network is constructed based on the factorized VTI medium eikonal equation as part of the PINN loss function; The constructed PINN network is used to sample and train the known VTI medium depth domain velocity field to obtain the trained PINN network; The trained PINN network is used to obtain the multi-source travel time information of the velocity field in the depth domain of the VTI medium. Based on the travel time information of multiple seismic source points in the depth domain velocity field of VTI media and seismic records, the Kirchhoff integral migration formula is used to realize the migration imaging of VTI media.

2. The VTI medium Kirchhoff integrated prestack depth migration imaging method based on PINN travel time calculation according to claim 1 is characterized in that: The PINN network is constructed based on the factorized VTI medium eikonal equation as part of the PINN loss function using the formula: In the formula, I represents the training set, N I represents the number of sampling points in the training set, (x s ,z s ) is the earthquake source coordinate, (x * ,z * ) represents the spatial coordinates of the sample points in the training set, and L represents the predicted value of the unknown factor in the neural network training process. The residual of the factorized VTI medium eikonal equation is obtained. is the factor value given at the earthquake source.

3. The VTI medium Kirchhoff integrated prestack depth migration imaging method based on PINN travel time calculation according to claim 2 is characterized in that: The constructed PINN network is used to sample and train the known VTI medium depth domain velocity field to obtain the trained PINN network. The specific process is as follows: The constructed PINN network is used to sample the known VTI medium depth domain velocity field. A set proportion of grid points in the VTI medium depth domain velocity field are randomly selected as the training set for network training. The input parameters are the velocity grid point coordinates (x, z) and the source position coordinates (x s ,z s ), the output parameter is the predicted value of the unknown factor 4. The VTI medium Kirchhoff integrated prestack depth migration imaging method based on PINN travel time calculation according to claim 3 is characterized in that: The trained PINN network is used to obtain the multi-source travel time information of the VTI medium depth domain velocity field, including: The coordinates of the grid points in the area to be solved are used as the input parameters of the trained PINN network to obtain the unknown factor solutions of all grid points. The final travel time solution is composed of the known factor part T0(x,z) calculated analytically and the unknown factor part predicted by the network Multiplying them gives: Among them, T(x,z) is the final travel time of the area to be solved.

5. The VTI medium Kirchhoff integrated prestack depth migration imaging method based on PINN travel time calculation according to claim 1, characterized in that: According to the multi-source travel time information and seismic records of the VTI medium depth velocity field, the Kirchhoff integral migration formula is used to realize the migration imaging of the VTI medium. The formula is: Among them, (S i ,G j ) represents the shot-check pair, IM(P) is the multi-shot stacking imaging result of the profile sample point, IG(i,P) is a data in the imaging gather, the migration result of the i-th shot record, P is the coordinate of the profile sample point, d i is the shot collection record, t is the time, i and j are the shot number and track number respectively, τ VTI (S i ,G j , P) is the travel time of the sample point in the VTI medium in the corresponding gun-detection pair, W M (S i ,G j ,P) is the weight factor.

6. The VTI medium Kirchhoff integrated prestack depth migration imaging method based on PINN travel time calculation according to claim 5, characterized in that: W M (S i ,G j ,P) is expressed as: Where d is the vertical distance from the profile sample point to the observation horizontal plane, v(P) is the velocity of the profile sample point, r is the distance from the profile sample point to the detection point, and r0 is the distance from the profile sample point to the shot point.

7. A VTI medium Kirchhoff integrated prestack depth migration imaging device based on PINN travel time calculation, characterized in that: The device includes: A network construction unit is configured to construct a PINN network based on the factorized VTI medium eikonal equation as a part of the PINN loss function; A network training unit is configured to perform sampling training on a known VTI medium depth domain velocity field using the constructed PINN network to obtain a trained PINN network; The travel time calculation unit is configured to obtain the travel time information of multiple earthquake source points of the velocity field in the depth domain of the VTI medium by using the trained PINN network; The migration imaging unit is configured to realize the migration imaging of the VTI medium by using the Kirchhoff integral migration formula according to the multi-source point travel time information and seismic records of the VTI medium depth domain velocity field.

8. The VTI medium Kirchhoff integrated prestack depth migration imaging device based on PINN travel time calculation according to claim 7, characterized in that: The PINN network is constructed based on the factorized VTI medium eikonal equation as part of the PINN loss function using the formula: In the formula, I represents the training set, N I represents the number of sampling points in the training set, x s 、z s are the x and z coordinates of the earthquake source, respectively. * ,z * ) represents the spatial coordinates of the sample points in the training set, and L represents the predicted value of the unknown factor in the neural network training process. The residuals of the factorized VTI medium eikonal equation are obtained.

9. An electronic device, characterized in that: include: at least one processor; as well as A memory communicatively connected to the processor; wherein, The memory stores instructions executable by the processor, and the instructions are executed by the processor to enable the processor to perform the method according to any one of claims 1 to 6.

10. A computer-readable storage medium storing one or more programs, characterized in that: The one or more programs include computer instructions for causing a computer to execute the method according to any one of claims 1-6.