Multiple suppression method and device, electronic equipment and storage medium

Abstract

The invention provides a multiple suppression method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining pre-stack common shot gather data through a pre-established and trained suppression model; determining multiple data in the pre-stack common shot gather data; and inputting the pre-stack common shot gather data and the predicted multiple wave model data into a pre-trained suppression model to obtain primary wave data, thereby realizing multiple wave suppression and improving multiple wave suppression precision.

Description

Technical Field

[0001] The present application relates to the field of geophysical exploration technology, and in particular to a multiple wave suppression method, device, electronic equipment and storage medium. Background Art

[0002] The multiple wave suppression technology of seismic data has also been a key research topic in the geophysics community. The existence of multiple waves reduces the signal-to-noise ratio of seismic data, creates structural illusions, and seriously affects the authenticity and reliability of seismic data imaging. In practical applications, existing technical software still has practical problems such as unsatisfactory multiple wave suppression and difficulty in suppressing multiple waves between layers. Summary of the invention

[0003] In view of the above problems, the present application provides a multiple wave suppression method, device, electronic device and storage medium, which can suppress multiple waves and improve the accuracy of multiple wave suppression.

[0004] The present application provides a multiple wave suppression method, comprising:

[0005] Obtain pre-stack common shot gather data;

[0006] determining predicted multiple wave model data in the prestack common shot gather data;

[0007] The pre-stack common shot gather data and the predicted multiple wave model data are input into a pre-trained suppression model to obtain primary wave data.

[0008] In some embodiments, the method further comprises:

[0009] Acquire synthetic data and primary wave data corresponding to the synthetic data, wherein the synthetic data includes: sample pre-stack common shot gather data and sample predicted multiple wave model data in the sample pre-stack common shot gather data;

[0010] Determine a sample training set based on the synthetic data and the primary wave data corresponding to the synthetic data;

[0011] Performing data enhancement processing on the sample training set to obtain a target sample training set;

[0012] The network model is trained based on the target sample training set to obtain a trained suppression model.

[0013] In some embodiments, the network model includes: an input end, the input end includes: a first channel and a second channel, the first channel is used to input sample pre-stack common shot data, the second channel is used to input sample predicted multiple wave model data, and the output end of the network model is used to output primary wave data corresponding to the synthetic data.

[0014] In some embodiments, the network model is a fully convolutional neural network, and the structure of the fully convolutional neural network model includes: a Unet network structure.

[0015] In some embodiments, the loss function of the network model includes: an L1 loss function.

[0016] In an embodiment of the present application, a multiple wave suppression device is provided, comprising:

[0017] An acquisition module is used to acquire pre-stack common shot gather data;

[0018] A determination module, used for determining predicted multiple wave model data in the pre-stack common shot gather data;

[0019] The suppression module is used to input the pre-stack common shot gather data and the predicted multiple wave model data into a pre-trained suppression model to obtain primary wave data.

[0020] In some embodiments, the multiple wave suppression device is further used to:

[0021] Acquire synthetic data and primary wave data corresponding to the synthetic data, wherein the synthetic data includes: sample pre-stack common shot gather data and sample predicted multiple wave model data in the sample pre-stack common shot gather data;

[0022] Determine a sample training set based on the synthetic data and the primary wave data corresponding to the synthetic data;

[0023] Performing data enhancement processing on the sample training set to obtain a target sample training set;

[0024] The network model is trained based on the target sample training set to obtain a trained suppression model.

[0025] In some embodiments, the network model includes: an input end, the input end includes: a first channel and a second channel, the first channel is used to input sample pre-stack common shot data, the second channel is used to input sample predicted multiple wave model data, the output end of the network model is used to output primary wave data corresponding to the synthetic data, the network model is a fully convolutional neural network, the structure of the fully convolutional neural network model includes: Unet network structure, and the loss function of the network model includes: L1 loss function.

[0026] An embodiment of the present application provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, any one of the multiple wave suppression methods described above is executed.

[0027] An embodiment of the present application provides a storage medium, wherein a computer program stored in the storage medium can be executed by one or more processors and can be used to implement any of the multiple wave suppression methods described above.

[0028] The present application provides a multiple wave suppression method, device, electronic device and storage medium, which achieves suppression of multiple waves and can improve the suppression accuracy of multiple waves by pre-establishing a trained suppression model, obtaining pre-stack common shot gather data, determining predicted multiple wave model data in the pre-stack common shot gather data, and inputting the pre-stack common shot gather data and the multiple wave data into the pre-trained suppression model to obtain primary wave data. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Hereinafter, the present application will be described in more detail based on embodiments and with reference to the accompanying drawings.

[0030] Figure 1 A schematic diagram of a multiple wave suppression method according to an embodiment of the present application;

[0031] Figure 2 A schematic diagram of the implementation flow of another multiple wave suppression method provided in an embodiment of the present application;

[0032] Figure 3 A schematic diagram of the structure of a network model provided in an embodiment of the present application;

[0033] Figure 4 A schematic diagram of pre-processing data provided in an embodiment of the present application;

[0034] Figure 5 A schematic diagram of predicted multiple wave model data provided in an embodiment of the present application;

[0035] Figure 6 A schematic diagram of primary wave data provided in an embodiment of the present application;

[0036] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0037] In the drawings, the same reference numerals are used for the same components, and the drawings are not drawn to scale. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0039] In the following description, reference is made to “some embodiments”, which describe a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0040] If similar descriptions of "first\second\third" appear in the application documents, the following instructions will be added. In the following description, the terms "first\second\third" involved are merely used to distinguish similar objects and do not represent a specific ordering of the objects. It can be understood that "first\second\third" can be interchanged in a specific order or sequence where permitted, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.

[0041] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0042] Before introducing the embodiments of the present application, a brief introduction to the related technologies is given:

[0043] At present, with the strengthening of exploration requirements, the technology of suppressing multiple waves at home and abroad is also gradually developing. The main multiple wave suppression technologies abroad can be summarized into two categories: (1) Filtering method based on signal analysis. Among them, the predictive deconvolution method uses the periodic characteristics of multiple waves to predict the pure interference part based on the information of the first reflection and interference of the seismic record. The physical meaning is clear, but the period of the multiple waves is difficult to determine. The Radon transform method uses the velocity difference between the multiple waves and the effective waves to transform the wave field into the Radon domain to eliminate the multiple waves. There are three transformation methods: linear, hyperbolic, and parabolic Radon. The advantages of this type of method are small calculation amount, high efficiency, and easy implementation, but the multiple wave removal effect at close offset is not good and it is easy to damage the effective signal. (2) Predictive subtraction method based on wave theory. Among them, the wave field extrapolation method suppresses the water layer multiple waves on the CDP gather through wave field extension and adaptive subtraction, and has achieved significant results. The feedback iteration method (SRME) is based on an accurate multiple wave model and wave theory. It uses the wave field propagation information of seismic waves between the free surface and the layer interface of the model to construct surface-related multiple waves, and then subtracts the calculated multiple waves from the collected seismic data. The advantage of this type of method is that it can adapt to any complex underground structure, but the disadvantage is that the amount of calculation is large and the data preprocessing requirements are high.

[0044] Based on the problems existing in the related art, the embodiment of the present application provides a multiple wave suppression method, which is applied to electronic devices, such as computers, mobile terminals, etc. The functions implemented by the multiple wave suppression method provided in the embodiment of the present application can be implemented by calling program codes by a processor of the electronic device, wherein the program codes can be stored in a computer storage medium.

[0045] Embodiment 1

[0046] The present application provides a multiple wave suppression method. Figure 1 A schematic diagram of a multiple wave suppression method provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, including:

[0047] Step S1, obtaining pre-stack common shot gather data.

[0048] In the embodiment of the present application, the pre-stack common shot gather data includes: multiple waves and primary waves.

[0049] In the embodiment of the present application, the pre-stack common shot gather data can be obtained through input from a testing device, a storage device, etc. The testing device can be a shot gather acquisition device.

[0050] In some embodiments, the electronic device may be connected to a database for communication and acquire pre-stack common shot gather data from the database.

[0051] Step S2, determining predicted multiple wave model data in the pre-stack common shot gather data.

[0052] In the embodiment of the present application, the pure interference part can be predicted based on the information of primary reflection and interference recorded in the pre-stack common shot gather data to obtain multiple wave data.

[0053] In some embodiments, the Radon transformation may be used to transform the wave field into the Radon domain for processing by utilizing the velocity difference between the multiple waves and the effective waves to obtain predicted multiple wave model data.

[0054] Step S3, inputting the pre-stack common shot gather data and the multiple wave data into a pre-trained suppression model to obtain primary wave data.

[0055] In an embodiment of the present application, the pre-trained suppression model may be a neural network model.

[0056] The present application provides a multiple wave suppression method, which achieves suppression of multiple waves by pre-establishing a trained suppression model, obtaining pre-stack common shot gather data; determining multiple wave data in the pre-stack common shot gather data; and inputting the pre-stack common shot gather data and the multiple wave data into the pre-trained suppression model to obtain primary wave data, thereby achieving suppression of multiple waves and improving the suppression accuracy of multiple waves.

[0057] Embodiment 2

[0058] Based on the above embodiments, the present application further provides a multiple wave suppression method. Figure 2 A schematic diagram of the implementation process of a multiple wave suppression method provided for the example of this application is as follows: Figure 2 As shown, the method includes:

[0059] Step S11, obtaining synthetic data and primary wave data corresponding to the synthetic data, wherein the synthetic data includes: sample pre-stack common shot gather data and sample predicted multiple wave model data in the sample pre-stack common shot gather data.

[0060] Step S12: determining a sample training set based on the synthetic data and the primary wave data corresponding to the synthetic data.

[0061] Step S13, performing data enhancement processing on the sample training set to obtain a target sample training set.

[0062] In the embodiment of the present application, the enhancement processing may be to perform block processing on the sample data in the sample training set, for example, to divide the sample data into data of 256*256 size.

[0063] Step S14, training the network model based on the target sample training set to obtain a trained suppression model.

[0064] In the embodiment of the present application, Figure 3 A schematic diagram of a network model provided in an embodiment of the present application is shown in FIG. Figure 3 As shown, the network model includes: an input end, the input end includes: a first channel and a second channel, the first channel is used to input sample pre-stack common shot data, the second channel is used to input sample predicted multiple wave model data, and the output end of the network model is used to output the primary wave data corresponding to the synthetic data.

[0065] The network model is a fully convolutional neural network, and the structure of the fully convolutional neural network model includes: a Unet network structure. The loss function of the network model includes: an L1 loss function.

[0066] In the embodiment of the present application, the network model is divided into two parts: downsampling and upsampling. The basic structural unit in the downsampling path is 1 convolution layer (convolution kernel size 4x4), 1 BatchNorm layer, and 1 LeaklyRelu layer, a total of 5 groups, and the number of convolution layer channels has increased from the original 64 to 512, and the number of channels in each group is twice that of the previous group. In the upsampling path, a deconvolution algorithm is used for image upsampling, and its basic structural unit is 1 deconvolution layer (convolution kernel size 4x4), 1 BatchNorm layer, and 1 Relu layer. In addition, a replication method is further used to short-circuit the data in the left downsampling to improve the richness of data information in the upsampling process.

[0067] In the embodiment of the present application, the network model is a dual-channel input and single-channel output structure, and the size of the input channel data and the output data are consistent. Based on the sparse characteristics of the latter wave, the loss function used by the objective function is the L1 norm, and the Adam optimizer is used for iterative update.

[0068] Step S15, obtaining pre-stack common shot gather data.

[0069] In the embodiment of the present application, the pre-stack common shot gather data includes: predicted multiple wave model data and primary waves.

[0070] In the embodiment of the present application, the pre-stack common shot gather data can be obtained through input from a testing device, a storage device, etc. The testing device can be a shot gather acquisition device.

[0071] In some embodiments, the electronic device may be connected to a database for communication and acquire pre-stack common shot gather data from the database.

[0072] Step S16, determining predicted multiple wave model data in the pre-stack common shot gather data.

[0073] In the embodiment of the present application, the pure interference part can be predicted based on the information of primary reflection and interference recorded in the pre-stack common shot gather data to obtain the multiple wave model data.

[0074] In some embodiments, the Radon transformation may be used to transform the wave field into the Radon domain for processing by utilizing the velocity difference between the multiple waves and the effective waves to obtain the multiple wave model data.

[0075] Step S17, inputting the pre-stack common shot gather data and the multiple wave data into a pre-trained suppression model to obtain primary wave data.

[0076] In an embodiment of the present application, the pre-trained suppression model may be a neural network model.

[0077] The present application provides a multiple wave suppression method, which achieves suppression of multiple waves by pre-establishing a trained suppression model, obtaining pre-stack common shot gather data; determining multiple wave data in the pre-stack common shot gather data; and inputting the pre-stack common shot gather data and the multiple wave data into the pre-trained suppression model to obtain primary wave data, thereby achieving suppression of multiple waves and improving the suppression accuracy of multiple waves.

[0078] Embodiment 3

[0079] Based on the above-mentioned embodiments, the embodiments of the present application further provide a multiple wave suppression method. The present invention introduces multiple wave model information constraints in a conventional pure end-to-end data-driven multiple wave suppression deep learning network that uses the data before multiple wave suppression as input and the primary wave data as a label to improve the stability and accuracy of network training. The network input data is dual-channel data, which are full wave field data containing multiple waves and corresponding multiple wave model data. The output of the network is the primary wave data of the corresponding channel gather. The network is a UNET network structure of a fully convolutional neural network, which is divided into two parts: downsampling and upsampling. The basic structural unit in the downsampling path is 1 convolution layer (convolution kernel size 4x4), 1 BatchNorm layer, and 1 LeaklyRelu layer, a total of 5 groups, and the number of channels in the convolution layer increases from the original 64 to 512, and the number of channels in each group is twice that of the previous group. In the upsampling path, the deconvolution algorithm is used for image upsampling. Its basic structural unit is 1 deconvolution layer (convolution kernel size 4x4), 1 BatchNorm layer, and 1 Relu layer. In addition, the replication method is further used to short-circuit the data in the left downsampling to improve the richness of data information in the upsampling process.

[0080] In the embodiment of the present application, the designed network is a dual-channel input and single-channel output structure, and the size of the input channel data and the output data are consistent. Based on the sparse characteristics of the latter wave, the loss function used by the objective function is the L1 norm, and the Adam optimizer is used for iterative update.

[0081] In the embodiment of the present application, the basic process of the algorithm includes:

[0082] Data preparation and data screening: Data preparation is prepared from two channels: synthetic data and actual data to finally form the sample label set required by the present invention. Synthetic data can enhance the accuracy of multiple wave suppression, and actual data can enhance the generalization ability of the network by selecting data with good suppression effect as sample labels.

[0083] Preparation of target sample training set: Aiming at the small sample problem of seismic data, the screened and prepared data set is processed in blocks, and the input data and corresponding label data are divided into 256*256 blocks to achieve data enhancement to improve sample diversity and enhance network generalization ability, and generate sample label data pairs according to the network input and output data requirements;

[0084] Network training: Set network training hyperparameters, feed the prepared training data set into the network for training, and finally obtain a high-precision intelligent multiple wave suppression network;

[0085] Network reasoning: Reload the saved network, perform multiple wave suppression processing on the test data, and verify the effect of the trained network.

[0086] The method provided in the embodiment of the present application establishes a network architecture with multi-channel input and introduces multiple wave data constraints to enhance the learning information richness of the network. Through network training, the intrinsic relationship between the two input channel data is fully explored to achieve intelligent and high-precision suppression of multiple waves.

[0087] Based on the multiple wave suppression method provided in the above embodiments, the present application embodiment provides a specific example application. The input data size of this training is 256x256, the above network is established, and the learning rate is set to 0.0002, batch_size is set to 5, and the number of learning rounds is set to 30. In addition, the cluster GPU used in this test is RTX 2080Ti with a memory size of 11G.

[0088] Based on the above parameters, the network is trained and the trained model is saved. During the test, the data size is 266*2001, and the input pre-processed data is as follows Figure 4 The multiple wave data used are shown in Figure 5 As shown, the data compressed by the method provided in the embodiment of the present application is as follows Figure 6 As shown in the figure, it can be clearly seen that the method of the present invention has a good effect on multiple wave suppression. The multiple waves after processing are suppressed cleanly without loss of effective waves.

[0089] Embodiment 4

[0090] Based on the foregoing embodiments, the embodiments of the present application provide a multiple wave suppression device, and the modules included in the device, as well as the units included in the modules, can be implemented by a processor in a computer device; of course, they can also be implemented by a specific logic circuit; in the implementation process, the processor can be a central processing unit (CPU, Central Processing Unit), a microprocessor (MPU, Microprocessor Unit), a digital signal processor (DSP, Digital Signal Processing) or a field programmable gate array (FPGA, Field Programmable Gate Array), etc.

[0091] The embodiment of the present application provides a multiple wave suppression device, the multiple wave suppression device comprising:

[0092] An acquisition module is used to acquire pre-stack common shot gather data;

[0093] A determination module, used for determining predicted multiple wave model data in the pre-stack common shot gather data;

[0094] The suppression module is used to input the pre-stack common shot gather data and the multiple wave data into a pre-trained suppression model to obtain primary wave data.

[0095] In the embodiment of the present application, the pre-stack common shot gather data includes: multiple wave model data and primary waves.

[0096] In the embodiment of the present application, the pre-stack common shot gather data can be obtained through input from a testing device, a storage device, etc. The testing device can be a shot gather acquisition device.

[0097] In some embodiments, the electronic device may be connected to a database for communication and acquire pre-stack common shot gather data from the database.

[0098] In the embodiment of the present application, the pure interference part can be predicted based on the information of primary reflection and interference recorded in the pre-stack common shot gather data to obtain multiple wave data.

[0099] In some embodiments, the Radon transformation may be used to transform the wave field into the Radon domain for processing by utilizing the velocity difference between the multiple waves and the effective waves to obtain the multiple wave model data.

[0100] In some embodiments, the multiple wave suppression device is further used to:

[0101] The multiple wave suppression device is also used for:

[0102] Acquire synthetic data and primary wave data corresponding to the synthetic data, wherein the synthetic data includes: sample pre-stack common shot gather data and sample multiple wave model data in the sample pre-stack common shot gather data;

[0103] Determine a sample training set based on the synthetic data and the primary wave data corresponding to the synthetic data;

[0104] Performing data enhancement processing on the sample training set to obtain a target sample training set;

[0105] The network model is trained based on the target sample training set to obtain a trained suppression model.

[0106] In the embodiment of the present application, the enhancement processing may be to perform block processing on the sample data in the sample training set, for example, to divide the sample data into data of 256*256 size.

[0107] In the embodiment of the present application, the network model is divided into two parts: downsampling and upsampling. The basic structural unit in the downsampling path is 1 convolution layer (convolution kernel size 4x4), 1 BatchNorm layer, and 1 LeaklyRelu layer, a total of 5 groups, and the number of convolution layer channels has increased from the original 64 to 512, and the number of channels in each group is twice that of the previous group. In the upsampling path, a deconvolution algorithm is used for image upsampling, and its basic structural unit is 1 deconvolution layer (convolution kernel size 4x4), 1 BatchNorm layer, and 1 Relu layer. In addition, a replication method is further used to short-circuit the data in the left downsampling to improve the richness of data information in the upsampling process.

[0108] In the embodiment of the present application, the network model is a dual-channel input and single-channel output structure, and the size of the input channel data and the output data are consistent. Based on the sparse characteristics of the latter wave, the loss function used by the objective function is the L1 norm, and the Adam optimizer is used for iterative update.

[0109] In some embodiments, the network model includes: an input end, the input end includes: a first channel and a second channel, the first channel is used to input sample pre-stack common shot data, the second channel is used to input sample multiple wave data, the output end of the network model is used to output primary wave data corresponding to the synthetic data, the network model is a fully convolutional neural network, the structure of the fully convolutional neural network model includes: a Unet network structure, and the loss function of the network model includes: an L1 loss function.

[0110] The present application provides a multiple wave suppression device, which achieves suppression of multiple waves by pre-establishing a trained suppression model, obtaining pre-stack common shot gather data; determining multiple wave data in the pre-stack common shot gather data; and inputting the pre-stack common shot gather data and the multiple wave model data into the pre-trained suppression model to obtain primary wave data, thereby achieving suppression of multiple waves and improving the suppression accuracy of multiple waves.

[0111] It should be noted that, in the embodiments of the present application, if the above-mentioned multiple wave suppression method is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the embodiments of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a U disk, a mobile hard disk, a read-only memory (ROM, Read Only Memory), a magnetic disk or an optical disk. In this way, the embodiments of the present application are not limited to any specific combination of hardware and software.

[0112] Accordingly, an embodiment of the present application provides a storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps in the multiple wave suppression method provided in the above embodiment are implemented.

[0113] Embodiment 5

[0114] An embodiment of the present application provides an electronic device; Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application is shown in FIG. Figure 7 As shown, the electronic device 700 includes: a processor 701, at least one communication bus 702, a user interface 703, at least one external communication interface 704, and a memory 705. The communication bus 702 is configured to realize the connection and communication between these components. The user interface 703 may include a display screen, and the external communication interface 704 may include a standard wired interface and a wireless interface. The processor 701 is configured to execute the program of the multiple wave suppression method stored in the memory to implement the steps in the multiple wave suppression method provided in the above embodiment.

[0115] Wherein, the multiple wave suppression method comprises:

[0116] Obtain pre-stack common shot gather data;

[0117] Determining multiple wave model data in the pre-stack common shot gather data;

[0118] The pre-stack common shot gather data and the multiple wave model data are input into a pre-trained suppression model to obtain primary wave data.

[0119] In the embodiment of the present application, the pre-stack common shot gather data includes: multiple wave model data and primary waves.

[0120] In the embodiment of the present application, the pre-stack common shot gather data can be obtained through input from a testing device, a storage device, etc. The testing device can be a shot gather acquisition device.

[0121] In some embodiments, the electronic device may be connected to a database for communication and acquire pre-stack common shot gather data from the database.

[0122] In the embodiment of the present application, the pure interference part can be predicted based on the information of primary reflection and interference recorded in the pre-stack common shot gather data to obtain the multiple wave model data.

[0123] In some embodiments, the Radon transformation may be used to transform the wave field into the Radon domain for processing by utilizing the velocity difference between the multiple waves and the effective waves to obtain the multiple wave model data.

[0124] In some embodiments, the method further comprises:

[0125] Acquire synthetic data and primary wave data corresponding to the synthetic data, wherein the synthetic data includes: sample pre-stack common shot gather data and sample predicted multiple wave model data in the sample pre-stack common shot gather data;

[0126] Determine a sample training set based on the synthetic data and the primary wave data corresponding to the synthetic data;

[0127] Performing data enhancement processing on the sample training set to obtain a target sample training set;

[0128] The network model is trained based on the target sample training set to obtain a trained suppression model.

[0129] In some embodiments, the network model includes: an input end, the input end includes: a first channel and a second channel, the first channel is used to input sample pre-stack common shot data, the second channel is used to input sample predicted multiple wave model data, and the output end of the network model is used to output primary wave data corresponding to the synthetic data.

[0130] In some embodiments, the network model is a fully convolutional neural network, and the structure of the fully convolutional neural network model includes: a Unet network structure.

[0131] In some embodiments, the loss function of the network model includes: an L1 loss function.

[0132] The network model is a fully convolutional neural network, and the structure of the fully convolutional neural network model includes: a Unet network structure. The loss function of the network model includes: an L1 loss function.

[0133] In the embodiment of the present application, the network model is divided into two parts: downsampling and upsampling. The basic structural unit in the downsampling path is 1 convolution layer (convolution kernel size 4x4), 1 BatchNorm layer, and 1 LeaklyRelu layer, a total of 5 groups, and the number of convolution layer channels has increased from the original 64 to 512, and the number of channels in each group is twice that of the previous group. In the upsampling path, a deconvolution algorithm is used for image upsampling, and its basic structural unit is 1 deconvolution layer (convolution kernel size 4x4), 1 BatchNorm layer, and 1 Relu layer. In addition, a replication method is further used to short-circuit the data in the left downsampling to improve the richness of data information in the upsampling process.

[0134] In the embodiment of the present application, the network model is a dual-channel input and single-channel output structure, and the size of the input channel data and the output data are consistent. Based on the sparse characteristics of the latter wave, the loss function used by the objective function is the L1 norm, and the Adam optimizer is used for iterative update.

[0135] The description of the above electronic device and storage medium embodiments is similar to the description of the above method embodiments, and has similar beneficial effects as the method embodiments. For technical details not disclosed in the computer device and storage medium embodiments of this application, please refer to the description of the method embodiments of this application for understanding.

[0136] It should be understood that "one embodiment" or "an embodiment" mentioned throughout the specification means that specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. It should be understood that in various embodiments of the present application, the size of the sequence number of the above-mentioned processes does not mean the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application. The above-mentioned sequence numbers of the embodiments of the present application are only for description and do not represent the advantages and disadvantages of the embodiments.

[0137] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0138] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0139] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0140] In addition, all functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may be separately configured as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0141] A person of ordinary skill in the art can understand that: all or part of the steps of implementing the above-mentioned method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above-mentioned method embodiment; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read Only Memory), disks or optical disks, etc. Various media that can store program codes.

[0142] Alternatively, if the above-mentioned integrated unit of the present application is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application can essentially or in other words, the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for a controller to execute all or part of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROMs, magnetic disks, or optical disks.

[0143] The above is only an implementation method of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A multiple wave suppression method, characterized in that: include: Obtain pre-stack common shot gather data; determining predicted multiple wave model data in the prestack common shot gather data; The pre-stack common shot gather data and the predicted multiple wave model data are input into a pre-trained suppression model to obtain primary wave data.

2. The method according to claim 1, characterized in that The method further comprises: Acquire synthetic data and primary wave data corresponding to the synthetic data, wherein the synthetic data includes: sample pre-stack common shot gather data and sample predicted multiple wave model data in the sample pre-stack common shot gather data; Determine a sample training set based on the synthetic data and the primary wave data corresponding to the synthetic data; Performing data enhancement processing on the sample training set to obtain a target sample training set; The network model is trained based on the target sample training set to obtain a trained suppression model.

3. The method according to claim 2, characterized in that The network model includes: an input end, the input end includes: a first channel and a second channel, the first channel is used to input sample pre-stack common shot gather data, the second channel is used to input sample predicted multiple wave model data, and the output end of the network model is used to output primary wave data corresponding to the synthetic data.

4. The method according to claim 3, characterized in that The network model is a fully convolutional neural network, and the structure of the fully convolutional neural network model includes: a Unet network structure.

5. The method according to claim 4, characterized in that The loss function of the network model includes: L1 loss function.

6. A multiple wave suppression device, characterized in that: include: An acquisition module is used to acquire pre-stack common shot gather data; A determination module, used for determining predicted multiple wave model data in the pre-stack common shot gather data; The suppression module is used to input the pre-stack common shot gather data and the predicted multiple wave model data into a pre-trained suppression model to obtain primary wave data.

7. The multiple wave suppression device according to claim 6, characterized in that: The multiple wave suppression device is also used for: Acquire synthetic data and primary wave data corresponding to the synthetic data, wherein the synthetic data includes: sample pre-stack common shot gather data and sample predicted multiple wave model data in the sample pre-stack common shot gather data; Determine a sample training set based on the synthetic data and the primary wave data corresponding to the synthetic data; Performing data enhancement processing on the sample training set to obtain a target sample training set; The network model is trained based on the target sample training set to obtain a trained suppression model.

8. The multiple wave suppression device according to claim 7, characterized in that: The network model includes: an input end, the input end includes: a first channel and a second channel, the first channel is used to input sample pre-stack common shot gather data, the second channel is used to input sample predicted multiple wave model data, the output end of the network model is used to output primary wave data corresponding to the synthetic data, the network model is a fully convolutional neural network, the structure of the fully convolutional neural network model includes: a Unet network structure, and the loss function of the network model includes: an L1 loss function.

9. An electronic device, characterized in that: The method comprises a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the multiple wave suppression method according to any one of claims 1 to 5 is executed.

10. A storage medium, characterized in that: The computer program stored in the storage medium can be executed by one or more processors and can be used to implement the multiple wave suppression method as claimed in any one of claims 1 to 5.

Images