Reflection wave waveform inversion method and system based on demigration denoising, medium and equipment

Abstract

The invention relates to the comprehensive field of exploration geophysics and inversion, and discloses a reflected wave waveform inversion method and system based on demigration denoising, a medium and equipment, and the method comprises the steps: generating an offset domain common imaging point gather for an observation record through an extended reverse time migration imaging condition; local tilt superposition transformation is applied to the offset domain common imaging point gather to obtain an angle domain common imaging point gather, a tilt filter based on similarity measurement is calculated to obtain a new angle domain common imaging point gather, and a new offset domain common imaging point gather is obtained through local tilt superposition inverse transformation; demigration data is generated for the new offset distance domain common imaging point gather through demigration based on an extended imaging condition; and establishing a least square objective function of the prediction data and the demigration data, alternately iteratively updating the disturbance model and the background speed model, and outputting the background speed model after an iteration termination condition is met. According to the method, the problem that the inversion result is influenced by noise in the iteration process of reflected wave waveform inversion can be solved.

Description

Technical Field

[0001] The present invention relates to the field of exploration geophysics and inversion integrated technology, and in particular to a reflection wave waveform inversion method, system, medium and equipment based on de-migration and denoising. Background Art

[0002] Reflection waveform inversion is a key technology in seismic data processing. It uses reflection wave data from deep underground to invert subsurface parameters. It has important applications in oil and gas exploration and deep geological structure research, providing strong support for high-precision subsurface modeling and reservoir prediction. The signal-to-noise ratio (SNR) of the data is a key factor affecting the accuracy of reflection waveform inversion. When noise is present in the observed data, it is reflected in the gradient calculated using the adjoint state method, causing artifacts in the inversion results or preventing convergence.

[0003] To improve the stability and accuracy of inversion, data denoising is necessary. Currently, conventional denoising methods used in the industry include fx-domain filtering, fk-domain filtering, and median filtering. These methods generally rely on manual parameter adjustment, making it difficult to balance denoising effectiveness with signal fidelity. This is particularly true in areas with overlapping noise and signal bands or complex structures. In the presence of strong energy noise in seismic data, it is impossible to generate high-quality data to meet the requirements of reflection waveform inversion. Summary of the Invention

[0004] In view of the above problems, the purpose of the present invention is to provide a reflection wave waveform inversion method, system, medium and equipment based on de-migration denoising to solve the problem that noise may affect the inversion result during the iterative process of reflection wave waveform inversion.

[0005] To achieve the above-mentioned purpose, in the first aspect, the technical solution adopted by the present invention is: a reflection wave waveform inversion method based on de-migration denoising, which comprises: s The offset domain common imaging point gather H is generated by extending the reverse time migration imaging condition. The angle domain common imaging point gather A is obtained by applying the local tilt stacking transform to the offset domain common imaging point gather H. The tilt filter F based on the similarity metric is calculated. spec , through the tilt filter F spec The new angle domain common imaging point gather A′ is obtained, and the new offset domain common imaging point gather H′ is obtained by local tilt stacking inverse transformation; the new offset domain common imaging point gather H′ is de-migrated based on the extended imaging condition to generate the de-migrated data d obs ; Based on the initial background velocity model v, establish the prediction data d pre and de-migration data d obsThe least squares objective function φ is used. In each iteration, a temporary disturbance model r is first established, and then the background velocity model v is updated using the established disturbance model r. The disturbance model r and the background velocity model v are updated alternately and iteratively until the iteration termination condition is met and the background velocity model v is output.

[0006] Furthermore, the observation record d s The offset domain common imaging point gather H is generated by extending the reverse time migration imaging conditions, including:

[0007] Solve the source wave field u by using the source wave field equation s , solve the detector wave field u by the detector wave field equation r ;

[0008] Through the source wave field u s and detector wave field u r , using the extended reverse time migration imaging condition, the offset domain common imaging point gather H is obtained.

[0009] Furthermore, the tilt filter F based on the similarity metric is calculated spec :Apply the local tilt stacking transformation to the offset domain common imaging point gather H to obtain the angle domain common imaging point gather A; calculate the similarity coefficient S based on the angle domain common imaging point gather A; and use the Gaussian weighting function W h The tilt filter F is calculated by the similarity coefficient S spec ,include:

[0010] Calculate the Gaussian weighting function W h , the underground offset h direction of the similarity coefficient S is converted to the Gaussian function W h Perform weighted summation to calculate the tilt filter F spec .

[0011] Furthermore, the new offset domain common imaging point gather H′ is de-migrated based on the extended imaging condition to generate de-migrated data d obs ,include:

[0012] Based on the demigration method under extended imaging conditions, the demigration wave field u is calculated. s ';

[0013] At the detector position x r Extract the record and get the de-shifted data d obs .

[0014] Furthermore, the disturbance model r and the background velocity model v are updated alternately and iteratively until the iteration termination condition is met and the background velocity model v is output, including:

[0015] Input the initial background velocity model v, set the number of iterations k, and establish the unconstrained augmented objective function φ;

[0016] Set the perturbation model r=0;

[0017] Update the disturbance model r and background velocity model v in sequence;

[0018] If the number of iterations is still less than k, the iteration is repeated until the termination condition is met, and the background velocity v is finally output.

[0019] Furthermore, the disturbance model r is updated, including:

[0020] Solve the background wave field equation, calculate the residual after amplitude matching of the predicted data and the observed data, solve the adjoint equation of the disturbance wave field, and calculate the gradient of the objective function φ with respect to the disturbance model r

[0021] According to the gradient Update the perturbation model r by the steepest descent method.

[0022] Furthermore, the background velocity model v is updated, including:

[0023] Solve the background wave field, disturbance wave field and corresponding adjoint wave field equations, and calculate the gradient of the objective function φ with respect to the background velocity model v

[0024] According to the gradient Update the background velocity model v by the steepest descent method.

[0025] In the second aspect, the technical solution adopted by the present invention is: a reflection wave waveform inversion system based on de-migration denoising, which includes: a first processing module, which processes the observation record d s The offset domain common imaging point gather H is generated by extending the reverse time migration imaging condition. The second processing module applies the local tilt stacking transform to the offset domain common imaging point gather H to obtain the angle domain common imaging point gather A, and calculates the tilt filter F based on the similarity metric. spec , through the tilt filter F spec The new angle domain common imaging point gather A′ is obtained, and the new offset domain common imaging point gather H′ is obtained by local tilt stacking inverse transformation. The third processing module performs demigration based on the extended imaging condition on the new offset domain common imaging point gather H′ to generate demigration data d obs ; Inversion output module, based on the initial background velocity model v, establishes the predicted data d pre and de-migration data d obs The least squares objective function φ is used. In each iteration, a temporary disturbance model r is first established, and then the background velocity model v is updated using the established disturbance model r. The disturbance model r and the background velocity model v are updated alternately and iteratively until the iteration termination condition is met and the background velocity model v is output.

[0026] In a third aspect, the technical solution adopted by the present invention is: a computer-readable storage medium storing one or more programs, wherein the one or more programs include instructions, and when the instructions are executed by a computing device, the computing device executes any one of the above methods.

[0027] In a fourth aspect, the technical solution adopted by the present invention is: a computing device, comprising: one or more processors, a memory and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include instructions for executing any of the above methods.

[0028] The present invention has the following advantages due to the adoption of the above technical solution:

[0029] This paper suppresses high-energy noise in seismic data by integrating demigration into reflection waveform inversion. This demigration method, based on extended imaging conditions, effectively removes various types of noise from seismic data by directly reconstructing the reflection signal. The demigrated data retains the same travel time and phase information as the original data, significantly mitigating the impact of noise on reflection waveform inversion and improving the accuracy of the inversion background velocity model. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 This is a flow chart of a method for inverting a reflected wave waveform based on de-migration and denoising according to an embodiment of the present invention;

[0031] Figure 2 1 is a diagram of a true velocity model and a migration and demigration velocity model provided by an embodiment of the present invention; wherein, Figure (a) is the true velocity model, which is used for forward simulation to obtain observed seismic records, and Figure (b) is the migration and demigration velocity model, which also serves as the initial velocity model for reflection wave waveform inversion;

[0032] Figure 3 Figure 1 is a comparison of different seismic data provided by an embodiment of the present invention, with all refracted waves removed from the records. Figure (a) shows the original noise-free data generated by forward simulation of the true velocity model with a shot point located at 6 km. Figure (b) shows the noise-free data in Figure (a) with random noise added, resulting in a signal-to-noise ratio of -35.1 dB. Figure (c) shows the denoised data after de-migration of the noisy data in Figure (b).

[0033] Figure 4 1 is a comparison of the results of reflection wave waveform inversion based on different seismic data provided by an embodiment of the present invention; wherein, Figure (a) is a model inverted using noisy data, Figure (b) is a model inverted using de-migration data, and Figure (c) is a model inverted using noise-free data;

[0034] Figure 5 3. The figures are comparative diagrams of reverse time migration results based on different background velocities provided by an embodiment of the present invention, wherein Figure (a) shows the imaging result of reverse time migration using the initial model; Figure (b) shows the imaging result of the model inverted using noisy data; Figure (c) shows the imaging result of the model inverted using reverse-migrated data; and Figure (d) shows the imaging result of the model inverted using noise-free data. DETAILED DESCRIPTION

[0035] To address the problem that existing technologies cannot generate high-quality data sufficient for reflection waveform inversion in scenarios where strong energy noise exists in seismic data, the present invention provides a reflection waveform inversion method, system, medium, and device based on demigration denoising. The method comprises: extending reverse time migration imaging conditions to generate offset-domain common imaging point gathers; obtaining formation dip information by transforming local tilt stacking into angle-domain common imaging point gathers, eliminating false imaging structures using a similarity-based dip filter, and restoring the data to offset-domain common imaging point gathers through inverse transformation; and reconstructing the reflection wave data using demigration for reflection waveform inversion. The method generates demigrated data with a high signal-to-noise ratio that has consistent travel time and phase information with the reflection waves in the original data, effectively alleviating the impact of noise on reflection waveform inversion and improving the accuracy of background velocity inversion.

[0036] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the described embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field are within the scope of protection of the present invention.

[0037] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0038] In one embodiment of the present invention, a method for inverting reflected wave waveforms based on de-migration denoising is provided. In this embodiment, Figure 1 As shown, the method includes the following steps:

[0039] 1) Observation record d s The offset domain common imaging point gather H is generated by extending the reverse time migration imaging condition;

[0040] 2) Apply the local tilt stacking transform to the offset domain common imaging point gather H to obtain the angle domain common imaging point gather A, and calculate the tilt filter F based on the similarity metric spec , through the tilt filter F spec The new angle domain common imaging point gather A' is obtained, and the new offset domain common imaging point gather H' is obtained by local tilt stacking and inverse transformation;

[0041] 3) The new offset domain common imaging point gather H′ is de-migrated based on the extended imaging condition to generate de-migrated data d obs ;

[0042] 4) Based on the initial background velocity model v, establish the prediction data d pre and de-migration data d obs The least squares objective function φ is used. In each iteration, a temporary disturbance model r is first established, and then the background velocity model v is updated using the established disturbance model r. The disturbance model r and the background velocity model v are updated alternately and iteratively until the iteration termination condition is met and the background velocity model v is output.

[0043] In the above step 1), the observation record d s The offset domain common imaging point gather H is generated by extending the reverse time migration imaging condition, including the following steps:

[0044] 1.1) Solve the source wave field u by using the source wave field equation s , solve the detector wave field u by the detector wave field equation r In this embodiment, the source wave field equation is:

[0045] Lu s (x,t;x s )=s(t;x s ) (1)

[0046] Where x = (x, y, z) represents the spatial position, t represents the current moment of wave field propagation, and x s represents the location of the earthquake source, L is the forward operator of the wave equation, and s represents the earthquake source.

[0047] In this embodiment, the detector wave field equation is:

[0048] L * u r (x,t;x s )=d s (x r ,t;x s ) (2)

[0049] Where x = (x, y, z) represents the spatial position, t represents the current moment of wave field propagation, and x s represents the location of the earthquake source, xr represents the detector position, d s Indicates observation record, L * Indicates the observation record d s By the detector position x r Reverse time propagation operator.

[0050] 1.2) Through the source wave field u s and detector wave field u r , using the extended reverse time migration imaging condition, the offset domain common imaging point gather H is obtained.

[0051] Among them, the offset domain common imaging point gather H is:

[0052]

[0053] Where x=(x,y,z) represents the spatial position, h=(h x ,h y ,h z ) represents the underground offset, t represents the current moment of wave field propagation, x s represents the location of the earthquake source, u s represents the source wave field, u r Represents the detector wavefield.

[0054] In step 2), the local tilt stacking transform is applied to the offset domain common imaging point gather H to obtain the angle domain common imaging point gather A, and the tilt filter F based on the similarity metric is calculated. spec , apply the tilt filter F spec The new angle domain common imaging point gather A′ is obtained, and the new offset domain common imaging point gather H′ is obtained by local tilt stacking inverse transformation. Specifically, the following steps are included:

[0055] 2.1) Calculate the angle domain common imaging point gather A and apply the local tilt stack transform to the offset domain common imaging point gather H:

[0056]

[0057] Where x and z represent spatial positions, ν represents the decomposition angle, q = tanν represents the integration path in the zx plane, h represents the subsurface offset, x′ represents the local window along the x direction of the tilt stack transformation, and Δx represents the size of the local window.

[0058] 2.2) Calculate the similarity coefficient S based on the angle domain common imaging point gather A. The similarity coefficient S is expressed as:

[0059]

[0060] Where x and z represent spatial positions, ν represents the decomposition angle, h represents the underground offset, and N z and N ν are represented as the half-window size in the depth direction and the dip direction respectively;

[0061] 2.3) Through Gaussian weighting function W h The tilt filter F is calculated by the similarity coefficient S spec , the specific steps are as follows:

[0062] 2.3.1) Calculate the Gaussian weighting function W h , the expression is:

[0063]

[0064] Where h represents the underground offset, σ 2 is the variance of the Gaussian weighting function;

[0065] 2.3.2) The similarity coefficient S is converted to the underground offset h direction by the Gaussian function W h Perform weighted summation to calculate the tilt filter F spec :

[0066]

[0067] Where x and z represent spatial positions, ν represents the decomposition angle, and h represents the underground offset.

[0068] 2.4) Apply the dip filter F to the angle domain common imaging point gather A spec , and obtain the new angle domain common imaging point gather A′:

[0069] A′(x,z,ν,h)=F spec ·A(x,z,ν,h) (8)

[0070] Where x and z represent spatial positions, ν represents the decomposition angle, and h represents the underground offset;

[0071] 2.5) Apply the local tilt stack inverse transform to the new angle-domain common imaging point gather A′ to obtain the corresponding new offset-domain common imaging point gather H′:

[0072]

[0073] Where x and z represent spatial positions, h represents the underground offset, and ρ(z) represents the |k in the corresponding wavenumber domain. z |filter, * indicates convolution in the spatial domain, q = tanν represents the integral path in the zx plane, ν represents the decomposition angle, x′ represents the local window along the x direction of the oblique stacking transformation, and Δx represents the local window size.

[0074] In step 3) above, the new offset domain common imaging point gather H′ is de-migrated based on the extended imaging condition to generate de-migrated data d obs , including the following steps:

[0075] 3.1) Based on the extended imaging condition, the demigration wave field u is calculated. s ′:

[0076] Lu s ′(x,t;x s )=∫H(xh,h)u s (x-2h,t;x s )dh (10)

[0077] Where x=(x,y,z) represents the spatial position, h=(h x ,h y ,h z ) represents the underground offset, t represents the current moment of wave field propagation, x s represents the location of the earthquake source, L represents the wave equation forward operator, H represents the offset domain common imaging point gather, u s Represents the source wavefield.

[0078] 3.2) At the detector position x r Extract the record and get the de-shifted data d obs :

[0079] d obs (x r ,t;x s )=p(x,x r ;x s )u s ′(x,t;x s ) (11)

[0080] Where x r represents the detector position, x=(x,y,z) represents the spatial position, t represents the current time of wave field propagation, x s represents the location of the earthquake source, p represents the detector extraction operator, u s ′ represents the demigrated wave field.

[0081] In the above step 4), the alternating iterative updating of the disturbance model r and the background velocity model v until the iteration termination condition is satisfied and the background velocity model v is outputted includes the following steps:

[0082] 4.1) Input the initial background velocity model v, set the number of iterations k, and establish the unconstrained augmented objective function φ:

[0083]

[0084] Where v represents the background velocity model, r represents the disturbance model, and u b represents the background wave field, u p represents the perturbation wave field, and represents the corresponding accompanying wave field, x=(x,y,z) represents the spatial position, x r represents the detector position, x s represents the location of the earthquake source, t represents the current moment of wave field propagation, p represents the detector extraction operator, d obs represents the observation data generated by demigration, <,> represents the inner product, L represents the wave equation forward operator, s represents the source wavelet, ü b Represents the background wave field u b Second-order partial derivative in the time direction.

[0085] 4.2) Set the perturbation model r = 0;

[0086] 4.3) Update the disturbance model r and background velocity model v in sequence;

[0087] 4.4) If the number of iterations is still less than k, repeat steps 4.2) to 4.3) until the termination condition is met, and finally output the background velocity v.

[0088] In this embodiment, updating the disturbance model r in step 4.3) includes the following steps:

[0089] 4.3.1.1) Solve the background wave field equation Lu b (x,t;x s )=s(t;x s ), calculate the residual after amplitude matching of the predicted data and the observed data, and solve the adjoint equation of the disturbance wave field Calculate the gradient of the objective function φ with respect to the perturbation model r

[0090] gradient The expression is:

[0091]

[0092] Where x = (x, y, z) represents the spatial position, t represents the current moment of wave field propagation, and x s Indicates the location of the earthquake source, ü b Represents the background wave field u b The second-order partial derivative in the time direction, represents the disturbance wave field u p The accompanying wave field;

[0093] 4.3.1.2) According to the gradient Update the perturbation model r by the steepest descent method.

[0094] In this embodiment, updating the background velocity model v in step 4.3) includes the following steps:

[0095] 4.3.2.1) Solve the background wave field, disturbance wave field and corresponding adjoint wave field equations, and calculate the gradient of the objective function φ with respect to the background velocity model v

[0096] gradient The expression is:

[0097]

[0098] Where x = (x, y, z) represents the spatial position, t represents the current moment of wave field propagation, and x s represents the location of the earthquake source, L represents the wave equation forward operator, u b represents the background wave field, u p represents the perturbation wave field, and represents the corresponding companion wave field;

[0099] 4.3.2.2) According to the gradient Update the background velocity model v by the steepest descent method.

[0100] The following is a detailed description of the reflected wave waveform inversion method based on de-migration denoising of the present invention through specific embodiments. Figure 2 As shown by Figure 2 As shown in (a), the true velocity model shows a flat stratum, with multiple faults penetrating the shallow strata and well-developed deep fractures and caverns. The accuracy of the background velocity modeling affects the focusing of the migration imaging on the fractures and caverns. A seismic source and receivers were placed on the surface. The source used a Ricker wavelet with a dominant frequency of 15 Hz. The excitation range was from 0 km to 12 km, with a shot spacing of 100 m, for a total of 121 shots. An observation system with center-shot and two-side receivers was used, with a maximum offset of 6.01 km, a grid spacing of 10 m, and a trace spacing of 10 m. Absorbing boundary conditions were used on the surface. Figure 2 Middle (b) is the background velocity model used for migration and demigration.

[0101] like Figure 3 As shown, Figure 3 The original noise-free record with the shot point at 6 km is shown in (a). Figure 3 (b) is the noisy data obtained by adding random noise to the noise-free data. The signal-to-noise ratio is -35.1dB, and the effective signal is basically masked by the noise. Figure 3(c) in the middle shows the de-migrated data obtained by de-migrating the noisy data, which successfully restores the reflected wave signal in the original data.

[0102] like Figure 4 As shown, Figure 4 (a) is the result of inversion based on noisy data. The strong energy noise leads to erroneous updates. Figure 4 (b) is the result of inversion based on the de-migration data. Figure 4 The result of inversion using noise-free data in (c) is closer and restores the background variation trend of the low wavenumber of the real model.

[0103] like Figure 5 As shown, Figure 5 As shown in (a), the initial velocity model still has errors, and the imaging results have poor focusing on low-velocity bodies and fracture-cavity bodies; Figure 5 (b) shows the background velocity model imaging results of the inversion of noisy data. There is a large amount of arcing noise in the migration profile, and the imaging of deep fractures and caves shows a scattered phenomenon. Figure 5 (c) The imaging results of the inversion model of the back-migration data are focused and restored at the low-velocity body and the fracture-cavity body, which is consistent with the Figure 5 The imaging results of the noise-free data inversion model in (d) are very close. The present invention can effectively alleviate the influence of noise data on the reflection wave waveform inversion and generate an accurate background velocity model.

[0104] In one embodiment of the present invention, a reflection waveform inversion system based on demigration denoising is provided, comprising:

[0105] The first processing module records the observation data. s The offset domain common imaging point gather H is generated by extending the reverse time migration imaging condition;

[0106] The second processing module applies the local tilt stacking transform to the offset domain common imaging point gather H to obtain the angle domain common imaging point gather A, and calculates the tilt filter F based on the similarity metric. spec , through the tilt filter F spec The new angle domain common imaging point gather A' is obtained, and the new offset domain common imaging point gather H' is obtained by local tilt stacking and inverse transformation;

[0107] The third processing module generates the de-migration data d by de-migration based on the extended imaging condition for the new offset domain common imaging point gather H′. obs ;

[0108] The inversion output module builds the prediction data d based on the initial background velocity model v pre and de-migration data d obsThe least squares objective function φ is used. In each iteration, a temporary disturbance model r is first established, and then the background velocity model v is updated using the established disturbance model r. The disturbance model r and the background velocity model v are updated alternately and iteratively until the iteration termination condition is met and the background velocity model v is output.

[0109] In the above embodiment, the observation record d s The offset domain common imaging point gather H is generated by extending the reverse time migration imaging conditions, including:

[0110] Solve the source wave field u by using the source wave field equation s , solve the detector wave field u by the detector wave field equation r ;

[0111] Through the source wave field u s and detector wave field u r , using the extended reverse time migration imaging condition, the offset domain common imaging point gather H is obtained.

[0112] In the above embodiment, the tilt filter F based on the similarity metric is calculated. spec :Apply the local tilt stacking transformation to the offset domain common imaging point gather H to obtain the angle domain common imaging point gather A; calculate the similarity coefficient S based on the angle domain common imaging point gather A; and use the Gaussian weighting function W h The tilt filter F is calculated by the similarity coefficient S spec ,include:

[0113] Calculate the Gaussian weighting function W h , the underground offset h direction of the similarity coefficient S is converted to the Gaussian function W h Perform weighted summation to calculate the tilt filter F spec .

[0114] In the above embodiment, the new offset-domain common imaging point gather H′ is subjected to demigration based on the extended imaging condition to generate demigration data d obs ,include:

[0115] Based on the demigration method under extended imaging conditions, the demigration wave field u is calculated. s ';

[0116] At the detector position x r Extract the record and get the de-shifted data d obs .

[0117] In the above embodiment, the alternating iterative updating of the disturbance model r and the background velocity model v until the iteration termination condition is satisfied and the background velocity model v is outputted includes:

[0118] Input the initial background velocity model v, set the number of iterations k, and establish the unconstrained augmented objective function φ;

[0119] Set the perturbation model r=0;

[0120] Update the disturbance model r and background velocity model v in sequence;

[0121] If the number of iterations is still less than k, the iteration is repeated until the termination condition is met, and the background velocity v is finally output.

[0122] In this embodiment, updating the disturbance model r includes:

[0123] Solve the background wave field equation, calculate the residual after amplitude matching of the predicted data and the observed data, solve the adjoint equation of the disturbance wave field, and calculate the gradient of the objective function φ with respect to the disturbance model r

[0124] According to the gradient Update the perturbation model r by the steepest descent method.

[0125] In this embodiment, updating the background velocity model v includes:

[0126] Solve the background wave field, disturbance wave field and corresponding adjoint wave field equations, and calculate the gradient of the objective function φ with respect to the background velocity model v

[0127] According to the gradient Update the background velocity model v by the steepest descent method.

[0128] The system provided in this embodiment is used to execute the above-mentioned method embodiments. Please refer to the above-mentioned embodiments for specific processes and detailed contents, which will not be repeated here.

[0129] In one embodiment of the present invention, a computing device is provided. The computing device may be a terminal and may include: a processor, a communications interface, a memory, a display screen, and an input device. The processor, communications interface, and memory communicate with each other via a communications bus. The processor is configured to provide computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and a computer program. When executed by the processor, the computer program implements the methods described in the above embodiments. The internal memory provides an environment for the operating system and computer program in the non-volatile storage medium to run. The communications interface is configured to communicate with an external terminal via wired or wireless communication. The wireless communication may be achieved via Wi-Fi, a network management service provider, NFC (near field communication), or other technologies. The display screen may be a liquid crystal display or an electronic ink display. The input device may be a touchscreen layer covering the display screen, or may be buttons, a trackball, or a touchpad provided on the computing device housing, or may be an external keyboard, touchpad, or mouse. The processor may invoke logic instructions stored in the memory.

[0130] In addition, the logical 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 sold or used as an independent product. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0131] In one embodiment of the present invention, a computer program product is provided, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the methods provided by the above-mentioned method embodiments.

[0132] In one embodiment of the present invention, a non-transitory computer-readable storage medium is provided. The non-transitory computer-readable storage medium stores server instructions. The computer instructions enable a computer to execute the methods provided in the above embodiments.

[0133] The above embodiment provides a computer-readable storage medium, whose implementation principle and technical effects are similar to those of the above method embodiment, and will not be repeated here.

[0134] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, 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 produce 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 flowcharts and / or block diagrams. 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.

[0135] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0136] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0137] 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 various embodiments of the present invention.

Claims

1. A reflection wave waveform inversion method based on de-migration denoising, characterized in that: It includes: Observation record d s The offset domain common imaging point gather H is generated by extending the reverse time migration imaging condition; Apply the local tilt stacking transform to the offset domain common imaging point gather H to obtain the angle domain common imaging point gather A, and calculate the tilt filter F based on the similarity measure. spec , through the tilt filter F spec The new angle domain common imaging point gather A' is obtained, and the new offset domain common imaging point gather H' is obtained by local tilt stacking and inverse transformation; The new offset domain common imaging point gather H′ is de-migrated based on the extended imaging condition to generate de-migrated data d obs ; Based on the initial background velocity model v, establish the prediction data d pre and de-migration data d obs The least squares objective function φ is used. In each iteration, a temporary disturbance model r is first established, and then the background velocity model v is updated using the established disturbance model r. The disturbance model r and the background velocity model v are updated alternately and iteratively until the iteration termination condition is met and the background velocity model v is output.

2. The reflected wave waveform inversion method based on de-migration denoising according to claim 1, characterized in that: Observation record d s The offset domain common imaging point gather H is generated by extending the reverse time migration imaging conditions, including: Solve the source wave field u by using the source wave field equation s , solve the detector wave field u by the detector wave field equation r ; Through the source wave field u s and detector wave field u r , using the extended reverse time migration imaging condition, the offset domain common imaging point gather H is obtained.

3. The reflected wave waveform inversion method based on de-migration denoising according to claim 1, characterized in that: Calculate the tilt filter F based on the similarity measure spec :Apply the local tilt stacking transformation to the offset domain common imaging point gather H to obtain the angle domain common imaging point gather A; calculate the similarity coefficient S based on the angle domain common imaging point gather A; and use the Gaussian weighting function W h The tilt filter F is calculated by the similarity coefficient S spec ,include: Calculate the Gaussian weighting function W h , the underground offset h direction of the similarity coefficient S is converted to the Gaussian function W h Perform weighted summation to calculate the tilt filter F spec .

4. The method for reflecting wave waveform inversion based on de-migration denoising according to claim 1, wherein: The new offset domain common imaging point gather H′ is de-migrated based on the extended imaging condition to generate de-migrated data d obs ,include: Based on the demigration method under extended imaging conditions, the demigration wave field u is calculated. s '; At the detector position x r Extract the record and get the de-shifted data d obs .

5. The reflected wave waveform inversion method based on de-migration denoising according to claim 1, characterized in that: The disturbance model r and the background velocity model v are updated alternately and iteratively until the iteration termination condition is met and the background velocity model v is output, including: Input the initial background velocity model v, set the number of iterations k, and establish the unconstrained augmented objective function φ; Set the perturbation model r=0; Update the disturbance model r and background velocity model v in sequence; If the number of iterations is still less than k, the iteration is repeated until the termination condition is met, and the background velocity v is finally output.

6. The reflected wave waveform inversion method based on de-migration denoising according to claim 5, characterized in that: Update the perturbation model r, including: Solve the background wave field equation, calculate the residual after amplitude matching of the predicted data and the observed data, solve the adjoint equation of the disturbance wave field, and calculate the gradient of the objective function φ with respect to the disturbance model r According to the gradient Update the perturbation model r by the steepest descent method.

7. The reflected wave waveform inversion method based on de-migration denoising according to claim 5, characterized in that: Update the background velocity model v, including: Solve the background wave field, disturbance wave field and corresponding adjoint wave field equations, and calculate the gradient of the objective function φ with respect to the background velocity model v According to the gradient Update the background velocity model v by the steepest descent method.

8. A reflection wave waveform inversion system based on de-migration denoising, characterized in that: It includes: The first processing module records the observation data. s The offset domain common imaging point gather H is generated by extending the reverse time migration imaging condition; The second processing module applies the local tilt stacking transform to the offset domain common imaging point gather H to obtain the angle domain common imaging point gather A, and calculates the tilt filter F based on the similarity metric. spec , through the tilt filter F spec The new angle domain common imaging point gather A' is obtained, and the new offset domain common imaging point gather H' is obtained by local tilt stacking and inverse transformation; The third processing module generates the de-migration data d by de-migration based on the extended imaging condition for the new offset domain common imaging point gather H′. obs ; The inversion output module builds the prediction data d based on the initial background velocity model v pre and de-migration data d obs The least squares objective function φ is used. In each iteration, a temporary disturbance model r is first established, and then the background velocity model v is updated using the established disturbance model r. The disturbance model r and the background velocity model v are updated alternately and iteratively until the iteration termination condition is met and the background velocity model v is output.

9. A computer-readable storage medium storing one or more programs, characterized in that: The one or more programs include instructions that, when executed by a computing device, cause the computing device to perform any one of the methods of claims 1 to 7 .

10. A computing device, characterized in that include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include instructions for executing any one of the methods according to claims 1 to 7.

Images