Vibroseis black triangular area noise suppression processing method, device, equipment and medium

Abstract

The invention relates to a vibroseis black triangular area noise suppression processing method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring seismic reflected wave data; respectively constructing models for the effective signal data and the black triangular area noise data; carrying out sparse inversion on the noise data of the black triangular area to obtain a sparse item; according to the sparse item, black triangular area noise prediction data are determined; and obtaining effective signal wave data by adopting adaptive subtraction. According to the method, the sparse term is obtained through sparse inversion, the boundary of the black triangular area noise data is further obtained, the black triangular area noise prediction data is obtained accordingly, the effective signal wave data is obtained by subtracting the black triangular area noise prediction data from the seismic reflection wave data, the constraint effect on the black triangular noise data is improved, the separation effect of the black triangular noise and the effective signal is improved, and the method is suitable for large-scale popularization and application. And moreover, the damage to effective signals in the seismic reflected wave data is reduced, so that the subsequent imaging resolution and precision can be improved.

Description

Technical Field

[0001] The present application relates to the field of geophysical exploration technology, and in particular to a method, device, computer equipment and storage medium for noise suppression processing in a black triangle area of ​​a controllable seismic source. Background Art

[0002] In geophysical exploration, compared with traditional explosive sources, vibroseis has many advantages such as wavelet controllability, environmental protection, and high efficiency. Since vibroseis can only be excited on the surface, in the case of complex surface topography and near-surface lateral inhomogeneity, a strong amplitude noise area will be generated that gradually increases with the increase of offset distance and propagation time. In the vibroseis seismic record, it appears as a triangular area with the shot point as the vertex, which is called "black triangle" noise. The existence of this noise brings difficulties to the subsequent vibroseis data processing, and significantly affects the imaging quality of deep target reservoirs, which seriously hinders the large-scale application of vibroseis.

[0003] When a vibrator is excited on the surface, most of the energy generated is propagated along the surface of the earth in the form of surface waves. When encountering surface obstacles, scattered surface waves are generated. These scattered surface waves interfere with each other to form a "black triangle" noise area of ​​the vibrator with the direct surface wave as the boundary. Considering the complexity of the "black triangle" noise characteristics, traditional noise suppression methods, such as FK filtering, Radon transform, and curvelet transform, cannot effectively constrain the "black triangle" noise, making it impossible to completely separate the "black triangle" noise from the effective reflection wave signal, resulting in damage to the effective signal during suppression, and ultimately affecting the imaging resolution and accuracy of the vibrator data. Summary of the invention

[0004] Based on this, it is necessary to provide a controllable seismic source black triangle area noise suppression processing method, device, computer equipment and storage medium that can improve the noise constraint effect, improve the separation effect of black triangle noise and effective signals, reduce damage to effective signals, and improve subsequent imaging resolution and accuracy.

[0005] In a first aspect, the present application provides a method for suppressing noise in a black triangle area of ​​a vibrator, comprising the following steps:

[0006] Acquire seismic reflection wave data;

[0007] Models are constructed for the effective signal data of seismic reflection wave data and the noise data of the black triangle area respectively, so as to separate the effective signal data of seismic reflection wave data and the noise data of the black triangle area in the model space;

[0008] Based on the model of the effective signal data and the model of the black triangle area noise data, sparse inversion is performed on the black triangle area noise data to obtain a sparse term;

[0009] According to the sparse items, the noise prediction data of the black triangle area is determined;

[0010] Adaptive subtraction is used to subtract the noise prediction data of the black triangle area from the seismic reflection wave data to obtain the effective signal wave data.

[0011] In one embodiment, the steps of constructing models for the effective signal data of the seismic reflection wave data and the noise data of the black triangle area respectively, and realizing separation of the effective signal data of the seismic reflection wave data and the noise data of the black triangle area in the model space include:

[0012] Effective signal data d for seismic reflection wave data d s Construct model operator L s ;

[0013] The noise data d in the black triangle area of ​​the seismic reflection wave data d bt Convert to the curvelet domain to construct the model operator L c , to achieve the separation of effective signal data of seismic reflection wave data and noise data in the black triangle area in the model space.

[0014] In one embodiment, the effective signal data d for the seismic reflection wave data d s Construct model operator L s In the step, the model operator L s It is defined by the following formula:

[0015] d s =L s x s

[0016] L s =N T S T

[0017] Among them, x s is the reflectivity, N T is the dynamic correction operator, S T is the superposition operator.

[0018] In one embodiment, the black triangle noise data d of the seismic reflection wave data d is bt Convert to the curvelet domain to construct the model operator L c In the step, the model operator L c It is defined by the following formula:

[0019] d bt =L c x c

[0020] Among them, x cis the sparse term of the noise data in the black triangle area in the curvelet domain.

[0021] In one embodiment, the step of performing sparse inversion on the black triangle area noise data based on the model of the effective signal data and the model of the black triangle area noise data to obtain the sparse terms comprises:

[0022] In the curvelet domain, an objective function is constructed for the black triangle area noise data based on the model of the effective signal data and the model of the black triangle area noise data, wherein in the objective function, a sparse constraint is added to the black triangle area noise data;

[0023] The final iterative inversion is used to solve the objective function and obtain the sparse terms of the noise data in the black triangle area in the curvelet domain.

[0024] In one embodiment, the objective function is constructed for the black triangle noise data in the curvelet domain, and the objective function J is expressed by the following formula:

[0025] J=||dL s x s -L c x c || 2 +μ||x c ||1

[0026] Among them, μ is the inversion regularization operator, ||x c ||1 is the sparse constraint on the noise data in the black triangle area.

[0027] In one embodiment, in the step of determining the black triangle area noise prediction data according to the sparse items, the black triangle area noise prediction data is determined using the following formula:

[0028] d c =L c x c

[0029] Among them, d c This is the noise prediction data for the black triangle area.

[0030] In a second aspect, the present application provides a controllable vibroseis black triangle area noise suppression processing device, the device comprising:

[0031] A raw data acquisition module, used to acquire seismic reflection wave data;

[0032] A model building module is used to build models for effective signal data of seismic reflection wave data and noise data of black triangle area respectively, so as to separate effective signal data of seismic reflection wave data and noise data of black triangle area in the model space;

[0033] A sparse inversion operation module, used for performing sparse inversion on the black triangle area noise data based on the model of the effective signal data and the model of the black triangle area noise data to obtain a sparse term;

[0034] A noise prediction module is used to determine the noise prediction data of the black triangle area according to the sparse items;

[0035] The effective signal wave data generation module is used to subtract the black triangle area noise prediction data from the seismic reflection wave data by adaptive subtraction to obtain the effective signal wave data.

[0036] In a third aspect, the present application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method described in any of the above embodiments when executing the computer program.

[0037] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the method described in any of the above embodiments are implemented.

[0038] The above-mentioned controllable source black triangle area noise suppression processing method aims at the problem that the traditional transform domain filtering method cannot solve the "black triangle" noise, and provides a controllable source "black triangle" noise suppression method based on sparse inversion. By constructing models for the effective signal data of the original seismic reflection wave data and the black triangle area noise data respectively, the effective signal data of the seismic reflection wave data and the black triangle area noise data in the model space are separated. Through sparse inversion, sparse terms are obtained, and then the boundary of the black triangle area noise data is obtained, and the black triangle area noise prediction data is obtained thereby, and then the black triangle area noise prediction data is subtracted from the seismic reflection wave data by adaptive subtraction to obtain the effective signal wave data. Through the method based on sparse inversion, the black triangle area noise data can be predicted and calculated relatively accurately, and the separation of the black triangle area noise data and the effective signal data in the original seismic reflection wave data is better realized. The sparse terms are solved by sparse inversion, the constraint effect on the black triangle noise data is improved, the separation effect of the black triangle noise and the effective signal is improved, and the damage to the effective signal in the seismic reflection wave data is reduced, thereby improving the subsequent imaging resolution and accuracy.

[0039] In other embodiments, a method for suppressing the "black triangle" noise of a vibroseis source based on sparse inversion in the curvelet domain is provided. The method models the effective signal and the "black triangle" noise in the seismic data in different model space domains and constructs an objective function, and then solves the objective function through sparse inversion, thereby converting the "black triangle" noise suppression problem into a least squares solution problem of the objective function, and replacing the traditional transform domain filtering process with a multi-dimensional inversion process. The inversion results are constrained by adding sparse constraints and linear search steps, and the output of each iteration is used as feedback to enter the next iteration, so as to protect the effective signal to the greatest extent in the internal algorithm. The processing results of actual data show that this method can effectively suppress the "black triangle" noise in the vibroseis source seismic data. Compared with the traditional transform domain filtering suppression method, this method can protect the effective signal to the greatest extent and further improve the resolution of the vibroseis source seismic data. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a flow chart of a method for suppressing noise in a black triangle area of ​​a vibrator in one embodiment;

[0041] Figure 2 It is a schematic diagram of sparse inversion of a noise suppression processing method for a vibroseis black triangle region in an embodiment;

[0042] Figure 3a-3f The frequency division processing result of the single shot data of the vibroseis in the desert area in one embodiment;

[0043] Figure 4a-4c This is a processing effect diagram of a controllable vibrator black triangle area noise suppression processing method applied to single shot data in one embodiment;

[0044] Figure 5a This is the superimposed profile of the effective signal obtained by the traditional transform domain filtering method;

[0045] Figure 5b This is a superimposed cross-sectional diagram of effective signals obtained by using the controllable vibrator black triangle area noise suppression processing method of the present application;

[0046] Figure 6 It is a structural schematic diagram of a controllable vibrator black triangle area noise suppression processing device in one embodiment;

[0047] Figure 7 Schematic diagram of the structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0048] In order to facilitate the understanding of the present application, in order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present application, and the preferred implementation methods of the present application are given in the accompanying drawings. However, the present application can be implemented in many different forms and is not limited to the implementation methods described herein. On the contrary, the purpose of providing these implementation methods is to make the disclosure of the present application more thoroughly and comprehensively understood. The present application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the present application, so the present application is not limited to the specific embodiments disclosed below. In addition, the terms "first" and "second" are used only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. When an element is considered to be "connected" to another element, it may be directly connected to another element or there may be a centering element at the same time. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present application. The terms used herein are only for the purpose of describing specific embodiments and are not intended to limit the present application. The term "and / or" used herein includes any and all combinations of one or more of the associated listed items.

[0049] Example 1

[0050] In one embodiment, if Figure 1 As shown, the controllable vibrator black triangle area noise suppression processing method specifically includes the following steps:

[0051] S210: Acquire seismic reflection wave data;

[0052] In this embodiment, the seismic reflection wave data is the vibroseis data containing black triangle noise, and the vibroseis data containing black triangle noise is the actual seismic data collected. Specifically, when the vibroseis is excited on the surface, in the case of complex surface topography and near-surface lateral unevenness, a strong amplitude noise area that gradually increases with the increase of offset distance and propagation time will be generated, which is manifested as a triangular area with the shot point as the vertex in the vibroseis seismic record, called "black triangle" noise. When the vibroseis is excited on the surface, most of the energy generated is propagated along the surface of the earth in the form of surface waves, and scattered surface waves are generated when encountering surface obstacles. These scattered surface waves interfere with each other to form a vibroseis "black triangle" noise area with direct surface waves as the boundary. The seismic reflection wave data of this embodiment is the reflection wave signal data received by the instrument.

[0053] S220: constructing models for the effective signal data of the seismic reflection wave data and the noise data of the black triangle area respectively; achieving separation of the effective signal data of the seismic reflection wave data and the noise data of the black triangle area in the model space;

[0054] In this embodiment, by constructing models for the effective signal data of the seismic reflection wave data and the noise data of the black triangle area, the separation of the effective signal data of the seismic reflection wave data and the noise data of the black triangle area in the model space is achieved. Specifically, the seismic reflection wave data is the received original data, which is the sum of the effective signal data and the noise data of the black triangle area, that is, the original seismic data can be expressed as the sum of the effective signal of the reflection wave and the noise data of the "black triangle" area. By constructing a model for the original seismic reflection wave data, the separation of the effective signal data of the seismic reflection wave data and the noise data of the black triangle area in the model space domain is achieved.

[0055] In specific applications, the model and related methods include, but are not limited to: filter models, statistical models, neural network models, wavelet transform models, Fourier transforms, etc., and are not limited thereto. In actual applications, by constructing a signal data model, the original seismic data is divided into data within the black triangle area and data outside the black triangle area. The data in the black triangle area is the noise data in the black triangle area, and the data within the black triangle area is the effective signal data, but in fact there is also a mixed area of ​​data containing data within the black triangle area and data outside the black triangle area. Therefore, after the present application realizes the separation of effective signal data and noise data in the black triangle area of ​​seismic reflection wave data in space through a model, it further uses optimization and iterative calculation methods to find the boundary between the noise data in the black triangle area and the effective signal data, thereby better achieving the suppression of noise in the black triangle area and reducing its impact on the effective signal, thereby better separating the noise in the black triangle area from the effective signal. In a specific embodiment, the effective signal data and the black triangle area noise data can be modeled separately to achieve the separation of the effective signal data and the black triangle area noise data of the seismic reflection wave data in the model space. For example, the effective signal data and the black triangle area noise data are modeled in different model space domains and the objective function is constructed, and then the objective function is solved by sparse inversion, so as to convert the problem of suppressing the black triangle area noise into the least squares solution problem of the objective function, and replace the traditional transform domain filtering process with a multi-dimensional inversion process, so as to ensure the suppression of the black triangle area noise data. It should be noted that the black triangle area noise data of the present application can also be understood as the black triangle area noise data. In a specific embodiment, based on the different dispersion and distribution characteristics of the effective signal and the "black triangle" noise, the data is converted to the curvelet domain to achieve the separation of the effective signal and the "black triangle" noise.

[0056] S230: performing sparse inversion on the black triangle area noise data based on the model of the effective signal data and the model of the black triangle area noise data to obtain a sparse term;

[0057] In the present application, after separating the effective signal data and the black triangle area noise data in the model space, the black triangle area noise data is sparsely inverted and sparsely solved to obtain sparse terms, so as to convert the "black triangle" noise suppression problem into a mathematical solution problem, and replace the traditional transform domain filtering process with a multi-dimensional inversion process, so as to effectively suppress the "black triangle" noise in the controllable source seismic data. Compared with the traditional transform domain filtering suppression method, this method can protect the effective signal to the greatest extent and improve the resolution of the controllable source seismic data.

[0058] S240: Determine the noise prediction data of the black triangle area according to the sparse item;

[0059] In the present application, a sparse solution is performed according to sparse inversion, and the noise prediction data of the black triangle area is determined thereby, that is, based on the sparse terms, the noise prediction data of the black triangle area is determined thereby.

[0060] S250: Adaptively subtract the black triangle area noise prediction data from the seismic reflection wave data to obtain effective signal wave data.

[0061] In the present application, the effective signal wave data is the effective signal data after the black triangle area noise prediction data is removed from the original seismic reflection wave data. In the present application, the sparse solution is determined through sparse inversion, thereby determining the boundary of the black triangle area noise data, and then determining the black triangle area noise prediction data, thereby realizing the separation of the black triangle area noise prediction data. After determining the black triangle area noise prediction data, the black triangle area noise prediction data is subtracted from the seismic reflection wave data by using adaptive subtraction to obtain the effective signal wave data. It should be noted that after determining the black triangle area noise prediction data, how to remove the black triangle area noise prediction data based on adaptive subtraction to obtain the effective signal wave data, please refer to the prior art, and this application will not repeat it here.

[0062] The above-mentioned controllable source black triangle area noise suppression processing method aims at the problem that the traditional transform domain filtering method cannot solve the "black triangle" noise, and provides a controllable source "black triangle" noise suppression method based on sparse inversion. By constructing models for the effective signal data of the original seismic reflection wave data and the black triangle area noise data respectively, the effective signal data of the seismic reflection wave data and the black triangle area noise data in the model space are separated. Through sparse inversion, sparse terms are obtained, and then the boundary of the black triangle area noise data is obtained, and the black triangle area noise prediction data is obtained thereby, and then the black triangle area noise prediction data is subtracted from the seismic reflection wave data by adaptive subtraction to obtain the effective signal wave data. Through the method based on sparse inversion, the black triangle area noise data can be predicted and calculated relatively accurately, and the separation of the black triangle area noise data and the effective signal data in the original seismic reflection wave data is better realized. The sparse terms are solved by sparse inversion, the constraint effect on the black triangle noise data is improved, the separation effect of the black triangle noise and the effective signal is improved, and the damage to the effective signal in the seismic reflection wave data is reduced, thereby improving the subsequent imaging resolution and accuracy.

[0063] In one embodiment, in step S220, the effective signal data of the seismic reflection wave data and the noise data of the black triangle area are respectively constructed into models, and the step of separating the effective signal data of the seismic reflection wave data and the noise data of the black triangle area in the model space includes:

[0064] Effective signal data d for seismic reflection wave data d s Construct model operator L s ;

[0065] The noise data d in the black triangle area of ​​the seismic reflection wave data d bt Convert to the curvelet domain to construct the model operator L c , to achieve the separation of effective signal data of seismic reflection wave data and noise data in the black triangle area in the model space.

[0066] In this application, the seismic reflection wave data d can be expressed as the reflection wave effective signal data d s Compared with the noise data in the “black triangle” area d bt The sum of , that is, can be expressed by the following formula:

[0067] d=d s +d bt

[0068] In this embodiment, the effective signal data d s and black triangle area noise data d btThe models are established separately, that is, the effective signal and the "black triangle" noise in the seismic data are modeled in different model space domains, and the black triangle noise data d of the seismic reflection wave data d is converted into bt The model operator is constructed in the curvelet domain, and then the model is optimized and solved through sparse inversion, thereby converting the "black triangle" noise suppression problem into a least squares solution problem of the objective function. The multidimensional inversion process replaces the traditional transform domain filtering process, so as to effectively suppress the "black triangle" noise in the vibroseis data and further improve the resolution of the vibroseis data.

[0069] In a specific embodiment, the effective signal data d for the seismic reflection wave data d s Construct model operator L s In the step, the model operator L s It is defined by the following formula:

[0070] d s =L s x s

[0071] L s =N T S T

[0072] Among them, x s is the reflectivity, N T is the dynamic correction operator, S T is the superposition operator.

[0073] Using the above model operator, according to the hyperbolic NMO assumption, the reflected wave effective signal data d s It can be expressed as the model operator L s With reflectivity x S The product of the model operator is composed of the moving average operator N T and the superposition operator S T Composition, thus we can get:

[0074] d s =L s x s =N T S T x s

[0075] In this way, by constructing a model operator for effective signal data, it is convenient to provide a boundary optimization basis for suppressing noise data in the black triangle area, thereby ensuring that the suppression effect of noise data in the black triangle area is improved and the interference of noise on effective signals is reduced.

[0076] In one embodiment, the black triangle noise data d of the seismic reflection wave data d isbt Convert to the curvelet domain to construct the model operator L c In the step, the model operator L c It is defined by the following formula:

[0077] d bt =L c x c

[0078] Among them, x c is the sparse term of the noise data in the black triangle area in the curvelet domain.

[0079] In this embodiment, the black triangle noise data d contained in the original seismic data d bt Convert to the curvelet domain to construct the model operator L c , so that the noise data d in the black triangle area bt It is able to obtain sparse solutions based on sparse inversion and transform the "black triangle" noise suppression problem into a least squares solution problem of the objective function. The multidimensional inversion process replaces the traditional transform domain filtering process, thereby effectively suppressing the "black triangle" noise in the vibroseis seismic data and further improving the resolution of the vibroseis seismic data.

[0080] In one embodiment, the step of performing sparse inversion on the black triangle area noise data based on the model of the effective signal data and the model of the black triangle area noise data to obtain the sparse terms comprises:

[0081] Based on the model of the effective signal data and the model of the black triangle area noise data, constructing an objective function for the black triangle area noise data in the curvelet domain, wherein in the objective function, a sparse constraint is added to the black triangle area noise data;

[0082] The final iterative inversion is used to solve the objective function and obtain the sparse terms of the noise data in the black triangle area in the curvelet domain.

[0083] In this embodiment, an objective function is constructed, and sparse constraints are added to the black triangle area noise data in the objective function, and the objective function is solved by the final iterative inversion to obtain a sparse solution of the black triangle area noise in the curve domain, that is, a sparse term. In this embodiment, in the final iterative inversion, the inversion result is constrained by adding sparse constraints and linear search steps, and the output of each iteration is used as feedback to enter the next iteration, so that the effective signal data is protected to the greatest extent in the internal algorithm to ensure that the objective function obtains the optimal solution to obtain the sparse term. It should be noted that the specific process of the final iterative inversion please refer to the prior art, and this application will not go into details here. Specifically, for example, the final iterative inversion is an iterative inversion based on the least squares solution. Please refer to Figure 2As shown, this is the final iterative inversion process of the present application. After inputting the initial corresponding model operator data, the least squares method is used to solve based on the constructed objective function, and the final iterative inversion is performed through sparse inversion to finally obtain a sparse solution.

[0084] In one embodiment, the objective function is constructed for the black triangle noise data in the curvelet domain, and the objective function J is expressed by the following formula:

[0085] J=||dL s x s -L c x c || 2 +μ||x c ||1

[0086] Among them, μ is the inversion regularization operator, ||x c ||1 is the sparse constraint on the noise data in the black triangle area.

[0087] In this way, the above objective function is adopted, by modeling the effective signal and "black triangle" noise in the seismic data in different model space domains and constructing the objective function, and then solving the objective function through sparse inversion, so as to transform the "black triangle" noise suppression problem into the least squares solution problem of the objective function, and replace the traditional transform domain filtering process with a multi-dimensional inversion process. By adding sparse constraints and linear search steps to constrain the inversion results, and taking the output of each iteration as feedback to enter the next iteration, the effective signal is protected to the greatest extent in the internal algorithm. The processing results of actual data show that this method can effectively suppress the "black triangle" noise in the vibroseis seismic data. Compared with the traditional transform domain filtering suppression method, this method can protect the effective signal to the greatest extent and further improve the resolution of the vibroseis seismic data.

[0088] In one embodiment, in the step of determining the black triangle area noise prediction data according to the sparse items, the black triangle area noise prediction data is determined using the following formula:

[0089] d c =L c x c

[0090] Among them, d c This is the noise prediction data for the black triangle area.

[0091] In this embodiment, based on the sparse solution obtained, the sparse term x c , thus through the model operator L c With the sparse term x cBy multiplying the black triangle noise prediction data, the noise in the black triangle area can be suppressed by the black triangle noise prediction data, and the effective signal data can be obtained after the black triangle noise prediction data is removed from the original seismic reflection wave data. This method can protect the effective signal to the greatest extent, minimize the impact of the black triangle noise removal on the effective signal, and further improve the resolution of the vibroseis seismic data.

[0092] The above-mentioned controllable source black triangle area noise suppression processing method aims at the problem that the traditional transform domain filtering method cannot solve the "black triangle" noise, and provides a controllable source "black triangle" noise suppression method based on sparse inversion. By constructing models for the effective signal data of the original seismic reflection wave data and the black triangle area noise data respectively, the effective signal data of the seismic reflection wave data and the black triangle area noise data in the model space are separated. Through sparse inversion, sparse terms are obtained, and then the boundary of the black triangle area noise data is obtained, and the black triangle area noise prediction data is obtained thereby, and then the black triangle area noise prediction data is subtracted from the seismic reflection wave data by adaptive subtraction to obtain the effective signal wave data. Through the method based on sparse inversion, the black triangle area noise data can be predicted and calculated relatively accurately, and the separation of the black triangle area noise data and the effective signal data in the original seismic reflection wave data is better realized. The sparse terms are solved by sparse inversion, the constraint effect on the black triangle noise data is improved, the separation effect of the black triangle noise and the effective signal is improved, and the damage to the effective signal in the seismic reflection wave data is reduced, thereby improving the subsequent imaging resolution and accuracy.

[0093] In other embodiments, a method for suppressing the "black triangle" noise of a vibroseis source based on sparse inversion in the curvelet domain is provided. The method models the effective signal and the "black triangle" noise in the seismic data in different model space domains and constructs an objective function, and then solves the objective function through sparse inversion, thereby converting the "black triangle" noise suppression problem into a least squares solution problem of the objective function, and replacing the traditional transform domain filtering process with a multi-dimensional inversion process. The inversion results are constrained by adding sparse constraints and linear search steps, and the output of each iteration is used as feedback to enter the next iteration, so as to protect the effective signal to the greatest extent in the internal algorithm. The processing results of actual data show that this method can effectively suppress the "black triangle" noise in the vibroseis source seismic data. Compared with the traditional transform domain filtering suppression method, this method can protect the effective signal to the greatest extent and further improve the resolution of the vibroseis source seismic data.

[0094] Example 2

[0095] In another embodiment, a method for suppressing noise in a black triangle area of ​​a vibrator includes the following steps:

[0096] Step 1: Obtain original seismic reflection wave data.

[0097] Step 2: Calculate the effective reflected wave signal data d in the original seismic reflection wave data d s Construct model operator L s ; Among them, the original seismic reflection wave data d can be expressed as the reflection wave effective signal data d s Compared with the noise data in the “black triangle” area d bt The sum is:

[0098] d=d s +d bt

[0099] According to the hyperbolic NMO assumption, the reflected wave effective signal data can be further expressed as the model operator L s With reflectivity x S The product of s By the NMO correction operator N T and the superposition operator S T Composition, which can be specifically expressed as:

[0100] d s =L s x s =N T S T x s

[0101] Step 3: Convert the noise data d in the “black triangle” area in the original seismic reflection wave data d bt Convert to the curvelet domain to construct the model operator L c , realizing the separation of signal and noise in the model space domain; wherein, the noise data d in the “black triangle” area contained in the original seismic reflection wave data d bt Convert to the curvelet domain to construct the model operator L c , specifically expressed as:

[0102] d bt =L c x c

[0103] Among them, L c is the curvelet domain model operator of the noise data in the “black triangle” area, which is composed of curvelet domain coefficients, x c It is the sparse representation of the noise data in the “black triangle” area in the curvelet domain, that is, the sparse term.

[0104] Step 4: Construct the objective function in the curvelet domain and add a sparse constraint Δx to the “black triangle” noise data c, the problem of suppressing the “black triangle” noise is converted into the problem of solving the objective function; by constructing the objective function in the curvelet domain, the problem of suppressing the “black triangle” noise data can be converted into the problem of solving the objective function. The objective function J can be expressed as

[0105] J=||dL s x s -L c x c || 2 +μ||x c ||1

[0106] Among them, μ is the inversion regularization operator, ||x c ||1 means sparse constraint on the “black triangle” noise.

[0107] Step 5: Solve the objective function and obtain the sparse term x of the "black triangle" noise through sparse inversion prediction c ; Using the x obtained by the final iterative inversion c , directly obtain the noise data of the “black triangle” area, specifically:

[0108] d c =L c x c

[0109] Among them, d c This is the noise prediction data for the black triangle area.

[0110] Step 6: Use adaptive subtraction to subtract the predicted “black triangle” area noise prediction data d c Subtract it from the original seismic reflection wave data d to obtain the effective signal wave data d s1 , specifically:

[0111] d s1 =dd c

[0112] In this way, the purpose of suppressing the "black triangle" noise is achieved; then the suppression of the "black triangle" noise of the seismic data is achieved through adaptive subtraction.

[0113] In this embodiment, a method for suppressing the "black triangle" noise of a vibroseis source based on sparse inversion in the curvelet domain is provided to solve the problem that the traditional transform domain filtering method cannot solve the "black triangle" noise. The method models the effective signal and the "black triangle" noise in the seismic data in different model space domains and constructs an objective function, and then solves the objective function through sparse inversion, thereby converting the suppression problem of the "black triangle" noise into a least square solution problem of the objective function, and replacing the traditional transform domain filtering process with a multidimensional inversion process. The inversion result is constrained by adding sparse constraints and linear search steps, and the output of each iteration is used as feedback to enter the next iteration, so as to protect the effective signal to the greatest extent in the internal algorithm. The processing results of actual data show that this method can effectively suppress the "black triangle" noise in the vibroseis source seismic data. Compared with the traditional transform domain filtering suppression method, this method can protect the effective signal to the greatest extent and improve the resolution of the vibroseis source seismic data. The present invention provides a method for suppressing the "black triangle" noise of a vibroseis source based on sparse inversion in the curvelet domain to solve the problem that the traditional transform domain filtering method cannot solve the "black triangle" noise. This method models the effective signal and "black triangle" noise in the seismic data in different model space domains and constructs the objective function, and then solves the objective function through sparse inversion, thereby converting the "black triangle" noise suppression problem into the least squares solution problem of the objective function, and replacing the traditional transform domain filtering process with a multi-dimensional inversion process. The inversion results are constrained by adding sparse constraints and linear search steps, and the output of each iteration is used as feedback to enter the next iteration, so as to protect the effective signal to the greatest extent in the internal algorithm. The processing results of actual data show that this method can effectively suppress the "black triangle" noise in the vibroseis seismic data. Compared with the traditional transform domain filtering suppression method, this method can protect the effective signal to the greatest extent and improve the resolution of the vibroseis seismic data.

[0114] Example 3

[0115] The following is an application of the controllable vibrator black triangle area noise suppression processing method provided by the present application, and the method is verified using the controllable vibrator seismic data of the actual desert area.

[0116] The actual data of vibroseis from a work area in western China were selected to test the method. By analyzing the actual data in different frequency bands, it was found that ( Figures 3a to 3f ), the effective signal and the “black triangle” noise have different dispersion and distribution characteristics, so converting the data into the curvelet domain can separate the effective signal from the “black triangle” noise. Figures 3a to 3f This is the result of frequency division processing of single-shot data of vibroseis in desert area. Figure 3a This is the single shot data for the full frequency range of 1.5-90Hz. Figure 3b-3f These are the single-shot data of vibroseis in different frequency bands.

[0117] Select the shot collection data of one of the guns as follows Figure 4a As shown, it is used as input data to suppress the "black triangle" noise. The "black triangle" suppression effect and the removed noise are shown as follows: Figure 4b (valid signal) and Figure 4c (Noise in the black triangle area) is shown. Figure 4a is the input single shot data with “black triangle noise”, Figure 4b This is the single shot data after the “black triangle” noise is suppressed by this method. Figure 4c The suppressed “black triangle” noise. It can be seen that the strong amplitude noise in the “black triangle” area has been significantly suppressed, and the effective signal has also been well restored and reconstructed.

[0118] Further, we can see from the zero offset stacking section that Figure 5a and Figure 5b , Figure 5a This is an effective signal diagram of suppressing the "black triangle" noise through the traditional transform domain filtering method. The traditional transform domain filtering method suppresses the "black triangle" noise. Figure 5b This is an effective signal diagram for suppressing the "black triangle" noise based on the controllable seismic source black triangle area noise suppression processing method provided in this application. Figure 5a represents the stacked profile obtained by the traditional transform domain filtering method, Figure 5b It represents the stacked profile obtained by the noise suppression processing method of the black triangle area of ​​the vibroseis provided in this application. Figure 5a and 5b It can be seen that the "black triangle" noise suppression method based on curvelet domain sparse inversion in the present application is more thorough in suppressing the "black triangle" noise, and compared with the zero-offset profile generated by the input data, the "black triangle" noise suppression method based on curvelet domain sparse inversion can suppress strong energy noise while better restoring the effective signal, thereby increasing the contribution of the "black triangle" area and improving the imaging quality.

[0119] This method is purely data-driven and does not require prior information such as the underground medium structure and velocity model. Compared with the traditional noise suppression method based on transform domain filtering, this method can obtain denoised signal data through iterative inversion, effectively protecting the primary wave signal data and improving the fidelity of denoising.

[0120] Example 4

[0121] In one embodiment, the present application provides a controllable vibrator black triangle area noise suppression processing device, please refer to Figure 6The device comprises: a raw data acquisition module, a model construction module, a sparse inversion operation module, a noise prediction module and an effective signal wave data generation module. The raw data acquisition module, the model construction module, the sparse inversion operation module, the noise prediction module and the effective signal wave data generation module are connected in sequence.

[0122] The raw data acquisition module is used to acquire seismic reflection wave data;

[0123] The model building module is used to build models for the effective signal data of seismic reflection wave data and the noise data of the black triangle area respectively, so as to separate the effective signal data of seismic reflection wave data and the noise data of the black triangle area in the model space;

[0124] The sparse inversion operation module is used to perform sparse inversion on the black triangle area noise data based on the model of the effective signal data and the model of the black triangle area noise data to obtain a sparse term;

[0125] The noise prediction module is used to determine the noise prediction data of the black triangle area based on the sparse items;

[0126] The effective signal wave data generation module is used to subtract the black triangle area noise prediction data from the seismic reflection wave data by adaptive subtraction to obtain the effective signal wave data.

[0127] The above-mentioned controllable source black triangle area noise suppression processing device provides a controllable source "black triangle" noise suppression method based on sparse inversion to solve the problem that traditional transform domain filtering methods cannot solve the "black triangle" noise. By constructing models for the effective signal data of the original seismic reflection wave data and the black triangle area noise data respectively, the effective signal data of the seismic reflection wave data and the black triangle area noise data in the model space are separated. Through sparse inversion, sparse terms are obtained, and then the boundary of the black triangle area noise data is obtained, and the black triangle area noise prediction data is obtained thereby, so that the black triangle area noise prediction data is subtracted from the seismic reflection wave data by adaptive subtraction to obtain the effective signal wave data. Through the method based on sparse inversion, the black triangle area noise data can be predicted and calculated relatively accurately, and the separation of the black triangle area noise data and the effective signal data in the original seismic reflection wave data is better achieved. The sparse terms are solved by sparse inversion, so that the constraint effect on the black triangle noise data is improved, the separation effect of the black triangle noise and the effective signal is improved, and the damage to the effective signal in the seismic reflection wave data is reduced, so that the subsequent imaging resolution and accuracy can be improved.

[0128] In one embodiment, the model building module includes an effective signal model building unit and a noise model building unit;

[0129] The effective signal model building unit is used to construct the effective signal data d of the seismic reflection wave data d s Construct model operator L s ;

[0130] The noise model building unit is used to convert the black triangle noise data d of the seismic reflection wave data d bt Convert to the curvelet domain to construct the model operator L c , to achieve the separation of effective signal data of seismic reflection wave data and noise data in the black triangle area in the model space.

[0131] In one embodiment, the effective signal model building unit is used to construct the effective signal data d of the seismic reflection wave data d. s Construct model operator L s ; Among them, the model operator L s It is defined by the following formula:

[0132] d s =L s x s

[0133] L s =N T S T

[0134] Among them, x s is the reflectivity, N T is the dynamic correction operator, S T is the superposition operator.

[0135] Using the above model operator, according to the hyperbolic NMO assumption, the reflected wave effective signal data d s It can be expressed as the model operator L s With reflectivity x S The product of the model operator is composed of the moving average operator N T and the superposition operator S T Composition, thus we can get:

[0136] d s =L s x s =N T S T x s

[0137] In this way, by constructing a model operator for effective signal data, it is convenient to provide a boundary optimization basis for suppressing noise data in the black triangle area, thereby ensuring that the suppression effect of noise data in the black triangle area is improved and the interference of noise on effective signals is reduced.

[0138] In one embodiment, the noise model building unit is used to convert the black triangle noise data dbt Convert to the curvelet domain to construct the model operator L c In the model operator L c It is defined by the following formula:

[0139] d bt =L c x c

[0140] Among them, x c is the sparse term of the noise data in the black triangle area in the curvelet domain.

[0141] In this embodiment, the black triangle noise data d contained in the original seismic data d bt Convert to the curvelet domain to construct the model operator L c , so that the noise data d in the black triangle area bt It is able to obtain sparse solutions based on sparse inversion and transform the "black triangle" noise suppression problem into a least squares solution problem of the objective function. The multidimensional inversion process replaces the traditional transform domain filtering process, thereby effectively suppressing the "black triangle" noise in the vibroseis seismic data and further improving the resolution of the vibroseis seismic data.

[0142] In one embodiment, the sparse inversion operation module includes an objective function construction unit and an iterative inversion solution unit;

[0143] The objective function construction unit is used to construct an objective function for the black triangle area noise data based on the model of the effective signal data and the model of the black triangle area noise data in the curvelet domain, wherein in the objective function, a sparse constraint is added to the black triangle area noise data;

[0144] The iterative inversion solving unit is used to solve the objective function by adopting the final iterative inversion to obtain the sparse items of the noise data in the black triangle area in the curvelet domain.

[0145] In the present embodiment, by constructing an objective function, and adding a sparse constraint to the black triangle area noise data in the objective function, and using the final iterative inversion to solve the objective function, a sparse solution of the black triangle area noise in the curve domain, that is, a sparse term, is obtained. In the present embodiment, in the final iterative inversion, the inversion result is constrained by adding a sparse constraint and a linear search step, and the output of each iteration is used as feedback to enter the next iteration, and the effective signal data is protected to the greatest extent in the internal algorithm to ensure that the objective function obtains the optimal solution to obtain the sparse term. It should be noted that the specific process of the final iterative inversion refers to the prior art, and this application will not go into details here. Specifically, for example, the final iterative inversion is based on the iterative inversion solved by the least squares method.

[0146] In one embodiment, the objective function construction unit is used to construct an objective function for the black triangle area noise data in the curvelet domain, and the objective function J is expressed by the following formula:

[0147] J=||dL s x s -L c x c || 2 +μ||x c ||1

[0148] Among them, μ is the inversion regularization operator, ||x c ||1 is the sparse constraint on the noise data in the black triangle area.

[0149] In this way, the above-mentioned objective function is adopted, by modeling the effective signal and "black triangle" noise in the seismic data in different model space domains and constructing the objective function, and then solving the objective function through sparse inversion, so as to convert the suppression problem of the "black triangle" noise into the least squares solution problem of the objective function, and replace the traditional transform domain filtering process with a multi-dimensional inversion process. The inversion result is constrained by adding sparse constraints and linear search steps, and the output of each iteration is used as feedback to enter the next iteration, so as to protect the effective signal to the greatest extent in the internal algorithm. The processing results of actual data show that the "black triangle" noise in the controllable source seismic data can be effectively suppressed in this way. Compared with the traditional transform domain filtering suppression method, the present application can protect the effective signal to the greatest extent and further improve the resolution of the controllable source seismic data.

[0150] In one embodiment, the noise prediction module is used to determine the black triangle area noise prediction data according to the sparse items, wherein the black triangle area noise prediction data is determined using the following formula:

[0151] d c =L c x c

[0152] Among them, d c This is the noise prediction data for the black triangle area.

[0153] In this embodiment, based on the sparse solution obtained, the sparse term x c , thus through the model operator L c With the sparse term x cBy multiplying the black triangle noise prediction data, the noise in the black triangle area can be suppressed by the black triangle noise prediction data, and the effective signal data can be obtained after the black triangle noise prediction data is removed from the original seismic reflection wave data. This method can protect the effective signal to the greatest extent, minimize the impact of the black triangle noise removal on the effective signal, and further improve the resolution of the vibroseis seismic data.

[0154] The above-mentioned controllable source black triangle area noise suppression processing device provides a controllable source "black triangle" noise suppression method based on sparse inversion to solve the problem that traditional transform domain filtering methods cannot solve the "black triangle" noise. By constructing models for the effective signal data of the original seismic reflection wave data and the black triangle area noise data respectively, the effective signal data of the seismic reflection wave data and the black triangle area noise data in the model space are separated. Through sparse inversion, sparse terms are obtained, and then the boundary of the black triangle area noise data is obtained, and the black triangle area noise prediction data is obtained thereby, so that the black triangle area noise prediction data is subtracted from the seismic reflection wave data by adaptive subtraction to obtain the effective signal wave data. Through the method based on sparse inversion, the black triangle area noise data can be predicted and calculated relatively accurately, and the separation of the black triangle area noise data and the effective signal data in the original seismic reflection wave data is better achieved. The sparse terms are solved by sparse inversion, so that the constraint effect on the black triangle noise data is improved, the separation effect of the black triangle noise and the effective signal is improved, and the damage to the effective signal in the seismic reflection wave data is reduced, so that the subsequent imaging resolution and accuracy can be improved.

[0155] The specific definition of the controllable seismic source black triangle area noise suppression processing device can refer to the definition of the controllable seismic source black triangle area noise suppression processing method mentioned above, which will not be repeated here. Each module in the above-mentioned controllable seismic source black triangle area noise suppression processing device can be fully or partially implemented by software, hardware and their combination. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0156] Example 5

[0157] In a third aspect, the present application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method described in any of the above embodiments when executing the computer program.

[0158] In one embodiment, the internal structure diagram of the computer device can be as follows: Figure 7As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate through a network connection. When the computer program is executed by the processor, the steps of a controllable vibrator black triangle area noise suppression processing method in any of the above embodiments are implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0159] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0160] Example 6

[0161] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the method described in any of the above embodiments are implemented.

[0162] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0163] The present invention aims at the problem that the traditional transform domain filtering method cannot solve the "black triangle" noise, and provides a controllable source "black triangle" noise suppression method based on curvelet domain sparse inversion. The method models the effective signal and "black triangle" noise in the seismic data in different model space domains and constructs the objective function, and then solves the objective function through sparse inversion, so as to convert the "black triangle" noise suppression problem into the least square solution problem of the objective function, and replaces the traditional transform domain filtering process with a multi-dimensional inversion process. The inversion result is constrained by adding sparse constraints and linear search steps, and the output of each iteration is used as feedback to enter the next iteration, so as to protect the effective signal to the maximum extent in the internal algorithm. The processing results of actual data show that the method can effectively suppress the "black triangle" noise in the controllable source seismic data. Compared with the traditional transform domain filtering suppression method, the method can protect the effective signal to the maximum extent and improve the resolution of the controllable source seismic data.

[0164] The technical features of the above-described embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above-described embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification. It should be noted that "in one embodiment", "for example", "for example", etc. in this application are intended to illustrate the present application, rather than to limit the present application. The above-described embodiments only express several implementation methods of the present application, and the description is relatively specific and detailed, but it cannot be understood as a limitation on the scope of the patent application. It should be pointed out that for a person of ordinary skill in the art, without departing from the concept of the present application, several variations and improvements can also be made, which all belong to the scope of protection of the present application. Therefore, the scope of protection of the patent in this application shall be subject to the attached claims.

Claims

1. A method for suppressing noise in the black triangle area of ​​a vibrator, characterized in that: The steps include: Acquire seismic reflection wave data; Models are constructed for the effective signal data of seismic reflection wave data and the noise data of black triangle area respectively; Based on the model of the effective signal data and the model of the black triangle area noise data, sparse inversion is performed on the black triangle area noise data to obtain a sparse term; According to the sparse items, the noise prediction data of the black triangle area is determined; Adaptive subtraction is used to subtract the noise prediction data of the black triangle area from the seismic reflection wave data to obtain the effective signal wave data.

2. The method according to claim 1, characterized in that The steps of constructing models for the effective signal data of the seismic reflection wave data and the black triangle area noise data include: Effective signal data d for seismic reflection wave data d s Construct model operator L s ; The noise data d in the black triangle area of ​​the seismic reflection wave data d bt Convert to the curvelet domain to construct the model operator L c .

3. The method according to claim 2, characterized in that The effective signal data d for the seismic reflection wave data d s Construct model operator L s In the step, the model operator L s It is defined by the following formula: d s =L s x s L s =N T S T Among them, x s is the reflectivity, N T is the dynamic correction operator, S T is the superposition operator.

4. The method according to claim 3, characterized in that The black triangle noise data d of the seismic reflection wave data d bt Convert to the curvelet domain to construct the model operator L c In the step, the model operator L c It is defined by the following formula: d bt =L c x c Among them, x c is the sparse term of the noise data in the black triangle area in the curvelet domain.

5. The method according to claim 4, characterized in that The step of performing sparse inversion on the black triangle area noise data based on the model of the effective signal data and the model of the black triangle area noise data to obtain sparse items comprises: In the curvelet domain, an objective function is constructed for the black triangle area noise data based on the model of the effective signal data and the model of the black triangle area noise data, wherein in the objective function, a sparse constraint is added to the black triangle area noise data; The final iterative inversion is used to solve the objective function and obtain the sparse terms of the noise data in the black triangle area in the curvelet domain.

6. The method according to claim 5, characterized in that The objective function is constructed for the black triangle noise data in the curvelet domain, and the objective function J is expressed by the following formula: J=||dL s x s -L c x c || 2 +μ||x c ||1 Among them, μ is the inversion regularization operator, ||x c ||1 is the sparse constraint on the noise data in the black triangle area.

7. The method according to claim 6, characterized in that In the step of determining the noise prediction data of the black triangle area according to the sparse items, the noise prediction data of the black triangle area is determined by the following formula: d c =L c x c Among them, d c This is the noise prediction data for the black triangle area.

8. A controllable vibrator black triangle area noise suppression processing device, characterized in that: The device comprises: A raw data acquisition module, used to acquire seismic reflection wave data; A model building module is used to build models for effective signal data of seismic reflection wave data and noise data of black triangle area respectively; A sparse inversion operation module, used for performing sparse inversion on the black triangle area noise data based on the model of the effective signal data and the model of the black triangle area noise data to obtain a sparse term; A noise prediction module is used to determine the noise prediction data of the black triangle area according to the sparse items; The effective signal wave data generation module is used to subtract the black triangle area noise prediction data from the seismic reflection wave data by adaptive subtraction to obtain the effective signal wave data.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

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

Images