Full-waveform inversion method and device based on nonlinear transformation and storage medium

Abstract

The invention discloses a full waveform inversion method and device based on nonlinear transformation and a storage medium, and relates to the technical field of petroleum seismic exploration, and the method is characterized in that the method comprises the steps: obtaining actual seismic source data and an initial model; carrying out wavelet estimation, and determining a seismic source wavelet of each shot in the seismic source data; performing forward modeling calculation on the seismic source wavelets to obtain synthetic seismic data and observed seismic data; performing nonlinear transformation on the synthesized seismic data and the observed seismic data to obtain seismic data of positive signals; setting a target function, and calculating the seismic data of the positive signal to obtain an adjoint field seismic source; and according to the adjoint field seismic source, performing gradient iteration on the initial model to obtain an actual model of the stratum medium. Through the full-waveform inversion method, the overall amplitude distribution of the seismic data can be adjusted, so that the strong amplitude is suppressed, the high precision and high fidelity of full-waveform inversion are effectively improved, and the problem of cycle jump in full-waveform inversion is solved.

Description

Technical Field

[0001] The invention relates to the technical field of petroleum seismic exploration, and in particular to a full waveform inversion method, device and storage medium based on nonlinear transformation. Background Art

[0002] This section is intended to provide a background or context to the embodiments recited in the claims. No admission is made that the description herein is prior art by inclusion in this section.

[0003] In the field of seismology, receivers are used to record seismic signals that propagate through a subsurface region. The structure, geology, and other characteristics of the subsurface region can then be inferred from these signals in a process known as inversion. Full waveform inversion (FWI) is a type of inversion that uses the entire contents of a seismic trace to extract the physical parameters of the medium that the seismic waves sampled.

[0004] Full waveform inversion relies on solving the wave equation to simulate the propagation of the wave field underground, and then compares the difference between the synthetic seismogram and the observed data. The error of the back-transmitted data forms a gradient, and the model is updated to reduce the data error. The new model is used as the initial model and it is iterated repeatedly until the algorithm converges to the optimal value. Usually, the full waveform inversion uses the least squares waveform error, which not only fits the amplitude of the seismic signal, but also fits the phase of the seismic signal; therefore, theoretically, the size that can be resolved by full waveform inversion can reach half a wavelength. Since the objective function compares the waveform, the conventional full waveform inversion requires that the event difference between the initial seismogram and the observed data is within half a waveform, otherwise cycle jumps are prone to occur. However, the existing initial model of full waveform inversion is poor and inaccurate, resulting in relatively poor imaging quality and serious cycle jump problems. Summary of the invention

[0005] In response to the above technical problems, the present invention proposes a full waveform inversion method, device and storage medium based on nonlinear transformation, which can convert the seismic signal into a positive signal through nonlinear transformation, and apply the optimization function to obtain a more convex objective function, thereby adjusting the overall amplitude distribution of the seismic data, so that the strong amplitude is suppressed to a certain extent, effectively improving the high precision and high fidelity of the full waveform inversion, and solving the problem of periodic jumping in the full waveform inversion.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention includes four aspects.

[0007] In a first aspect, a full waveform inversion method based on nonlinear transformation is provided, comprising the following steps:

[0008] S1. Obtain actual earthquake source data and initial model;

[0009] S2, performing wavelet estimation on the actual seismic source data to determine the source wavelet of each shot in the seismic source data;

[0010] S3, performing forward calculation on the source wavelet on the initial model to obtain synthetic seismic data and observed seismic data;

[0011] S4, performing nonlinear transformation on the synthetic seismic data and the observed seismic data to obtain seismic data of positive signals;

[0012] S5, setting the target function, and calculating the seismic data of the positive signal according to the target function to obtain the accompanying field source;

[0013] S6. Perform gradient iteration on the initial model according to the associated field source; repeat steps S2-S6 until a model that meets the requirements is iterated to obtain an actual model of the formation medium.

[0014] In some embodiments, step S2 includes:

[0015] For the earthquake source data and the initial model, the source wavelet is estimated according to the Green's function theorem to obtain the source wavelet of each shot.

[0016] In some embodiments, the synthetic seismic data and the observed seismic data are stored in a two-dimensional matrix, the synthetic seismic data is recorded as p(x, t), and the observed seismic data is recorded as d(y, τ); the transformation formulas for performing nonlinear transformation on the synthetic seismic data and the observed seismic data are respectively:

[0017]

[0018]

[0019] In the formula, p i =p(x i ,t) is a row of the two-dimensional matrix p(x,t), d j =d(y j ,τ) is a row of the two-dimensional matrix d(y,τ); α i and β j It is a parameter given by the user, used to automatically adjust the amplitude range.

[0020] In some embodiments, the parameters given by the user in the nonlinear transformation include using one parameter for each seismic channel or using the same parameter for all seismic source data; the nonlinear transformation includes a combination of one or more of a hyperbolic tangent function, an exponential function or a logarithmic function.

[0021] In some embodiments, step S5 includes:

[0022] The observed seismic data and the synthetic seismic data in the seismic data of the positive signal are compared by using a two-dimensional optimal transfer function to obtain the optimal transfer function value;

[0023] The accompanying field source is obtained by calculation based on the optimized transfer function value and the chain rule.

[0024] In some embodiments, the formula of the two-dimensional optimization transfer function is:

[0025]

[0026] In the formula, x, y represent two-dimensional coordinates, t represents time, τ represents space, and p(x, t) represents synthetic seismic data;

[0027] The formula for obtaining the accompanying field source by calculating according to the optimized transfer function value and the chain rule is:

[0028]

[0029] Where h is the two-dimensional optimal transfer function value; p is the positive signal seismic data.

[0030] In some embodiments, step S6 includes:

[0031] Back propagation is performed to calculate the gradient according to the associated field source;

[0032] The initial model is iterated according to the gradient.

[0033] In a second aspect, a full waveform inversion device is provided, the device comprising:

[0034] An acquisition module is used to obtain actual earthquake source data and an initial model;

[0035] A data processing module is used to perform wavelet estimation on the actual seismic source data and determine the source wavelet of each shot in the seismic source data;

[0036] A forward modeling module, used for performing forward modeling calculation on the source wavelet on the initial model to obtain synthetic seismic data and observed seismic data;

[0037] A transformation module, used for performing nonlinear transformation on synthetic seismic data and observed seismic data to obtain seismic data of positive signals;

[0038] A solution calculation module is used to set the target function and calculate the seismic data of the positive signal according to the target function to obtain the accompanying field source;

[0039] The iterative execution module is used to perform gradient iteration on the initial model according to the accompanying field source, and repeat steps S2-S6 in the aforementioned full waveform inversion method until a model that meets the requirements is iterated to obtain an actual model of the formation medium.

[0040] In a third aspect, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the aforementioned full waveform inversion method when executing the computer program.

[0041] In a fourth aspect, a computer-readable storage medium is provided, wherein the computer program, when executed by a processor, implements the steps of the full waveform inversion method as described above.

[0042] Compared with the prior art, one or more embodiments of the above scheme may have the following advantages or beneficial effects:

[0043] The present application provides a full waveform inversion method, device and storage medium based on nonlinear transformation, and the full waveform inversion method includes the following steps: S1, obtaining actual seismic source data and an initial model; S2, performing wavelet estimation on the actual seismic source data to determine the source wavelet of each shot in the seismic source data; S3, performing forward calculation on the source wavelet on the initial model to obtain synthetic seismic data and observed seismic data; S4, performing nonlinear transformation on the synthetic seismic data and the observed seismic data to obtain seismic data of positive signals; S5, setting an objective function, and calculating the seismic data of positive signals according to the objective function to obtain an accompanying field source; S6, performing gradient iteration on the initial model according to the accompanying field source; repeating the steps S2-S6 until a model that meets the requirements is iterated to obtain an actual model of the formation medium. Through this full waveform inversion method, the seismic signal can be converted into a signal through nonlinear transformation, and a more convex objective function can be obtained by applying the optimization function, which adjusts the overall amplitude distribution of the seismic data so that the strong amplitude is suppressed to a certain extent. This feature effectively improves the high precision and high fidelity of the full waveform inversion and solves the problem of periodic jumping in the full waveform inversion. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Hereinafter, the present application will be described in more detail based on embodiments and with reference to the accompanying drawings;

[0045] Figure 1 An exemplary flow chart of a full waveform inversion method based on nonlinear transformation provided in Embodiment 1 of the present invention;

[0046] Figure 2 An example diagram of a single shot gather in the actual seismic source data provided in the first embodiment of the present invention;

[0047] Figure 3 This is an example diagram of the initial model provided in the first embodiment of the present invention;

[0048] Figure 4 is an example diagram of the source wavelet estimated in step S2 in the first embodiment of the present invention;

[0049] Figure 5 In the embodiment of the present invention, Figure 1 An exemplary flow chart of step S5 shown in;

[0050] Figure 6 In the embodiment of the present invention, Figure 1 An exemplary flow chart of step S6 shown in FIG.

[0051] Figure 7 This is an example diagram of the velocity model obtained in the first frequency band in the first embodiment of the present invention;

[0052] Figure 8 This is an example diagram of the velocity model obtained in the second frequency band in the first embodiment of the present invention;

[0053] Fig. 9 A schematic block diagram of a full waveform inversion device provided in Embodiment 2 of the present invention;

[0054] Fig.10 A schematic block diagram of an electronic device provided in Embodiment 3 of the present invention;

[0055] Fig.11 This is a schematic diagram of a computer-readable storage medium provided in Embodiment 4 of the present invention.

[0056] In the drawings, the same reference numerals are used for the same components and the drawings are not drawn to scale. DETAILED DESCRIPTION

[0057] The present disclosure is further described below in conjunction with the embodiments shown in the accompanying drawings.

[0058] Embodiment 1:

[0059] The present application embodiment discloses a full waveform inversion method based on nonlinear transformation, such as Figure 1As shown, the method comprises the following steps: S1, obtaining actual seismic source data and an initial model; S2, performing wavelet estimation on the actual seismic source data to determine the source wavelet of each shot in the seismic source data; S3, performing forward calculation on the source wavelet on the initial model to obtain synthetic seismic data and observed seismic data; S4, performing nonlinear transformation on the synthetic seismic data and observed seismic data to obtain seismic data of positive signals; S5, setting an objective function, and calculating the seismic data of positive signals according to the objective function to obtain the accompanying field source; S6, performing gradient iteration on the initial model according to the accompanying field source; repeating the steps S2-S6 until a model that meets the requirements is iterated to obtain the actual model of the stratum medium. Through the full waveform inversion method, the seismic signal can be converted into a signal through nonlinear transformation, and a more convex objective function can be obtained by applying the optimization function, so that the overall amplitude distribution of the seismic data is adjusted, so that the strong amplitude is suppressed to a certain extent. This characteristic effectively improves the high precision and high fidelity of the full waveform inversion and solves the cycle jump problem in the full waveform inversion.

[0060] A full waveform inversion method based on nonlinear transformation provided by at least one embodiment of the present disclosure can be implemented in the form of software, hardware, firmware or any combination thereof, and loaded and executed by a processor in a device such as a mobile phone, a tablet computer, a laptop computer, a desktop computer, a network server, etc., so as to obtain a more convex objective function by applying the optimized transfer function, thereby improving the high precision and high fidelity of the full waveform inversion and solving the problem of periodic jumps in the full waveform inversion.

[0061] Reference below Figure 1 As shown, a full waveform inversion method based on nonlinear transformation provided by at least one embodiment of the present disclosure is described, and the full waveform inversion method includes steps S1 to S6.

[0062] S1. Obtain actual earthquake source data and initial model.

[0063] In some embodiments, after obtaining the actual seismic source data, the actual seismic source data can be preprocessed to delete abnormal data in the actual observed seismic record data to obtain normal actual seismic source data, which is beneficial to subsequent inversion results.

[0064] Furthermore, the initial model can be to establish a uniform velocity model of the formation medium after velocity analysis of the formation medium, and then establish a series of initial models on this basis to obtain the local optimal solution of the initial model of the formation medium, wherein the formation medium can be a natural source of land or sea, so as to facilitate imaging through full waveform inversion.

[0065] like Figure 2 is a single shot gather from the provided observation data. Figure 3This is the initial model provided. It is relatively smooth and almost no structural information of the interface can be seen.

[0066] S2. Perform wavelet estimation on the actual seismic source data to determine the source wavelet of each shot in the seismic source data.

[0067] In some embodiments, step S2 includes: estimating the source wavelet according to the Green's function theorem for the seismic source data and the initial model to obtain the source wavelet of each shot, and Green's function filtering can effectively improve the accuracy of full waveform inversion. Figure 4 It is the source wavelet estimated based on the Green's function theorem, and the wavelets of different shot gathers are relatively consistent.

[0068] S3. Perform forward calculation on the source wavelet on the initial model to obtain synthetic seismic data and observed seismic data.

[0069] In some embodiments, synthetic seismic data and observed seismic data are stored in a two-dimensional matrix, where the synthetic seismic data is recorded as p(x, t) and the observed seismic data is recorded as d(y, τ), and their two-dimensional indicators are space and time, respectively, to facilitate nonlinear transformation of the data.

[0070] S4. Perform nonlinear transformation on the synthetic seismic data and the observed seismic data to obtain seismic data of positive signals.

[0071] In some embodiments, the transformation formulas for performing nonlinear transformation on synthetic seismic data and observed seismic data are:

[0072]

[0073]

[0074] In the formula, p i =p(x i ,t) is a row of the two-dimensional matrix p(x,t), d j =d(y j ,τ) is a row of the two-dimensional matrix d(y,τ); α i and β j It is a parameter given by the user, used to automatically adjust the amplitude range.

[0075] Furthermore, the parameters given by the user in the nonlinear transformation include using one parameter for the transformation of each seismic channel alone or using the same parameter for the transformation of all seismic source data; the nonlinear transformation includes a combination of one or more of a hyperbolic tangent function (tanh function), an exponential function (exp function) or a logarithmic function (log function). In this embodiment, the hyperbolic tangent function (tanh function) is preferably used for the transformation.

[0076] S5. Set the objective function, and calculate the seismic data of the positive signal according to the objective function to obtain the accompanying field source.

[0077] In some embodiments, Figure 5 As shown, step S5 includes:

[0078] S51, using a two-dimensional optimal transfer function to compare the observed seismic data and the synthetic seismic data in the seismic data of the positive signal to obtain an optimal transfer function value;

[0079] S52. Calculate according to the optimized transfer function value and the chain rule to obtain the accompanying field source.

[0080] By comparing the observed seismic data with the synthetic seismic data in a global comparison method with optimized transmission, the correlation between the time axis and the space axis of the shot gather can be effectively extracted, thus improving the accuracy of full waveform inversion.

[0081] Furthermore, the formula of the two-dimensional optimal transfer function is:

[0082]

[0083] In the formula, x, y represent two-dimensional coordinates, t represents time, τ represents space, and p(x, t) represents synthetic seismic data;

[0084] The formula for obtaining the accompanying field source by calculating according to the optimized transfer function value and the chain rule is:

[0085]

[0086] Where h is the two-dimensional optimal transfer function value; p is the positive signal seismic data.

[0087] S6. Performing gradient iteration on the initial model according to the associated field source.

[0088] In some embodiments, Figure 6 As shown, step S6 includes:

[0089] S61, performing back propagation calculation of the gradient according to the associated field source;

[0090] S62: Iterate the initial model according to the gradient.

[0091] In this embodiment, after determining the objective function of each seismic trace, the accompanying field source of each seismic trace can be calculated based on the objective function. After the accompanying field source of each seismic trace is calculated, it can be back-propagated along time to calculate the gradient to determine the direction and magnitude of the initial velocity model update, and then the current velocity model can be updated and iterated according to the gradient, so that the objective function value gradually decreases.

[0092] After iterating the initial velocity model, repeat steps S2-S6 until a model that meets the requirements is iterated to obtain the actual model of the formation medium. Specifically, obtain the iterated synthetic seismic record data, determine the objective function based on the iterated synthetic seismic record data, and calculate the objective function value. Then determine whether the objective function value meets the preset accuracy requirements. When the objective function value does not meet the accuracy requirements, continue to iterate the velocity model; when the objective function value meets the accuracy requirements, stop iterating and obtain the final actual velocity model. Among them, when the objective function meets the requirements, the corresponding velocity model is the actual model of the formation medium.

[0093] Figure 7 Shown is the resulting velocity model for the first frequency band; more structure is apparent compared to the initial model. Figure 8 Shown is the velocity model obtained in the second frequency band; it can be seen that many fine structures have been inverted.

[0094] The full waveform inversion method based on nonlinear transformation provided by the embodiments of the present disclosure can effectively extract the correlation along the time axis and the space axis of the shot set based on the global comparison method of optimized transmission, and at the same time, convert the seismic signal into a signal through nonlinear transformation, and apply the optimization function to obtain a more convex objective function, thereby adjusting the overall amplitude distribution of the seismic data so that the strong amplitude is suppressed to a certain extent. This feature effectively improves the high precision and high fidelity of the full waveform inversion, solves the problem of periodic jumps in the full waveform inversion, and reconstructs a high-precision and high-fidelity underground velocity structure.

[0095] At least one embodiment of the present disclosure further provides a full waveform inversion device, wherein each module in the device corresponds to the steps in the above method, and the effects achieved and the problems solved have been described above and will not be repeated here. Fig. 9 As shown, including:

[0096] An acquisition module 11 is used to acquire actual earthquake source data and an initial model;

[0097] The data processing module 12 is used to perform wavelet estimation on the actual seismic source data and determine the source wavelet of each shot in the seismic source data;

[0098] A forward modeling module 13 is used to perform forward modeling calculations on the source wavelet on the initial model to obtain synthetic seismic data and observed seismic data;

[0099] A transformation module 14 is used to perform nonlinear transformation on the synthetic seismic data and the observed seismic data to obtain seismic data of positive signals;

[0100] The solution calculation module 15 is used to set the target function and calculate the seismic data of the positive signal according to the target function to obtain the accompanying field source;

[0101] The iterative execution module 16 is used to perform gradient iteration on the initial model according to the accompanying field source, and repeat steps S2-S6 in the full waveform inversion method provided in any embodiment of the present disclosure until a model that meets the requirements is iterated to obtain the actual model of the formation medium.

[0102] In some embodiments, the output end of the acquisition module 11 is connected to the input end of the data processing module 12, the output end of the data preprocessing module 12 is connected to the input end of the forward modeling module 13, the output end of the forward modeling module 13 is connected to the input end of the transformation module 14, the output end of the transformation module 14 is connected to the input end of the solution calculation module 15, and the output end of the solution calculation module 15 is connected to the input end of the iterative execution module 16, thereby executing the steps of the full waveform inversion method of embodiment 1, so as to improve the high precision and high fidelity of the full waveform inversion, solve the periodic jump problem in the full waveform inversion, and reconstruct a high-precision and high-fidelity underground velocity structure.

[0103] In some embodiments, the device may be a part of the software, or may be implemented in combination with corresponding hardware, which will not be described in detail here.

[0104] Embodiment three:

[0105] At least some embodiments of the present disclosure also provide an electronic device, such as Fig.10 As shown, the electronic device includes a memory 21 and a processor 22, wherein the memory 21 stores a computer program, and when the computer program is executed by the processor, the steps of the full waveform inversion method provided in any embodiment of the present disclosure are performed.

[0106] In some embodiments, the processor 22 is used to execute all or part of the steps in the full waveform inversion method of any embodiment of the present disclosure. The memory 21 is used to store various types of data, which may include instructions of any application or method in the electronic device, and data related to the application.

[0107] The processor 22 can be an application specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field programmable gate array (FPGA), a controller, a microcontroller, a microprocessor or other electronic components, and is used to execute the full waveform inversion method in the above-mentioned embodiment 1.

[0108] The memory 21 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, 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, disk or optical disk.

[0109] Embodiment 4:

[0110] At least some embodiments of the present disclosure also provide a computer-readable storage medium, such as Fig.11 As shown, a computer program 31 is stored on the readable storage medium, and when the computer program 31 is executed by a processor, the steps of the full waveform inversion method provided by any embodiment of the present disclosure are implemented.

[0111] In some embodiments, the storage medium can be a tangible medium that can contain or store a program for use by an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the above. A more specific example of a machine-readable storage medium can include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0112] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0113] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0114] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0115] Embodiment five:

[0116] An embodiment of the present invention further provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the full waveform inversion method provided in any embodiment of the present disclosure.

[0117] The various embodiments in the present disclosure are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0118] The protection scope of the present disclosure is not limited to the above-mentioned embodiments. Obviously, those skilled in the art can make various changes and modifications to the present disclosure without departing from the scope and spirit of the present disclosure. If these changes and modifications fall within the scope of the claims of the present disclosure and their equivalents, the intention of the present disclosure also includes these changes and modifications.

Claims

1. A full waveform inversion method based on nonlinear transformation, characterized in that: The following steps are involved: S1. Obtain actual earthquake source data and initial model; S2, performing wavelet estimation on the actual seismic source data to determine the source wavelet of each shot in the seismic source data; S3, performing forward calculation on the source wavelet on the initial model to obtain synthetic seismic data and observed seismic data; S4, performing nonlinear transformation on the synthetic seismic data and the observed seismic data to obtain seismic data of positive signals; S5, setting the target function, and calculating the seismic data of the positive signal according to the target function to obtain the accompanying field source; S6. performing gradient iteration on the initial model according to the associated field source; Repeat steps S2-S6 until a model that meets the requirements is iterated to obtain the actual model of the formation medium.

2. The full waveform inversion method based on nonlinear transformation according to claim 1 is characterized in that: The step S2 comprises: For the earthquake source data and the initial model, the source wavelet is estimated according to the Green's function theorem to obtain the source wavelet of each shot.

3. The full waveform inversion method based on nonlinear transformation according to claim 1 is characterized in that: The synthetic seismic data and the observed seismic data are stored in a two-dimensional matrix, the synthetic seismic data is recorded as p(x, t), and the observed seismic data is recorded as d(y, τ); the transformation formulas for performing nonlinear transformation on the synthetic seismic data and the observed seismic data are respectively: In the formula, p i =p(x i , t) is a row of the two-dimensional matrix p(x, t), d j =d(y j ,τ) is a row of the two-dimensional matrix d(y,τ); α i and β j It is a parameter given by the user, used to automatically adjust the amplitude range.

4. The full waveform inversion method based on nonlinear transformation according to claim 1 is characterized in that: The parameters given by the user in the nonlinear transformation include using one parameter for transformation of each seismic channel or using the same parameter for transformation of all seismic source data; the nonlinear transformation includes a combination of one or more of a hyperbolic tangent function, an exponential function or a logarithmic function.

5. The full waveform inversion method based on nonlinear transformation according to claim 1, characterized in that: The step S5 comprises: The observed seismic data and the synthetic seismic data in the seismic data of the positive signal are compared by using a two-dimensional optimal transfer function to obtain the optimal transfer function value; The accompanying field source is obtained by calculation based on the optimized transfer function value and the chain rule.

6. A full waveform inversion method based on nonlinear transformation according to claim 5, characterized in that: The formula of the two-dimensional optimized transfer function is: In the formula, x, y represent two-dimensional coordinates, t represents time, τ represents space, and p(x, t) represents synthetic seismic data; The formula for obtaining the accompanying field source by calculating according to the optimized transfer function value and the chain rule is: Where h is the two-dimensional optimal transfer function value; p is the positive signal seismic data.

7. The full waveform inversion method based on nonlinear transformation according to claim 1, characterized in that: The step S6 comprises: Back propagation is performed to calculate the gradient according to the associated field source; The initial model is iterated according to the gradient.

8. A full waveform inversion device, characterized in that: The device comprises: An acquisition module is used to obtain actual earthquake source data and an initial model; A data processing module is used to perform wavelet estimation on the actual seismic source data and determine the source wavelet of each shot in the seismic source data; A forward modeling module, used for performing forward modeling calculation on the source wavelet on the initial model to obtain synthetic seismic data and observed seismic data; A transformation module, used for performing nonlinear transformation on synthetic seismic data and observed seismic data to obtain seismic data of positive signals; A solution calculation module is used to set the target function and calculate the seismic data of the positive signal according to the target function to obtain the accompanying field source; An iterative execution module is used to perform gradient iteration on the initial model according to the accompanying field source, and repeat steps S2-S6 in the full waveform inversion method described in any one of claims 1 to 7 until a model that meets the requirements is iterated to obtain an actual model of the formation medium.

9. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the full waveform inversion method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: When the computer program is executed by a processor, the steps of the full waveform inversion method according to any one of claims 1 to 7 are implemented.

Images