Seismic signal recovery method, system, device, storage medium and program product

Abstract

The application discloses a seismic signal recovery method, system, device, storage medium and program product, and the method comprises the following steps: converting a noisy seismic record into a frequency space domain seismic record; based on the frequency space domain seismic record, constructing a target function of a signal prediction operator and solving to obtain the signal prediction operator; after filtering processing of the signal prediction operator, a prediction error operator is calculated and obtained; based on the prediction error operator and the frequency space domain seismic record, a frequency space domain seismic signal recovered from the noisy seismic record is obtained. The application can improve the denoising capability of the seismic record, and meanwhile, the fidelity of the seismic signal recovery is improved.

Description

Technical Field

[0001] This invention relates to the field of ejection oil and gas exploration technology, and in particular to seismic signal recovery methods, systems, equipment, storage media, and program products. Background Technology

[0002] Compared to random noise, seismic signals exhibit continuity and coherence in spatial direction. Based on the difference in spatial coherence between seismic signals and random noise, numerous methods for suppressing random noise in seismic records have been developed.

[0003] Among them, temporal-spatial predictive filtering and frequency-spatial predictive filtering are currently the most widely used random noise attenuation methods in industry. Although these methods have different mathematical algorithms, they essentially use spatial filtering for noise suppression and signal recovery. However, these methods share a common drawback: while suppressing noise, they also damage the seismic signal, reducing the accuracy of seismic signal recovery.

[0004] Therefore, there is an urgent need for a new method for seismic signal recovery that can recover effective seismic signals from noise-contaminated seismic records, avoid damage to the effective seismic signals during the denoising process, and improve the fidelity of the recovered seismic signals. Summary of the Invention

[0005] The purpose of this invention is to provide a method, system, device, storage medium, and program product for seismic signal recovery, which can improve the denoising capability of seismic records and improve the fidelity of seismic signal recovery.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] In a first aspect, embodiments of the present invention provide a seismic signal recovery method, comprising:

[0008] Convert noisy seismic records into frequency spatial domain seismic records;

[0009] The objective function of the signal prediction operator is constructed based on the frequency spatial domain seismic records, and the signal prediction operator is obtained by solving the objective function of the signal prediction operator.

[0010] After filtering the signal prediction operator, the prediction error operator is calculated.

[0011] Based on the prediction error operator and the frequency spatial domain seismic record, the frequency spatial domain seismic signal recovered from the noisy seismic record is obtained.

[0012] Secondly, embodiments of the present invention provide an earthquake signal recovery system, comprising:

[0013] The first conversion unit is used to convert noisy seismic records into frequency spatial domain seismic records.

[0014] The solution unit is used to construct the objective function of the signal prediction operator based on frequency spatial domain seismic records and solve for the signal prediction operator.

[0015] The calculation unit is used to calculate the prediction error operator after filtering the signal prediction operator;

[0016] The recovery unit is used to obtain the frequency spatial domain seismic signal recovered from the noisy seismic record based on the prediction error operator and the frequency spatial domain seismic record.

[0017] Thirdly, embodiments of the present invention also provide an electronic device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program or instructions to implement the aforementioned seismic signal recovery method.

[0018] Fourthly, embodiments of the present invention also provide a computer storage medium storing a computer program or instructions, which, when executed by a processor, implement the aforementioned seismic signal recovery method.

[0019] Fifthly, embodiments of the present invention also provide a computer program product, including a computer program or instructions, which, when executed by a processor, implement the aforementioned seismic signal recovery method.

[0020] The technical effects and advantages of this invention are as follows: This invention no longer uses spatial filtering for noise attenuation and signal recovery. Instead, it uses the spatial coherence of the seismic signal as prior information and directly obtains the seismic signal from the seismic record using Bayesian inversion under the constraint of this prior information. Furthermore, compared with the spatial filtering-based methods in the prior art, this invention not only has stronger denoising capabilities but also improves the fidelity of the recovered seismic signal.

[0021] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures pointed out in the description, claims and drawings. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a flowchart of a seismic signal recovery method according to an embodiment of the present invention;

[0024] Figure 2 This is a schematic diagram of the synthesized seismic record in an embodiment of the present invention;

[0025] Figure 3 This is a schematic diagram of a synthetic seismic record containing random noise in an embodiment of the present invention;

[0026] Figure 4 A schematic diagram of a seismic signal recovered using conventional methods;

[0027] Figure 5 A schematic diagram illustrating random noise removal using conventional methods;

[0028] Figure 6 This is a schematic diagram of the recovered seismic signal in an embodiment of the present invention. Figure 1 ;

[0029] Figure 7 This is a schematic diagram of the random noise removal in an embodiment of the present invention;

[0030] Figure 8 This is a schematic diagram of pre-stack seismic records for a certain exploration block in an embodiment of the present invention;

[0031] Figure 9 This is a schematic diagram of the recovered seismic signal in an embodiment of the present invention. Figure 2 ;

[0032] Figure 10 This is a schematic diagram of the structure of an earthquake signal recovery system according to an embodiment of the present invention;

[0033] Figure 11 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0034] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0035] To address the shortcomings of existing technologies, this invention, using a specific exploration block as an example, discloses a seismic signal recovery method, such as... Figure 1 As shown, it includes the following steps:

[0036] Step S1: Convert the noisy seismic records into frequency spatial domain seismic records;

[0037] Step S2: Based on frequency spatial domain seismic records, construct the objective function of the signal prediction operator and solve for it to obtain the signal prediction operator;

[0038] Step S3: After filtering the signal prediction operator, calculate the prediction error operator.

[0039] Step S4: Based on the prediction error operator and the frequency spatial domain seismic record, obtain the frequency spatial domain seismic signal recovered from the noisy seismic record.

[0040] In some specific embodiments, step S1; converting noisy seismic records into frequency spatial domain seismic records includes:

[0041] like Figure 2 As shown, Figure 2 This is a schematic diagram of a synthetic seismic record. Random noise is added to the synthetic seismic record to generate a signal, such as... Figure 3 The image shows a noisy earthquake record.

[0042] The noisy seismic record is represented as d(ix,it); where it represents the time sampling number of the noisy seismic record, it∈(1,2,...,m), m represents the number of time samples, m=500; ix represents the seismic trace number of the seismic record, ix∈(1,2,...,n), n represents the number of seismic traces, n=201.

[0043] Perform a Fourier transform on the noisy seismic record d(ix,it) to obtain the frequency spatial domain seismic record d(ix,iω); where iω represents the frequency sampling number of the frequency spatial domain seismic record, iω∈(1,2,...,m), and m represents the number of frequency samples, m=500.

[0044] In some specific embodiments, step S2; based on frequency spatial domain seismic records, constructing the objective function of the signal prediction operator and solving for it to obtain the signal prediction operator includes:

[0045] Based on the spatial predictability of seismic signals and the frequency spatial domain seismic record d(ix,iω), an autoregressive model of the frequency spatial domain seismic record is established, that is, an autoregressive model of the seismic signal in the spatial frequency domain is established.

[0046] The autoregressive model is as follows:

[0047]

[0048] In the formula, d(ix,iω) represents the frequency spatial domain seismic record, that is, the frequency spatial domain seismic record value of the ixth seismic trace and the iωth frequency component; ix represents the seismic trace number of the seismic record, ix∈(1,2,...,n), and n represents the number of seismic traces; iω represents the frequency sampling number of the frequency spatial domain seismic record, iω∈(1,2,...,m), and m represents the number of frequency samples; h(jx,iω) represents the signal prediction operator, that is, the signal prediction operator value of the jxth spatial location and the iωth frequency component; jx represents the spatial sampling number, jx∈(1,2,...,l), and l represents the spatial length of the signal prediction operator;

[0049] d(ix-jx,iω) represents the frequency spatial domain seismic record, that is, the frequency spatial domain seismic record value of the i-th spatial location and the i-th frequency component.

[0050] The physical meaning of the autoregressive model is that a frequency spatial domain seismic record at a certain location can be linearly predicted using the preceding l frequency spatial domain seismic records.

[0051] Based on the aforementioned autoregressive model, the objective function of the signal prediction operator is established as follows:

[0052]

[0053] In the formula, e(iω) represents the objective function of the signal prediction operator; ix represents the seismic trace number of the seismic record, ix∈(1,2,...,n), and n represents the number of seismic traces; d(ix,iω) represents the frequency spatial domain seismic record; iω represents the frequency sampling number of the frequency spatial domain seismic record, iω∈(1,2,...,m), and m represents the number of frequency samples; jx represents the spatial sampling number, jx∈(1,2,...,l), l represents the spatial length (i.e., the length of the spatial direction) of the signal prediction operator, l=7; h(jx,iω) represents the signal prediction operator, λ1 represents the regularization coefficient, λ1=0.001; d(ix-jx,iω) represents the frequency spatial domain seismic record, that is, the frequency spatial domain seismic record value of the iω-th frequency component at the i-th spatial location.

[0054] The objective function of the signal prediction operator is solved using the conjugate gradient method, and the signal prediction operator h(jx,iω) is obtained.

[0055] In some specific embodiments, step S3; after filtering the signal prediction operator, calculating the prediction error operator includes:

[0056] Based on the signal prediction operator h(jx,iω), the objective function of the autoregressive filter is established as follows:

[0057]

[0058] In the formula, δ(jx) represents the objective function of the autoregressive filter; iω represents the frequency sampling number of the frequency spatial domain seismic record, iω∈(1,2,...,m), where m represents the number of frequency samples; h(jx,iω) represents the signal prediction operator, jx represents the spatial sampling number; jω represents the frequency sampling number of the autoregressive filter, jω∈(1,2,...,k), where k represents the length of the autoregressive filter p(jx,jω) in the frequency direction, k=9; p(jx,jω) represents the autoregressive filter, i.e., the autoregressive filter value at the jx-th spatial location and the iω-th frequency component; h(jx,iω-jω) represents the signal prediction operator, i.e., the signal prediction operator value at the jx-th spatial location and the iω-jω-th frequency component; λ2 represents the regularization coefficient, λ2=0.003.

[0059] The objective function of the autoregressive filter is solved using the conjugate gradient method to obtain the autoregressive filter p(jx,jω).

[0060] To eliminate the impact of noise interference on the signal prediction operator, an autoregressive filter p(jx,jω) is used to perform autoregressive filtering on the signal prediction operator h(jx,iω) to obtain the filtered signal prediction operator h'(jx,iω). The filtering of the signal prediction operator is performed using the following formula:

[0061]

[0062] In the formula, h'(jx,iω) represents the filtered signal prediction operator, that is, the value of the filtered signal prediction operator at the jx-th spatial position and the iω-th frequency component; jx represents the spatial sampling number, jx∈(1,2,...,l), and l represents the spatial length (i.e., the length in the spatial direction) of the signal prediction operator; iω represents the frequency sampling number of the frequency spatial domain seismic record, iω∈(1,2,...,m), and m represents the number of frequency samples; jω represents the frequency sampling number of the autoregressive filter, jω∈(1,2,...,k), and k represents the length of the autoregressive filter p(jx,jω) in the frequency direction; p(jx,jω) represents the autoregressive filter, jx represents the spatial sampling number; and h(jx,iω-jω) represents the signal prediction operator.

[0063] Since the signal prediction operator h(jx,iω) describes the spatial predictability of seismic signals, the prediction error operator can be calculated using the filtered signal prediction operator as follows:

[0064]

[0065] In the formula, q(jx,iω) represents the prediction error operator, that is, the prediction error operator value of the jx-th spatial location and the iω-th frequency component; jx represents the spatial sampling number, jx∈(1,2,...,l), and l represents the spatial length (i.e., the length of the spatial direction) of the signal prediction operator; iω represents the frequency sampling number of the frequency spatial domain seismic record, iω∈(1,2,...,m), and m represents the number of frequency samples; h'(jx,iω) represents the filtered signal prediction operator.

[0066] In some specific embodiments, step S4; obtaining the frequency spatial domain seismic signal recovered from the noisy seismic record based on the prediction error operator and the frequency spatial domain seismic record, includes:

[0067] The prediction error operator is introduced as prior information into the regularization condition for seismic signal recovery, and combined with the frequency spatial domain seismic record, the objective function for the frequency spatial domain seismic signal to be recovered is established as follows:

[0068]

[0069] In the formula, σ(iω) represents the objective function of the frequency spatial domain seismic signal to be recovered, ix represents the seismic trace number, ix∈(1,2,...,n), and n represents the number of seismic traces; d(ix,iω) represents the frequency spatial domain seismic record; s(ix,iω) represents the frequency spatial domain seismic signal to be recovered, that is, the frequency spatial domain seismic signal value to be recovered for the ixth seismic trace and the iωth frequency component; iω represents the frequency sampling sequence number of the frequency spatial domain seismic record, iω∈(1,2,...,m), and m represents the number of frequency samples; λ3 represents the regularization coefficient, λ3=0.05; jx represents the spatial sampling sequence number; q(jx,iω) represents the prediction error operator; l represents the spatial length of the autoregressive filter; s(ix-jx,iω) represents the frequency spatial domain seismic signal to be recovered, that is, the frequency spatial domain seismic signal value to be recovered for the ix-jxth seismic trace and the iωth frequency component.

[0070] The objective function of the frequency spatial domain seismic signal to be recovered is solved to obtain the frequency spatial domain seismic signal s(ix,iω) recovered from the noisy seismic record.

[0071] Then, the recovered frequency spatial domain seismic signal s(ix,iω) is transformed to obtain the time spatial domain seismic signal s(ix,it), which is then output, including:

[0072] The recovered frequency spatial domain seismic signal s(ix,iω) is subjected to an inverse Fourier transform to obtain the time spatial domain seismic signal s(ix,it). Here, it represents the time sampling sequence number of the seismic record, it∈(1,2,...,m), and m represents the number of time samples; ix represents the seismic trace sequence number of the seismic record, ix∈(1,2,...,n), and n represents the number of seismic traces; iω represents the frequency sampling sequence number of the frequency spatial domain seismic record, iω∈(1,2,...,m), and m represents the number of frequency samples, m=500.

[0073] like Figure 4-7 As shown, Figure 4 The seismic signal is recovered after noise attenuation using existing conventional methods. Figure 5 Random noise removed by existing conventional methods, Figure 6 The seismic signal recovered by the method of this invention, Figure 7 The random noise removed by the method of this invention is shown in the figure. Through observation and comparison, it can be found that the seismic signal recovered by conventional methods still contains noise interference with a certain amount of energy, and the noise interference contains the shadow of the seismic signal. However, the seismic signal recovered by the method of this invention has almost no residual noise interference, and the removed random noise does not contain the shadow of the seismic signal. Therefore, the technical solution provided by this invention has a more obvious effect.

[0074] Furthermore, this invention has conducted practical operational experiments using a specific exploration block as an example, such as... Figure 8 and Figure 9 As shown; Figure 8 This is the pre-stack seismic record of the exploration block, which contains strong noise interference. Figure 9 It is using the method of the present invention from Figure 8 The recovered seismic signals were observed. Figure 9 As can be seen, the present invention recovers the seismic signal with high fidelity from noisy seismic records.

[0075] The seismic signal was recovered with high fidelity from the noise-affected seismic records.

[0076] This invention discloses a seismic signal recovery system, such as... Figure 10 As shown, it includes:

[0077] The first conversion unit is used to convert noisy seismic records into frequency spatial domain seismic records.

[0078] The solution unit is used to construct the objective function of the signal prediction operator based on frequency spatial domain seismic records and solve for the signal prediction operator.

[0079] The calculation unit is used to calculate the prediction error operator after filtering the signal prediction operator;

[0080] The recovery unit is used to obtain the seismic signal recovered from the noisy seismic record based on the prediction error operator and the frequency spatial domain seismic record.

[0081] Regarding the system in the above embodiments, the specific manner in which each unit module performs operations has been described in detail in the embodiments related to the method, and will not be elaborated here.

[0082] Based on the same inventive concept, embodiments of the present invention also provide an electronic device, the structure of which is as follows: Figure 11 As shown, it includes a memory, a processor, and a computer program stored in the memory. The processor executes the computer program or instructions to implement the aforementioned seismic signal recovery method.

[0083] Based on the same inventive concept, embodiments of the present invention also provide a computer storage medium storing a computer program or instructions, which, when executed by a processor, implements the aforementioned seismic signal recovery method.

[0084] Based on the same inventive concept, embodiments of the present invention also provide a computer program product, including a computer program or instructions, which, when executed by a processor, implement the aforementioned seismic signal recovery method.

[0085] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for seismic signal recovery, characterized in that, include: Convert noisy seismic records into frequency spatial domain seismic records; Based on frequency spatial domain seismic records, an objective function for a signal prediction operator is constructed and solved to obtain the signal prediction operator. After filtering the signal prediction operator, the prediction error operator is calculated. Based on the prediction error operator and the frequency spatial domain seismic record, the frequency spatial domain seismic signal recovered from the noisy seismic record is obtained.

2. The seismic signal recovery method according to claim 1, characterized in that, Converting noisy seismic records into frequency spatial domain seismic records includes: The noisy seismic record is represented as d(ix,it); where it represents the time sampling number of the seismic record, it∈(1,2,...,m), m represents the number of time samples, ix represents the seismic trace number of the seismic record, ix∈(1,2,...,n), and n represents the number of seismic traces; Perform a Fourier transform on the noisy seismic record d(ix,it) to obtain the frequency spatial domain seismic record d(ix,iω); where iω represents the frequency sampling number of the frequency spatial domain seismic record, iω∈(1,2,...,m), and m represents the number of frequency samples.

3. The seismic signal recovery method according to claim 1, characterized in that, Based on frequency spatial domain seismic records, an objective function for a signal prediction operator is constructed and solved to obtain the signal prediction operator, including: Based on the spatial predictability of seismic signals and the frequency spatial domain seismic record d(ix,iω), an autoregressive model of the frequency spatial domain seismic record is established, that is, an autoregressive model of the seismic signal in the spatial frequency domain is established. Based on the autoregressive model, an objective function for the signal prediction operator is established. The objective function of the signal prediction operator is solved using the conjugate gradient method, and the signal prediction operator h(jx,iω) is obtained.

4. The seismic signal recovery method according to claim 3, characterized in that, The autoregressive model is as follows: In the formula, d(ix,iω) represents the frequency spatial domain seismic record; ix represents the seismic trace number, ix∈(1,2,...,n), and n represents the number of seismic traces; iω represents the frequency sampling number of the frequency spatial domain seismic record, iω∈(1,2,...,m), and m represents the number of frequency samples; h(jx,iω) represents the signal prediction operator, jx represents the spatial sampling number, jx∈(1,2,...,l), and l represents the spatial length of the signal prediction operator; d(ix-jx,iω) represents the frequency spatial domain seismic record. The objective function of the signal prediction operator is: In the formula, e(iω) represents the objective function of the signal prediction operator, and λ1 represents the regularization coefficient.

5. The seismic signal recovery method according to claim 1, characterized in that, Filtering the signal prediction operator includes: Based on the signal prediction operator h(jx,iω), the objective function of the autoregressive filter is established; The objective function of the autoregressive filter is solved using the conjugate gradient method to obtain the autoregressive filter p(jx,jω); The signal prediction operator h(jx,iω) is subjected to autoregressive filtering using an autoregressive filter p(jx,jω) to obtain the filtered signal prediction operator h'(jx,iω).

6. The seismic signal recovery method according to claim 5, characterized in that, The objective function of the autoregressive filter is: In the formula, δ(jx) represents the objective function of the autoregressive filter, iω represents the frequency sampling number of the frequency spatial domain seismic record, iω∈(1,2,...,m), and m represents the number of frequency samples; h(jx,iω) represents the signal prediction operator, jx represents the spatial sampling number; k represents the length of the autoregressive filter p(jx,jω) in the frequency direction, jω represents the frequency sampling number of the autoregressive filter, jω∈(1,2,...,k); jx represents the spatial sampling number, p(jx,jω) represents the autoregressive filter, h(jx,iω-jω) is the signal prediction operator, and λ2 represents the regularization coefficient; The signal prediction operator is filtered using the following formula: In the formula, h'(jx,iω) represents the filtered signal prediction operator.

7. A seismic signal recovery method according to claim 1 or 5, characterized in that, After filtering the signal prediction operator, the prediction error operator is calculated, including: Using the filtered signal prediction operator, the prediction error operator is calculated as follows: In the formula, q(jx,iω) represents the prediction error operator, jx represents the spatial sampling number, iω represents the frequency sampling number, iω∈(1,2,...,m), and m represents the number of frequency samples; h'(jx,iω) represents the filtered signal prediction operator.

8. The seismic signal recovery method according to claim 1, characterized in that, Based on the prediction error operator and the frequency spatial domain seismic record, the frequency spatial domain seismic signal recovered from the noisy seismic record is obtained, including: The prediction error operator is introduced as prior information into the regularization condition for seismic signal recovery, and combined with the frequency spatial domain seismic record, an objective function for the frequency spatial domain seismic signal to be recovered is established. The objective function of the frequency spatial domain seismic signal to be recovered is solved to obtain the frequency spatial domain seismic signal recovered from the noisy seismic record; The objective function of the frequency spatial domain seismic signal to be recovered is: In the formula, σ(iω) represents the objective function of the frequency spatial domain seismic signal to be recovered, ix represents the seismic trace number of the seismic record, ix∈(1,2,...,n), and n represents the number of seismic traces; d(ix,iω) represents the frequency spatial domain seismic record value, s(ix,iω) represents the frequency spatial domain seismic signal to be recovered; iω represents the frequency sampling sequence number of the frequency spatial domain seismic record, iω∈(1,2,...,m), and m represents the number of frequency samples; λ3 represents the regularization coefficient, jx represents the spatial sampling sequence number, q(jx,iω) represents the prediction error operator of the jx-th spatial and iω-th frequency components, l represents the spatial length of the autoregressive filter; and s(ix-jx,iω) represents the frequency spatial domain seismic signal to be recovered.

9. A seismic signal recovery system, characterized in that, include: The first conversion unit is used to convert noisy seismic records into frequency spatial domain seismic records. The solution unit is used to construct the objective function of the signal prediction operator based on frequency spatial domain seismic records and solve for the signal prediction operator. The calculation unit is used to calculate the prediction error operator after filtering the signal prediction operator; The recovery unit is used to obtain the frequency spatial domain seismic signal recovered from the noisy seismic record based on the prediction error operator and the frequency spatial domain seismic record.

10. An electronic device, characterized in that, The device includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program or instructions to implement a seismic signal recovery method according to any one of claims 1-8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program or instructions, which, when executed by a processor, implement the seismic signal recovery method according to any one of claims 1-8.

12. A computer program product, comprising a computer program or instructions, characterized in that, When the computer program or instructions are executed by the processor, they implement the earthquake signal recovery method according to any one of claims 1-8.

Images