A method for attenuating coherent noise in 3D pre-stack seismic records

Abstract

The present invention discloses a method for attenuating coherent noise in three-dimensional prestack seismic records, comprising the following steps: 1) performing wavelet transform on each seismic record in a three-dimensional prestack seismic data volume to decompose the three-dimensional prestack seismic data volume into frequency-divided seismic data volumes with different wavelet orders; 2) determining the wavelet order W of the frequency-divided seismic data volume and the apparent velocity scanning range of the coherent noise in this wavelet order according to the frequency and apparent velocity of the coherent noise; 3) determining the direction of coherent noise for each sample point of each trace according to the apparent velocity scanning of the coherent noise; 4) predicting the amplitude value of the coherent noise for each sample point of each trace along the direction of coherent noise and performing attenuation processing on the coherent noise; 5) performing the same processing on each sample point within the processing time range of each shot and each trace to obtain the result after coherent noise attenuation, and performing wavelet transform reconstruction on the result after coherent noise attenuation to obtain the seismic record. The coherent noise attenuation method of the present invention provides good protection for effective signals.

Description

Technical Field

[0001] The present invention relates to oil exploration and development technologies, and in particular, to a method for attenuating coherent noise in three-dimensional pre-stack seismic records. Background Art

[0002] In seismic data collected under complex surface geological conditions in mountainous areas, there are often strong coherent noises. Due to the limitation of complex landforms, the arrangement is somewhat irregular. At the same time, due to the large lateral variation in the lithology of excitation and reception, the coherent noise varies greatly in space. These strong-energy coherent noises seriously affect the estimation of amplitude and wavelet spectrum in subsequent surface-consistent amplitude compensation and surface-consistent deconvolution, affect the extraction of static correction amounts and dynamic correction velocities, and thus affect the superposition imaging effect.

[0003] Conventional processing methods use the F-Kx-Ky technique for filtering after cross-shaped arrangement. However, the application conditions of this technique are: in the spatial direction, the energy and waveform of seismic waves are roughly the same, and at the same time, the spatial sampling conditions must also be met. But the actual situation often fails to meet these conditions: without performing surface-consistent processing to eliminate the factors of shot points and geophone points, the spatial amplitude and waveform differences are very large, and insufficient spatial sampling leads to relatively serious spatial aliasing, especially in the vertical arrangement direction (Y direction). In addition, the mat phenomenon existing in FK filtering still exists in F-Kx-Ky filtering. These adverse factors restrict its application effect. In addition, since the denoising technique uses a fan-shaped excision method, signals with the same frequency and wave number as the noise are completely lost. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method for attenuating coherent noise in three-dimensional pre-stack seismic records in view of the defects in the prior art.

[0005] The technical solution adopted by the present invention to solve its technical problems is: a method for attenuating coherent noise in three-dimensional pre-stack seismic records, including the following steps:

[0006] 1) Perform wavelet transform on each seismic record on the three-dimensional pre-stack seismic data volume, and decompose the three-dimensional pre-stack seismic data volume into frequency-divided seismic data volumes with different wavelet orders;

[0007] 2) According to the frequency and apparent velocity of the coherent noise, determine the wavelet order W of the frequency-divided seismic data volume and the apparent velocity scanning range (V1, V2) of the coherent noise in this wavelet order, and determine the processing time range (T s , T e ) of each seismic record in the frequency-divided seismic data volume through the apparent velocity scanning range;

[0008] 3) Determine the coherent noise direction of each sample point of each trace according to the apparent velocity scanning of the coherent noise;

[0009] 4) Predict the coherent noise amplitude value of each sample point of each trace along the coherent noise direction, and perform attenuation processing on the coherent noise;

[0010] 5) Perform the same processing on each sample point within the processing time range of each shot and each trace, obtain the result after coherent noise attenuation, and perform wavelet transform reconstruction on the seismic record for the result after coherent noise attenuation.

[0011] According to the above scheme, in step 2), the processing time range (T s , T e ) of each seismic record in the frequency-divided data volume is determined through the apparent velocity scanning range, specifically as follows:

[0012] The starting time of processing for each frequency-divided seismic record is,

[0013] T s = D / V2 (1)

[0014] If T s < T0, where T0 is the first arrival time of this trace, then T s = T0, where D is the shot-receiver offset of this trace;

[0015] The ending time of processing for each frequency-divided seismic record is,

[0016] T e = D / V1 (2)

[0017] If T e > T1, where T1 is the length of this trace, then T e = T1.

[0018] According to the above scheme, in step 3), the coherent noise direction of each sample point of each trace is determined according to the apparent velocity scanning of the coherent noise, specifically as follows:

[0019] Suppose 2N + 1 consecutive seismic records are s i (t), i = -N,..., 0,..., N. For a certain moment t0 of the central trace s0 of 2N + 1 traces, the objective function is defined as

[0020]

[0021] The corresponding apparent velocity V0 that makes this function reach the minimum value through apparent velocity scanning is the coherent noise direction,

[0022] V0 = argmin V∈[V1,V2] E(V) (4)

[0023] Among them,

[0024] ti = t0 + (D i - D0) / V (5)

[0025] D0 is the offset of the central trace, D i is the offset of the i-th trace, and V is the apparent velocity.

[0026] According to the above scheme, in step 4), the amplitude value of the coherent noise of each sample point of each trace is predicted as follows:

[0027] Let the seismic records of 2N + 1 consecutive traces be s i (t), i = -N,..., 0,..., N. For the amplitude value of the coherent noise at time t0 of the central trace s0 of 2N + 1 traces, it is calculated according to the following formula,

[0028]

[0029] where,

[0030] t i = t0 + (D i - D0) / V0 (7)

[0031] The predicted coherent noise is weighted by the objective function E(V0) and subtracted from the original record, that is, the result A0(t0) after coherent noise attenuation is obtained.

[0032] A0(t0) = S0(t0) - E(V0)N0(t0).

[0033] The beneficial effects produced by the present invention are:

[0034] A method for attenuating coherent noise in 3D pre-stack seismic records of the present invention restricts the time range and frequency range for coherent noise attenuation processing, and makes appropriate weight processing on the identified coherent noise during coherent noise attenuation, achieving the protection of effective signals; the method of the present invention can adapt to the irregular changes of offsets and the irregular changes of coherent noise caused by missing traces, thus having strong adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The present invention will be further described below in conjunction with the drawings and embodiments. In the drawings:

[0036] Figure 1 is the flowchart of the method of the embodiment of the present invention;

[0037] Figure 2 is the seismic data diagram before coherent noise attenuation of 3D pre-stack seismic records of the embodiment of the present invention;

[0038] Figure 3It is the seismic data map after the coherent noise attenuation of the 3D prestack seismic record in the embodiment of the present invention. Detailed implementation manners

[0039] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0040] As Figure 1 shown, a method for attenuating coherent noise of 3D prestack seismic records provided by an embodiment of the present invention mainly includes the following steps:

[0041] S1: Perform wavelet transform on each seismic record on the 3D prestack seismic data volume, and decompose the 3D prestack seismic data volume into frequency-divided seismic data volumes with different wavelet orders;

[0042] S2: Calculate the processing time range through the apparent velocity scanning range [V1, V2], calculate the starting time of each trace through V2, and calculate the ending time of each trace through V1.

[0043] In the embodiment of the present invention, step S2 can be implemented in the following manner:

[0044] Let D be the offset of this trace, then the starting time of this trace is,

[0045] T s = D / V2 (1)

[0046] If T s < T0 (T0 is the first arrival time of this trace), then T s = T0,

[0047] The ending time of processing is,

[0048] T e = D / V1 (2)

[0049] If T e > T1 (T1 is the length of this trace), then T e = T1.

[0050] S3: Determine the coherent noise direction through apparent velocity scanning.

[0051] In the embodiment of the present invention, step S3 can be implemented in the following manner:

[0052] Let 2N + 1 consecutive seismic records be s i (t), i = -N,..., 0,..., N. For a certain moment t0 of the central trace s0, determine its coherent noise direction (defined by the apparent velocity V0) as follows:

[0053] Define the objective function as

[0054]

[0055] The V0 that minimizes this function through apparent velocity scanning is the coherent noise direction,

[0056] V0 = argmin V∈[V1,V2] E(V) (4)

[0057] where,

[0058] t i = t0 + (D i - D0) / V (5)

[0059] D0 is the offset of the central trace, D i is the offset of the i-th trace, and V is the apparent velocity.

[0060] S4: Predict coherent noise along the coherent noise direction and perform attenuation.

[0061] In the embodiments of the present invention, step S4 can be implemented in the following manner:

[0062] Let the continuous 2N + 1 trace seismic record be s i (t), i = -N,..., 0,..., N. For the amplitude value N0(t0) of the coherent noise at the central trace s0 of the 2N + 1 traces at time t0, it is calculated according to the following formula,

[0063]

[0064] where,

[0065] t i = t0 + (D i - D0) / V0 (7)

[0066] Use the objective function E(V0) to weight the predicted coherent noise and subtract it from the original record, that is, the value A0(t0) at time t0 of the s0 trace after coherent noise attenuation is,

[0067] A0(t0) = S0(t0) - E(V0)N0(t0) (8)

[0068] Perform the same processing for each sample point within the processing time range of each shot and each trace, that is, the coherent noise attenuation result is obtained.

[0069] S5: Reconstruct the seismic record by wavelet transform.

[0070] Figure 2It is a seismic data map before the attenuation of coherent noise in 3D pre-stack seismic records. Due to the influence of non-normal offsets, the coherent noise shows a non-linear form, and conventional linear denoising techniques are difficult to work effectively. If no denoising treatment is performed, it will seriously affect subsequent wavelet processing, velocity analysis, and stacking effects.

[0071] Figure 3 It is a seismic data map after the attenuation of coherent noise in 3D pre-stack seismic records. Compared with Figure 2 the coherent noise is attenuated, and the seismic information hidden under the coherent noise also appears. The denoising effect is obvious.

[0072] In summary, a method for attenuating coherent noise in 3D pre-stack seismic records of the present invention mainly has two improvements: (1) By restricting the time-frequency range and setting a certain threshold limit when identifying coherent noise, the protection of effective signals is achieved; (2) It can adapt to the irregular changes of coherent noise caused by irregular changes in shot-receiver offsets and missing channels, thus having strong adaptability.

[0073] This application also provides a computer-readable storage medium, such as flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disc, server, App application store, etc., on which a computer program is stored. When the program is executed by a processor, the method for attenuating coherent noise in 3D pre-stack seismic records in the method embodiment is implemented.

[0074] It should be noted that according to the needs of implementation, the various steps / components described in this application can be split into more steps / components, or two or more steps / components or partial operations of steps / components can be combined into new steps / components to achieve the purpose of the present invention.

[0075] It should be understood that for those of ordinary skill in the art, improvements or transformations can be made according to the above description, and all such improvements and transformations should fall within the protection scope of the appended claims of the present invention.

Claims

1. A method for attenuating coherent noise in three-dimensional pre-stack seismic records, characterized in that, Including the following steps: 1) Perform wavelet transform on each seismic record in the 3D prestack seismic data volume, and decompose the 3D prestack seismic data volume into frequency-divided seismic data volumes with different wavelet orders; 2) Determine the wavelet order W of the frequency-divided seismic data volume and the apparent velocity scanning range (V1, V2) of the coherent noise in this wavelet order according to the frequency and apparent velocity of the coherent noise, and determine the processing time range (T s , T e ) of each seismic record in the frequency-divided seismic data volume through the apparent velocity scanning range; 3) Determine the coherent noise direction of each sample point of each trace according to the apparent velocity scanning of the coherent noise; Specifically as follows: Let the continuous 2N + 1 seismic records be s i (t), where i = -N,..., 0,..., N. For a certain moment t0 of the central trace s0 of the 2N + 1 traces, the objective function is defined as The corresponding apparent velocity V0 that minimizes the function through apparent velocity scanning is the direction of coherent noise, V0 = argmin V∈[V1,V2] E(V) Wherein, t i = t0 + (D i - D0) / V $D_0$ is the offset of the central trace, and $D$ i is the offset of the $i$-th trace, and $V$ is the apparent velocity; 4) Predict the coherent noise amplitude value of each sample point of each trace along the coherent noise direction, and perform attenuation processing on the coherent noise; Wherein, the prediction of the coherent noise amplitude value of each sample point of each trace is as follows: Let the continuous 2N + 1 seismic records be s i (t), where i = -N,..., 0,..., N. The amplitude value of the coherent noise at the central trace s0 at time t0 of the 2N + 1 traces is calculated according to the following formula: Wherein, t i = t0 + (D i - D0) / V Use the objective function E(V0) to weight the predicted coherent noise, and subtract it from the original record, that is, the result A0(t0) after coherent noise attenuation is obtained, A0(t0) = S0(t0) - E(V0)N0(t0); 5) Perform the same processing on each sample point within the processing time range of each shot and each trace, obtain the result after coherent noise attenuation, and perform wavelet transform reconstruction on the result after coherent noise attenuation to obtain the seismic record.

2. The method for attenuating coherent noise of three-dimensional pre-stack seismic records according to claim 1, wherein In step 2), the processing time range (T s , T e ) of each seismic record in the frequency-divided data volume is determined by the apparent velocity scanning range, specifically as follows: The starting time of the processing of each frequency-divided seismic record is, T s = D / V2 (1) If T s < T0, where T0 is the first arrival time of this trace, then T s = T0, where D is the offset of this trace; the processing end time of each frequency-divided seismic record is T e = D / V1 If T e > T1, where T1 is the length of this track, then T e = T1.

3. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, the steps of the 3D prestack seismic record coherent noise attenuation method according to any one of claims 1 or 2 are implemented.

4. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, the 3D prestack seismic record coherent noise attenuation method according to any one of claims 1 or 2 is implemented.

Images