Intelligent first break tomographic statics method, device, equipment, medium and program

Abstract

This disclosure relates to an intelligent first-arrival tomography static correction method, apparatus, device, storage medium, and computer program. The method includes: acquiring first-arrival data, observation system files, and grid files; setting network training parameters, sampling interval, and maximum depth; performing intelligent inversion based on the first-arrival data, network training parameters, sampling interval, and maximum depth to obtain a near-surface velocity model; using the near-surface velocity model to pick a low-velocity zone bottom interface according to a preset stripping layer velocity; smoothing the low-velocity zone bottom interface according to preset smoothing parameters to obtain a smoothed low-velocity zone bottom interface; calculating a static correction amount based on the smoothed low-velocity zone bottom interface and the near-surface velocity; and using the static correction amount to perform time difference correction on the seismic data to be processed to obtain corrected data.

Description

Technical Field

[0001] This disclosure relates to the field of geophysical exploration technology, and in particular to an intelligent first-arrival tomographic static correction method, apparatus, equipment, storage medium, and computer program. Background Technology

[0002] In seismic exploration, the reflection time-distance curve of seismic waves is no longer a hyperbolic distribution due to factors such as topographic relief, varying source well depths, and the thickness and velocity variations of low-velocity zones. However, some important steps in seismic data processing (such as velocity analysis and dynamic correction) are performed under the premise that the reflection time-distance curve is hyperbolic, and surface undulations affect the accuracy of subsurface structure imaging. Therefore, static correction techniques are needed to correct the reflection time, ensuring that the shot point and receiver point are located on the same reference plane. Static correction methods can be divided into two main categories: primary static correction and residual static correction. Primary static correction is used to eliminate long-wavelength time difference components caused by lateral variations in elevation and low-velocity zones. Commonly used primary static correction methods include elevation static correction, model static correction, refraction static correction, and tomographic static correction. The time difference estimation caused by low-velocity zone variations requires the use of a near-surface velocity model; therefore, the main difference between these methods lies in the way the near-surface velocity model is established.

[0003] First-arrival travel-time tomography can be used to establish near-surface velocity structures, thereby obtaining long-wavelength static correction values. Therefore, the effectiveness of tomographic static correction largely depends on the accuracy of the first-arrival travel-time tomography inversion. First-arrival tomography inversion uses the first-arrival travel time and path to invert the velocity structure of the subsurface medium, avoiding the assumption of a layered velocity structure and better addressing static correction problems caused by topographic and low-velocity zone variations in the work area. However, conventional first-arrival travel-time tomography increases computational costs when dealing with undulating surfaces, and in cases of drastic topographic changes, it can easily lead to unstable calculation results. Summary of the Invention

[0004] This disclosure provides an intelligent first-arrival tomographic static correction method, apparatus, device, storage medium, and computer program to solve the problem of high computational cost of existing first-arrival tomographic static correction methods under complex surface conditions.

[0005] In a first aspect, this disclosure provides an intelligent first-arrival chromatography static correction method, including:

[0006] Acquire initial arrival data, observation system files, and grid files; set network training parameters, sampling interval, and maximum depth.

[0007] Based on the initial arrival data, the network training parameters, the sampling interval, and the maximum depth, an intelligent inversion is performed to obtain a near-surface velocity model;

[0008] The low-velocity zone bottom interface is picked up using the near-surface velocity model according to the preset peeling layer velocity, and the low-velocity zone bottom interface is smoothed according to the preset smoothing parameters to obtain a smooth low-velocity zone bottom interface.

[0009] Calculate the static correction amount based on the smooth low-speed zone bottom interface and the near-surface velocity;

[0010] The static correction value is used to perform time difference correction on the seismic data to be processed, and the corrected data is obtained.

[0011] In some embodiments, setting the network training parameters, sampling interval, and maximum depth includes:

[0012] Set the learning rate, batch size, and number of iterations to obtain the network training parameters;

[0013] The sampling interval and maximum depth are set according to the user's input parameters.

[0014] In some embodiments, the step of intelligently inverting the near-surface velocity model based on the initial arrival data, the network training parameters, the sampling interval, and the maximum depth includes:

[0015] The neural network model outputs the model results based on the grid file.

[0016] Calculate the error between the network model output and the initial arrival data, and update the network model weights based on the error using the minimum loss function;

[0017] Once the number of iterations specified in the network training parameters is reached, the model training is confirmed to be complete, and a near-surface velocity model is obtained. In some embodiments, smoothing the low-velocity band bottom interface according to preset smoothing parameters to obtain a smoothed low-velocity band bottom interface includes:

[0018] Set the smoothing window according to the smoothing parameters;

[0019] The smooth window is used to slide sequentially on the low-speed band bottom interface;

[0020] Calculate the Gaussian weighted value of the data within the smoothing window after each slide, and replace the original data in the center of the smoothing window with the calculated Gaussian weighted value.

[0021] After all data has been replaced, the smoothing process is confirmed to be complete, resulting in the smoothed low-speed bottom interface.

[0022] In some embodiments, calculating the static correction based on the smooth low-speed zone bottom interface and the near-surface velocity includes:

[0023] The static correction amount is calculated using the following formula:

[0024]

[0025] Where t is the static correction amount, M is the number of low-speed band layers at the smooth low-speed band bottom interface, and h i and v i E represents the thickness and velocity of the smooth low-speed band bottom interface, respectively. d E is the pre-acquired elevation of the receiver point. g For the elevation of the pre-obtained reference surface, v c This is the preset replacement speed.

[0026] In some embodiments, the step of using the static correction amount to perform time difference correction on the seismic data to be processed to obtain corrected data includes:

[0027] The corrected data is obtained by shifting the seismic data by time using the static correction amount.

[0028] Secondly, this disclosure provides an intelligent first-arrival chromatography static correction device, comprising:

[0029] The parameter setting module is used to acquire initial arrival data, observation system files, and grid files, and to set network training parameters, sampling intervals, and maximum depth.

[0030] The model training module is used to perform intelligent inversion based on the initial arrival data, the network training parameters, the sampling interval, and the maximum depth to obtain a near-surface velocity model;

[0031] The low-velocity zone picking module is used to pick up the low-velocity zone bottom interface according to the preset peeling layer velocity using the near-surface velocity model, and to smooth the low-velocity zone bottom interface according to the preset smoothing parameters to obtain a smooth low-velocity zone bottom interface.

[0032] The static correction calculation module is used to calculate the static correction based on the smooth low-speed zone bottom interface and the near-surface velocity.

[0033] The data correction module is used to perform time difference correction on the seismic data to be processed using the static correction amount to obtain corrected data.

[0034] Thirdly, this disclosure provides a computer device including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method described in the foregoing aspects.

[0035] Fourthly, this disclosure provides a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the steps of the methods described in the above aspects.

[0036] Fifthly, this disclosure provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the methods described in the above aspects.

[0037] This disclosure provides an intelligent first-arrival tomography static correction method, apparatus, device, storage medium, and computer program. It acquires first-arrival data, observation system files, and grid files; sets network training parameters, sampling intervals, and maximum depth; performs intelligent inversion based on the first-arrival data, network training parameters, sampling intervals, and maximum depth to obtain a near-surface velocity model; uses the near-surface velocity model to pick the low-velocity zone floor interface according to a preset stripping layer velocity; smooths the low-velocity zone floor interface according to preset smoothing parameters to obtain a smoothed low-velocity zone floor interface; calculates a static correction amount based on the smoothed low-velocity zone floor interface and the near-surface velocity; and uses the static correction amount to perform time difference correction on the seismic data to be processed to obtain corrected data. This method can flexibly adapt to surface conditions with varying degrees of undulation, improving universality. By using a neural network model to establish the near-surface velocity model, it improves model establishment efficiency, reduces correction costs, and increases correction accuracy.

[0038] The present disclosure will be described in more detail below based on embodiments and with reference to the accompanying drawings:

[0039] Figure 1 A flowchart illustrating an intelligent first-arrival chromatography static correction method provided in this embodiment of the disclosure;

[0040] Figure 2 A schematic diagram of a process for training a near-surface velocity model provided in an embodiment of this disclosure;

[0041] Figure 3 This is a functional block diagram of an intelligent first-arrival chromatography static correction device provided in an embodiment of the present disclosure.

[0042] In the accompanying drawings, the same parts are referred to by the same reference numerals, and the drawings are not drawn to scale. Detailed Implementation

[0043] To enable those skilled in the art to better understand the technical solutions of this disclosure, and to fully understand and implement the process of how this disclosure applies technical means to solve technical problems and achieve corresponding technical effects, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, not all embodiments. The embodiments of this disclosure and the various features within them can be combined with each other without conflict, and the resulting technical solutions are all within the protection scope of this disclosure. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure without creative effort should fall within the protection scope of this disclosure.

[0044] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0045] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0046] Example 1

[0047] Figure 1 This is a schematic flowchart illustrating an intelligent first-arrival chromatography static correction method provided in an embodiment of this disclosure. Figure 1 As shown, an intelligent first-arrival chromatography static correction method includes:

[0048] S1. Set the network training parameters, sampling interval, and maximum depth.

[0049] In this embodiment of the invention, the first arrival data refers to the recorded data of the first arrival of seismic waves from the epicenter to the seismic detector during seismic exploration. This data is crucial for seismic wave velocity analysis and imaging of subsurface structures. For example, in seismic tomography, since the propagation time of seismic waves is directly related to the velocity of the medium, the first arrival data is used to determine the velocity distribution of the subsurface medium.

[0050] In this embodiment of the invention, the observation system file mainly records the parameters and settings for seismic data acquisition, including but not limited to shot point layout, total number of shot points, shot firing method, detector arrangement and length, receiver point spacing, receiver line spacing, CMP (Common Mid Point) spacing, coverage times, and azimuth.

[0051] In this embodiment of the invention, the mesh file includes mesh resolution, number of mesh points, and mesh coverage area.

[0052] In this embodiment of the invention, setting the network training parameters, sampling interval, and maximum depth includes:

[0053] Set the learning rate, batch size, and number of iterations to obtain the network training parameters;

[0054] The sampling interval and maximum depth are set according to the user's input parameters.

[0055] In detail, the setting of learning rate, batch size, and number of iterations is based on the size of the grid resolution and the number of grid points contained in the grid file. The learning rate is set according to the size of the grid resolution and the number of grid points contained in the grid file. The number of iterations can be set according to the training accuracy requirements.

[0056] In detail, the learning rate is a crucial hyperparameter in a neural network model, affecting both the speed and stability of model training. The magnitude of the learning rate directly impacts the model's learning speed. A larger learning rate can accelerate learning but may also lead to error explosion and oscillations. A smaller learning rate can prevent overfitting and improve convergence speed but may result in excessively slow learning. Learning rate reduction can be achieved through epoch-based reduction, exponential reduction, and fractional reduction. Ethoch-based reduction halves the learning rate every N epochs; exponential reduction increases the learning rate exponentially with the number of training epochs; fractional reduction controls the magnitude of the reduction and is related to the number of iterations.

[0057] In detail, batch size is an important hyperparameter in the deep learning training process. It refers to the number of samples used to update the model weights simultaneously in one iteration. The choice of batch size has a significant impact on the model's training efficiency, memory consumption, training stability, and final performance. A larger batch size can improve memory utilization and computation speed, while a smaller batch size may require more iterations to process the entire training set but can provide more frequent model updates.

[0058] In detail, setting the maximum depth according to the grid file and the observation system file means setting the maximum depth according to the grid coverage contained in the grid file and the shot point layout, CMP spacing and azimuth angle contained in the observation system file.

[0059] In detail, setting the sampling interval and maximum depth according to the user's input parameters means obtaining the numerical data input by the user, and then setting the sampling interval and maximum depth according to the numerical data.

[0060] S2. Based on the initial arrival data, the network training parameters, the sampling interval, and the maximum depth, an intelligent inversion is performed to obtain the near-surface velocity model.

[0061] In this embodiment of the invention, the intelligent inversion based on the initial arrival data, the network training parameters, the sampling interval, and the maximum depth to obtain the near-surface velocity model is based on PINNtomo to obtain the near-surface velocity model.

[0062] In detail, PINNtomo is a seismic tomography method based on Physics-Informed Neural Networks (PINNs). This method utilizes the powerful approximation capabilities of neural networks to solve the seismic travel time tomography problem, that is, to construct a velocity model by solving the equation of motion. The equation of motion is a nonlinear partial differential equation that appears in the wave propagation problem.

[0063] In detail, the physical information neural network is a deep learning technique that embeds physical laws (such as partial differential equations PDEs and ordinary differential equations ODEs) into the loss function of the neural network, so that the network learning process not only conforms to the data distribution, but also follows physical laws.

[0064] In detail, the PINNtomo method improves the accuracy and efficiency of imaging by combining physical laws in seismology with a data-driven approach. Its advantage lies in its ability to utilize partially observed seismic travel times from computational models covering seismic station coverage. This allows for the approximation of travel time factors and velocity fields through training neural networks, while being constrained by a physical information regularizer formed by equations. This method not only better compensates for the ill-posedness of tomographic imaging problems.

[0065] In this embodiment of the invention, reference is made to Figure 2 The diagram illustrates a process for training a near-surface velocity model according to an embodiment of the present invention. The step of intelligently inverting the near-surface velocity model based on the initial arrival data, the network training parameters, the sampling interval, and the maximum depth includes:

[0066] S21. Utilize the neural network model to output the model output results based on the grid file;

[0067] S22. Calculate the error between the output of the network model and the initial arrival data, and update the weights of the network model based on the error using the minimum loss function;

[0068] S23. When the number of iterations in the network training parameters is reached, the model training is confirmed to be complete, and the near-surface velocity model is obtained.

[0069] In detail, the weight parameters are key components of a neural network, determining the network's ability to extract features and learn patterns from input data. The weight parameters are adjusted through a training process (typically backpropagation and gradient descent) to minimize the loss function, enabling the model to make accurate predictions on unseen data.

[0070] In this embodiment of the invention, a near-surface velocity model is obtained by intelligent inversion based on the initial arrival data, the network training parameters, the sampling interval, and the maximum depth, thereby improving the efficiency of establishing the near-surface velocity model.

[0071] S3. Using the near-surface velocity model, the low-velocity zone bottom interface is picked up according to the preset peeling layer velocity, and the low-velocity zone bottom interface is smoothed according to the preset smoothing parameters to obtain a smooth low-velocity zone bottom interface.

[0072] In this embodiment of the invention, the step of using the near-surface velocity model to pick the bottom interface of the low-velocity zone based on the preset stripping layer velocity refers to identifying and determining the depth boundary of the low-velocity zone (usually referring to a geological stratum with a slower velocity) by comparing the velocity in the near-surface velocity model with the stripping layer velocity during the seismic data processing and interpretation process.

[0073] In detail, low-velocity zones typically refer to geological strata with slower velocities, which can affect the propagation path and velocity of seismic waves. The existence of these low-velocity zones can be identified and their depth boundaries determined through in-depth analysis of near-surface velocity models. This step is crucial for subsequent seismic data processing and interpretation, as the presence of low-velocity zones can affect seismic wave propagation, thereby impacting imaging quality and the accuracy of interpretation.

[0074] In this embodiment of the invention, the stripping layer refers to the influence of surface or shallow velocity changes that needs to be removed from the data during seismic data processing in order to more clearly identify the deep geological structure.

[0075] In detail, the preset stripping layer velocity refers to an expected estimate of the velocity at this layer, which can be based on geological models, empirical data, or the results of previous seismic data processing.

[0076] In this embodiment of the invention, the preset smoothing parameter is a parameter used to control the smoothing accuracy. It can be set according to requirements. An excessively large smoothing parameter can improve the efficiency of smoothing processing, but will cause the data to lose more details; a smaller smoothing parameter can retain more details, but will reduce the efficiency of smoothing processing.

[0077] In this embodiment of the invention, the step of smoothing the low-speed band bottom interface according to preset smoothing parameters to obtain a smooth low-speed band bottom interface includes:

[0078] Set the smoothing window according to the smoothing parameters;

[0079] The smooth window is used to slide sequentially on the low-speed band bottom interface;

[0080] Calculate the Gaussian weighted value of the data within the smoothing window after each slide, and replace the original data in the center of the smoothing window with the calculated Gaussian weighted value.

[0081] After all data has been replaced, the smoothing process is confirmed to be complete, resulting in the smoothed low-speed bottom interface.

[0082] S4. Calculate the static correction amount based on the smooth low-speed zone bottom interface and the near-surface velocity.

[0083] In the embodiments of the invention, the static correction amount is a correction parameter used in seismic data processing to eliminate the influence of factors such as changes in surface elevation, changes in the thickness and velocity of the weathering layer, and changes in the depth of the excitation and reception points on the propagation time of reflected waves.

[0084] In this embodiment of the invention, the step of calculating the static correction amount based on the smooth low-speed zone bottom interface and the near-surface velocity includes:

[0085] The static correction amount is calculated using the following formula:

[0086]

[0087] Where t is the static correction amount, M is the number of low-speed band layers at the smooth low-speed band bottom interface, and h i and v i E represents the thickness and velocity of the smooth low-speed band bottom interface, respectively. d E is the pre-acquired elevation of the receiver point. g For the elevation of the pre-obtained reference surface, v c This is the preset replacement speed.

[0088] In this embodiment, the elevation of the detector point refers to the vertical distance of the detector point relative to a preset reference plane.

[0089] In this embodiment of the invention, the elevation of the reference surface refers to a reference surface used to measure other elevations, typically sea level or a specific geographic reference surface. The reference surface elevation provides a unified reference framework for seismic data, enabling the comparison and processing of elevation data from different locations.

[0090] S5. Use the static correction amount to perform time difference correction on the seismic data to be processed to obtain corrected data.

[0091] In this embodiment of the invention, the step of using the static correction amount to perform time difference correction on the seismic data to be processed to obtain corrected data includes:

[0092] The corrected data is obtained by shifting the seismic data by time using the static correction amount.

[0093] In this embodiment of the invention, by using the static correction amount to perform time difference correction on the seismic data to be processed, the corrected data is obtained, which improves the efficiency and accuracy of first-arrival tomographic static correction.

[0094] In this embodiment of the invention, a near-surface velocity model is obtained through intelligent inversion based on pre-acquired first-arrival data, observation system files, and grid files. A low-velocity bottom interface is then identified based on the near-surface velocity model. A static correction is calculated based on a preset replacement velocity and the low-velocity bottom interface. Finally, the static correction is used to correct the seismic data for time difference, resulting in corrected data. This approach can flexibly adapt to surface conditions with varying degrees of undulation, improving universality. By utilizing a neural network model to establish the near-surface velocity model, model establishment efficiency is improved, correction costs are reduced, and correction accuracy is enhanced.

[0095] Example 2

[0096] Based on the above embodiments, this embodiment provides an application example.

[0097] This invention addresses the problem that conventional first-arrival tomography static correction techniques are ineffective in complex surface conditions, especially when the terrain is undulating. It proposes an intelligent first-arrival tomography static correction method based on PINNtomo. By leveraging the flexibility and efficiency of PINNtomo, a near-surface velocity model is established for complex surface areas. Then, the low-velocity zone bottom interface is picked on this velocity model, and the static correction amount is obtained by combining information such as elevation, thereby improving the static correction processing effect.

[0098] PINNtomo, used for intelligent first-arrival tomography inversion, consists of two networks: one for estimating velocity and the other for travel time at any point within the target area. The relationship between velocity and travel time is established using the equations describing seismic wave propagation. Furthermore, at the shot point, the seismic travel time is 0, while at the receiver point, it is the first-arrival travel time recorded in the seismic data. This establishes a loss function for optimizing the network model. The spatial locations of data points at and below the surface, along with the first-arrival travel times received at the receiver points, are input into the network model. The weights of the velocity and travel time networks are updated by minimizing the loss function. After training, the output value of the velocity network is the near-surface velocity model inversion result.

[0099] After obtaining the near-surface velocity model, the low-velocity zone bottom interface is selected, and then the static corrections of the shot point and receiver point relative to the reference plane are calculated. The main idea is to first strip away the low-velocity zone, and then fill the space between the low-velocity zone bottom interface and the reference plane with a replacement velocity v. c The medium. The formula is as follows:

[0100]

[0101] Among them, E g E represents the elevation of the shot point or receiver point. d h is the elevation of the reference plane. i and v i These represent the thickness and velocity of each layer in the low-speed band, respectively, where M is the number of low-speed band layers, and v c For replacement speed.

[0102] Therefore, the basic process of the intelligent first-arrival chromatography static correction method can be summarized as follows:

[0103] (1) The near-surface velocity model is obtained by PINNtomo intelligent inversion. The first arrival data, observation system file and grid file are read, the network training parameters, the sampling interval and maximum depth of the velocity model are set, and the near-surface velocity model is output after the network training is completed.

[0104] (2) Picking the low-speed layer bottom interface. On the velocity model obtained by the above inversion, set the peeling layer velocity, interactively pick the low-speed band bottom interface, and smooth the automatically picked low-speed band bottom interface according to the set smoothing parameters.

[0105] (3) Calculate the static correction. Set the replacement velocity and calculate the static correction based on Formula 1 using the near-surface velocity model and the low-velocity zone bottom interface.

[0106] (4) Static correction processing. The static correction amount obtained above is used to perform time difference correction on the seismic data to obtain the first-arrival tomographic static correction seismic data.

[0107] Example 3

[0108] like Figure 3 The diagram shown is a functional block diagram of an intelligent first-arrival chromatography static correction device provided in this embodiment.

[0109] The intelligent first-arrival tomography static correction device 100 described in this invention can be installed in an electronic device. Depending on the functions implemented, the intelligent first-arrival tomography static correction device 100 may include a parameter setting module 101, a model training module 102, a low-speed band pickup module 103, a static correction amount calculation module 104, and a data correction module 105. The module described in this invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and are stored in the memory of the electronic device.

[0110] In this embodiment, the functions of each module / unit are as follows:

[0111] The parameter setting module 101 is used to acquire initial arrival data, observation system files and grid files, and set network training parameters, sampling interval and maximum depth;

[0112] The model training module 102 is used to perform intelligent inversion based on the initial arrival data, the network training parameters, the sampling interval, and the maximum depth to obtain a near-surface velocity model.

[0113] The low-velocity zone picking module 103 is used to pick the low-velocity zone bottom interface according to the preset peeling layer velocity using the near-surface velocity model, and to smooth the low-velocity zone bottom interface according to the preset smoothing parameters to obtain a smooth low-velocity zone bottom interface.

[0114] The static correction calculation module 104 is used to calculate the static correction based on the smooth low-speed zone bottom interface and the near-surface velocity.

[0115] The data correction module 104 is used to perform time difference correction on the seismic data to be processed using the static correction amount to obtain corrected data.

[0116] In this embodiment of the invention, a near-surface velocity model is obtained through intelligent inversion based on pre-acquired first-arrival data, observation system files, and grid files. A low-velocity bottom interface is then identified based on the near-surface velocity model. A static correction is calculated based on a preset replacement velocity and the low-velocity bottom interface. Finally, the static correction is used to correct the seismic data for time difference, resulting in corrected data. This approach can flexibly adapt to surface conditions with varying degrees of undulation, improving universality. By utilizing a neural network model to establish the near-surface velocity model, model establishment efficiency is improved, correction costs are reduced, and correction accuracy is enhanced.

[0117] Example 4

[0118] Based on the above embodiments, this embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method described in the above embodiments.

[0119] In some embodiments of this example, a computer-readable storage medium is provided, on which a computer program is stored, characterized in that the computer program, when executed by a processor, performs the following steps:

[0120] Acquire initial arrival data, observation system files, and grid files; set network training parameters, sampling interval, and maximum depth.

[0121] Based on the initial arrival data, the network training parameters, the sampling interval, and the maximum depth, an intelligent inversion is performed to obtain a near-surface velocity model;

[0122] The low-velocity zone bottom interface is picked up using the near-surface velocity model according to the preset peeling layer velocity, and the low-velocity zone bottom interface is smoothed according to the preset smoothing parameters to obtain a smooth low-velocity zone bottom interface.

[0123] Calculate the static correction amount based on the smooth low-speed zone bottom interface and the near-surface velocity;

[0124] The static correction value is used to perform time difference correction on the seismic data to be processed, and the corrected data is obtained.

[0125] In some embodiments of this example, a computer program product is provided, including a computer program, characterized in that the computer program, when executed by a processor, performs the following steps:

[0126] Acquire initial arrival data, observation system files, and grid files; set network training parameters, sampling interval, and maximum depth.

[0127] Based on the initial arrival data, the network training parameters, the sampling interval, and the maximum depth, an intelligent inversion is performed to obtain a near-surface velocity model;

[0128] The low-velocity zone bottom interface is picked up using the near-surface velocity model according to the preset peeling layer velocity, and the low-velocity zone bottom interface is smoothed according to the preset smoothing parameters to obtain a smooth low-velocity zone bottom interface.

[0129] Calculate the static correction amount based on the smooth low-speed zone bottom interface and the near-surface velocity;

[0130] The processor may include, but is not limited to, one or more processors or microprocessors. Each processor may be implemented as an Application Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), controller, microcontroller, microprocessor, or other electronic component, for executing the methods in the above embodiments.

[0131] Computer-readable storage media can be implemented by any type of volatile or non-volatile storage device or a combination thereof. Computer-readable storage media may include, but are not limited to, random access memory (RAM), read-only memory (ROM), flash memory, EPROM memory, EEPROM memory, registers, and computer storage media (e.g., hard disks, floppy disks, solid-state drives, removable disks, Blu-ray discs, etc.).

[0132] Computer-readable storage media may also store at least one computer-executable program, such as computer-readable instructions. Computer-readable storage media include, but are not limited to, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Computer-readable storage media may include, for example, read-only memory (ROM), hard disk, flash memory, etc. For example, a non-transitory computer-readable storage medium may be connected to a computing device such as a computer, and then, when the computing device executes the computer-readable instructions stored on the computer-readable storage medium, the various methods described above can be performed.

[0133] In addition, the computer device may include (but is not limited to) a data bus, an input / output (I / O) bus, a display, and input / output devices (e.g., keyboard, mouse, speakers, etc.).

[0134] The processor can communicate with external devices via the I / O bus through wired or wireless networks.

[0135] In one embodiment, the at least one computer-executable instruction may also be compiled into or comprise a software product / computer program product, wherein one or more computer-executable instructions are executed by a processor to perform the steps of the various functions and / or methods in the embodiments described herein.

[0136] In the embodiments provided in this disclosure, it should be understood that the disclosed apparatus and methods can also be implemented in other ways. The apparatus embodiments described above are merely illustrative; for example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0137] It should be noted that, in this disclosure, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element limited by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0138] While the embodiments disclosed herein are as described above, the foregoing content is merely for the purpose of facilitating understanding of this disclosure and is not intended to limit this disclosure. Any person skilled in the art to which this disclosure pertains may make any modifications and changes in form and detail of the implementation without departing from the spirit and scope of this disclosure; however, the scope of patent protection of this disclosure shall still be determined by the scope defined in the appended claims.

Claims

1. An intelligent first break analysis statics method, characterized in that, include: Acquire initial arrival data, observation system files, and grid files; set network training parameters, sampling interval, and maximum depth. Based on the initial arrival data, the network training parameters, the sampling interval, and the maximum depth, an intelligent inversion is performed to obtain a near-surface velocity model; The low-velocity zone bottom interface is picked up using the near-surface velocity model according to the preset peeling layer velocity, and the low-velocity zone bottom interface is smoothed according to the preset smoothing parameters to obtain a smooth low-velocity zone bottom interface. Calculate the static correction amount based on the smooth low-speed zone bottom interface and the near-surface velocity; The static correction value is used to perform time difference correction on the seismic data to be processed, and the corrected data is obtained.

2. The method of claim 1, wherein, The setting of network training parameters, sampling interval, and maximum depth includes: Set the learning rate, batch size, and number of iterations to obtain the network training parameters; The sampling interval and maximum depth are set according to the user's input parameters.

3. The method of claim 1, wherein, The intelligent inversion based on the initial arrival data, the network training parameters, the sampling interval, and the maximum depth to obtain the near-surface velocity model includes: The neural network model outputs the model results based on the grid file. Calculate the error between the network model output and the initial arrival data, and update the network model weights based on the error using the minimum loss function; Once the number of iterations specified in the network training parameters is reached, the model training is confirmed to be complete, and the near-surface velocity model is obtained.

4. The method of claim 1, wherein, The step of smoothing the low-speed band bottom interface according to preset smoothing parameters to obtain a smooth low-speed band bottom interface includes: Set the smoothing window according to the smoothing parameters; The smooth window is used to slide sequentially on the low-speed band bottom interface; Calculate the Gaussian weighted value of the data within the smoothing window after each slide, and replace the original data in the center of the smoothing window with the calculated Gaussian weighted value. After all data has been replaced, the smoothing process is confirmed to be complete, resulting in the smoothed low-speed bottom interface.

5. The method of claim 1, wherein, The calculation of the static correction based on the smooth low-speed zone bottom interface and the near-surface velocity includes: The static correction amount is calculated using the following formula: Where t is the static correction amount, M is the number of low-speed band layers at the smooth low-speed band bottom interface, and h i and v i E represents the thickness and velocity of the smooth low-speed band bottom interface, respectively. d E is the pre-acquired elevation of the receiver point. g For the elevation of the pre-obtained reference surface, v c This is the preset replacement speed.

6. The method of claim 1, wherein, The process of using the static correction amount to perform time difference correction on the seismic data to be processed, to obtain corrected data, includes: The corrected data is obtained by shifting the seismic data by time using the static correction amount.

7. An intelligent first break analysis statics device, characterized in that, include: The parameter setting module is used to acquire initial arrival data, observation system files, and grid files, and to set network training parameters, sampling intervals, and maximum depth. The model training module is used to perform intelligent inversion based on the initial arrival data, the network training parameters, the sampling interval, and the maximum depth to obtain a near-surface velocity model; The low-velocity zone picking module is used to pick up the low-velocity zone bottom interface according to the preset peeling layer velocity using the near-surface velocity model, and to smooth the low-velocity zone bottom interface according to the preset smoothing parameters to obtain a smooth low-velocity zone bottom interface. The static correction calculation module is used to calculate the static correction based on the smooth low-speed zone bottom interface and the near-surface velocity. The data correction module is used to perform time difference correction on the seismic data to be processed using the static correction amount to obtain corrected data.

8. A computer device, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 6.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, When executed by a processor, the computer program implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the steps of the method according to any one of claims 1 to 6.

Images