Black triangle noise attenuation method based on vibroseis force signal

Abstract

The invention provides a black triangle noise attenuation method based on a vibroseis force signal, relates to the field of seismic exploration, and aims to solve the problems that the black triangle noise of a vibroseis on a loose earth surface is serious and the indoor denoising difficulty is large, the noise attenuation method adopts different denoising modes to remove the noise of different causes through the application of the vibroseis force signal, and the noise attenuation efficiency is improved. And the black triangle noise suppression of the vibroseis can be realized.

Description

Technical Field

[0001] The present invention relates to the field of seismic exploration, and in particular to a "black triangle" noise attenuation method based on controllable seismic source force signals. Background Art

[0002] Affected by the excitation conditions, the vibroseis data on loose surface develops obvious "black triangle" noise. The noise has the characteristics of strong energy, large range, wide bandwidth, and no linear regularity, which seriously affects the signal-to-noise ratio of seismic data. Conventional noise suppression technology is difficult to effectively remove these noises, which brings great difficulties to the subsequent interpretation and inversion of seismic data.

[0003] In response to the strong energy noise of seismic data, geophysicists have proposed a variety of noise suppression methods. Most of these methods are based on filtering processing based on wave field characteristics. Yilmaz et al. used a high-pass filtering method based on low-frequency characteristics, Karsli et al. used Wiener filter and FK filtering technology to suppress noise, Deighan et al. used the characteristics of time-frequency analysis of wavelet transform to perform wavelet transform on seismic traces and filter low-frequency areas to suppress noise; in 2019, Liu Xiheng et al. proposed to use seismic data feature extraction technology to obtain signal and noise feature information, and finally use this feature information to remove noise. In 2021, Wang Lixin et al. developed a controllable source scattered surface wave interference prediction and matching subtraction technology based on the gun-detection seismic interference and pattern matching subtraction method.

[0004] Compared with explosive sources, vibroseis acquisition has the advantages of "safety, environmental protection, high efficiency and economy", and its excitation parameters are controllable. Vibroseis acquisition will become the development trend of seismic exploration in the future. However, vibroseis seismic data still have many deficiencies in noise interference, frequency band and phase. The nonlinear strong interference noise generated by vibroseis excitation is manifested as full-band "black triangle" strong energy noise on the data set, especially in loose surface areas, where the noise energy is stronger and the signal-to-noise ratio of the data is lower. At present, the suppression of "black triangle" noise is based on the analysis of wave field characteristics and is achieved by filtering. This method will cause certain damage to the effective signal.

[0005] In view of the problems existing in the existing technology, it is necessary to find a "black triangle" noise attenuation method based on controllable seismic source force signal. Summary of the invention

[0006] In view of the problems existing in the prior art, the present invention provides a "black triangle" noise attenuation method based on a controllable seismic source force signal. The method applies the controllable seismic source force signal and adopts different denoising methods to remove noise of different causes, thereby achieving the suppression of the "black triangle" noise of the controllable seismic source.

[0007] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0008] The present invention provides a noise attenuation method, comprising the following steps:

[0009] 1) Collect the original relevant records of vibroseis seismic data, and first perform vibroseis excitation sub-wave ω on the original relevant records. 0 (t) conjugate function convolution, and then divided by ω 0 (t) modulo, get the parent record;

[0010] 2) Calculate the average amplitude of the seismic trace and remove the mechanical noise E'(t) generated by the vibrator excitation in the mother record with an amplitude value more than ten times the average amplitude;

[0011] 3) Continue to remove the vibroseis noise similar to the well-shot excitation noise in the vibroseis mother record to obtain the mother record that only contains the influence of the mechanical system and near-surface factors;

[0012] 4) The distortion factor of the mechanical and hydraulic system of the vibrator is denoted by p 0 (t), the correlation factor near the surface is denoted as p 1 (t), respectively find p 0 (t), p 1 The conjugate function of (t)

[0013] 5) Combine the records including the mechanical system and near-surface factors obtained in step 3) with the records obtained in step 4) Convolution, then divide by p 0 (t) and p 1 The modulus of (t) is used to obtain the denoised mother record;

[0014] 6) Convolve the denoised mother record with the vibrator excitation wavelet to obtain the vibrator related record and achieve "black triangle" noise attenuation.

[0015] Furthermore, the calculation formula of the parent record in step 1) is as follows:

[0016] S1(t)=ω 0 (t)*e(t)*p 0 (t)*p 1 (t)+n′(t) Formula (1);

[0017] Among them, S1(t) is the parent record; ω 0 (t) represents the sub-wave excited by the vibrator, e(t) represents the formation impulse response, and p 0 (t) is the distortion factor of the mechanical and hydraulic system of the vibrator, p 1(t) is the correlation factor near the surface, and n′(t) is composed of the controllable source noise similar to the well gun excitation noise and the mechanical noise generated by the controllable source excitation.

[0018] Furthermore, the calculation formula for the record including the influence of the mechanical system and the near-surface factors in step 3) is as follows:

[0019] S2(t)=ω 0 (t)*e(t)*p 0 (t)*p 1 (t) Formula (2);

[0020] Among them, S2(t) is the parent record that only includes the influence of mechanical system and near-surface factors.

[0021] Furthermore, the distortion factor in step 4) is obtained by recording the vibration plate signal and the trigger signal of the controllable vibrator.

[0022] Specifically, the distortion factor in step 4) is obtained as follows: the trigger signal is the sub-wave ω excited by the vibrator 0 (t), the vibration plate signal of the controllable vibrator signal is the controllable vibrator force signal ω recorded in the field 0 ′(t), using the vibrator force signal ω recorded in the field 0 ′(t) and the sub-wave ω excited by the vibrator 0 (t) conjugate function convolution, and then divided by ω 0 (t), we get the distortion factor p 0 (t).

[0023] Furthermore, the near-surface correlation factors in step 4) are obtained by investigating the surface Q value.

[0024] Specifically, the method for obtaining the near-surface correlation factor in step 4) is as follows: using the record ω obtained from the surface Q value survey 0 ″(t) and force signal ω 0 The conjugate function convolution of ′(t) and then divided by ω 0 The modulus of ′(t) can be used to obtain the correlation factor p near the surface. 1 (t).

[0025] Among them, the surface Q value, that is, the surface quality factor, can reflect the quality factor of the physical characteristics of the surface layer's earthquake absorption and attenuation.

[0026] Furthermore, the calculation of the denoised mother record in step 5) includes the following steps:

[0027] Combine formula (2) with Perform convolution and get:

[0028]

[0029] That is ω 0 (t)*e(t)×[|p 0 (t)| 2 ×|p 1 (t)| 1 ] Formula (4);

[0030] Divide equation (4) by p 0 (t) and p 1 The modulus of (t) yields: S3(t) = ω 0 (t)*e(t) Formula (5);

[0031] Among them, S3(t) is the mother record after denoising.

[0032] Furthermore, the calculation formula for convolving the denoised mother record with the vibroseis excitation wavelet in step 6) is as follows:

[0033] S4(t)=ω 0 (t)*e(t)*ω 0 (t) Formula (6);

[0034] Among them, S4(t) is the relevant record of the controllable source after denoising.

[0035] Furthermore, n′(t)=n′ 1 (t)+n′ 2 (t)+……n′ N (t)+E'(t);

[0036] Among them, n′ 1 (t)+n′ 2 (t)+……n′ N (t) is the controllable source noise similar to the well gun excitation noise, E′(t) is the mechanical noise generated by the controllable source excitation; where n′ 1 (t)~n′ N (t) includes one or more of surface wave noise, multiple wave noise, linear interference noise, random noise, and environmental noise, and N is a positive integer.

[0037] The technical effects achieved by the present invention are:

[0038] The present invention proposes a new idea for suppressing the "black triangle" noise. Starting from the cause of the "black triangle" noise of the controllable source, this method makes full use of the force signal to suppress the noise in the parent record. This study divides the noise in the controllable source record into three categories: mechanical noise of the controllable source, controllable source noise similar to well gun excitation, and controllable source related noise generated by mechanical and hydraulic systems and near-surface. The mechanical noise is attenuated by removing the maximum value, the well gun denoising method is used to remove the noise similar to the well gun, and the conjugate function of the mechanical distortion factor and the near-surface correlation factor is used to suppress the controllable source related noise, thus realizing the suppression of the "black triangle" noise of the controllable source. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is the original relevant record of vibroseis seismic data;

[0040] Figure 2 is the original parent record;

[0041] Figure 3 To remove the strong amplitude mechanical noise contained in the record, (a) is before removing the mechanical noise, (b) is after removing the mechanical noise, and (c) is the removed mechanical noise;

[0042] Figure 4 To remove the surface waves contained in the record, (a) is before removing the surface waves, (b) is after removing the surface waves, and (c) is the removed surface waves;

[0043] Figure 5 To excite wavelets for vibrator;

[0044] Figure 6 It is the vibrator force signal recorded in the field;

[0045] Figure 7 This is a schematic diagram of the surface Q value survey;

[0046] Figure 8 Comparison of data before and after the “black triangle” noise attenuation, where (a) is the superimposed record before attenuation and (b) is the superimposed record after attenuation. DETAILED DESCRIPTION

[0047] The following describes the embodiments of the present invention through specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention.

[0048] Before further describing the specific embodiments of the present invention, it should be understood that the scope of protection of the present invention is not limited to the specific embodiments described below; it should also be understood that the terms used in the examples of the present invention are for describing specific embodiments rather than for limiting the scope of protection of the present invention.

[0049] When the embodiment gives a numerical range, it should be understood that, unless otherwise specified in the present invention, the two endpoints of each numerical range and any numerical value between the two endpoints can be selected. Unless otherwise defined, all technical and scientific terms used herein have the same meanings as those of ordinary skill in the art to which the present invention belongs.

[0050] Example 1

[0051] Step 1: If Figure 5 As shown, the sub-wave ω excited by the controllable vibrator is obtained 0 The conjugate function of (t);

[0052] Step 2: Use the vibrator to excite the sub-wave ω 0 The conjugate function of (t) and the original correlation record of the vibrator ( Figure 1 ) is convolved and then divided by the vibrator excitation wavelet ω 0 (t) to obtain the original mother record before correlation ( Figure 2 );

[0053] The calculation formula for the above parent record is as follows:

[0054] S1(t)=ω 0 (t)*e(t)*p 0 (t)*p 1 (t)+n′(t) Formula (1);

[0055] Among them, S1(t) is the parent record; ω 0 (t) represents the sub-wave excited by the vibrator, e(t) represents the formation impulse response, and p 0 (t) is the distortion factor of the mechanical and hydraulic system of the vibrator, p 1 (t) is the correlation factor near the surface, n′(t) is composed of the vibroseis noise similar to the well gun excitation noise and the mechanical noise generated by the vibroseis excitation;

[0056] Where n′(t)=n′ 1 (t)+E'(t);

[0057] In this embodiment, n′ 1 (t) is the surface wave noise (vibroseis noise similar to well gun excitation noise), and E′(t) is the mechanical noise generated by the vibroseis excitation.

[0058] Step 3: According to the time window of 2000ms-8000ms, the amplitude average value is calculated every 21 seismic channels, and the seismic channels with single-channel amplitude values ​​more than 10 times the average amplitude value are eliminated to remove the mechanical noise with strong amplitude ( Figure 3 ), and obtain the mother record after eliminating mechanical noise;

[0059] Step 4: Use a high-pass filter to suppress surface wave noise with a frequency below 8 Hz ( Figure 4 ), and obtain the mother record after suppressing the surface wave. The calculation formula is as follows:

[0060] S2(t)=ω 0 (t)*e(t)*p 0 (t)*p 1 (t) Formula (2);

[0061] Step 5: Use the vibrator to excite the sub-wave ω 0 The conjugate function of (t) and the vibrator force signal ω recorded in the field 0 ′(t)( Figure 6 ) is convolved and then divided by the vibrator excitation wavelet ω 0 (t), and the distortion factor p of the hydraulic and mechanical systems is obtained 0 (t);

[0062] Step 6: Convolve the conjugate function of the distortion factor with each trace in the mother record, and then divide it by the modulus of the distortion factor of the hydraulic and mechanical systems to obtain the mother record after eliminating the influence of the hydraulic and mechanical systems;

[0063] Step 7: Convert the vibrator force signal recorded in the field to 0 The conjugate function of ′(t) and the downhole record ω in the field surface Q value survey 0 ″(t) is convolved and then divided by the vibrator force signal ω recorded in the field 0 ′(t), we get the correlation factor p near the surface 1 (t); Schematic diagram of surface Q value survey is shown in Figure 7 ;

[0064] Step 8: Convolve the conjugate function of the near-surface correlation factor with each trace in the mother record, and then divide it by the modulus of the near-surface correlation factor to obtain the mother record with the near-surface effect eliminated;

[0065] The specific calculation formula for steps six and eight is: Perform convolution and get:

[0066]

[0067] That is ω 0(t)*e(t)×[|p 0 (t)| 2 ×|p 1 (t)| 2 ] Formula (4);

[0068] Divide equation (4) by p 0 (t) and p 1 The modulus of (t) yields: S3(t) = ω 0 (t)*e(t) Formula (5);

[0069] Among them, S3(t) is the mother record after denoising.

[0070] Step 9: Convolve the denoised mother record with the vibrator excitation wavelet to obtain the vibrator-related record after the "black triangle" noise is attenuated, thus achieving the "black triangle" noise attenuation. For details on the data comparison before and after the "black triangle" noise attenuation, see Figure 8 .

[0071] The calculation formula for convolving the denoised mother record with the vibrator-excited wavelet is as follows:

[0072] S4(t)=ω 0 (t)*e(t)*ω 0 (t) Formula (6);

[0073] Among them, S4(t) is the relevant record of the controllable source after denoising.

[0074] Finally, it should be noted that the above content is only used to illustrate the technical solution of the present invention, rather than to limit the scope of protection of the present invention. Simple modifications or equivalent substitutions of the technical solution of the present invention by ordinary technicians in this field do not deviate from the essence and scope of the technical solution of the present invention.

Claims

1. A noise attenuation method, characterized in that: The following steps are involved: 1) Collect the original correlation records of the vibrator seismic data, first perform convolution of the conjugate function of the vibrator excitation wavelet ω0(t) on the original correlation records, and then divide it by the modulus of ω0(t) to obtain the mother record; 2) Calculate the average amplitude of the seismic trace and remove the mechanical noise E'(t) generated by the vibrator excitation in the mother record with an amplitude value more than ten times the average amplitude; 3) Continue to remove the vibroseis noise similar to the well-shot excitation noise in the vibroseis mother record to obtain the mother record that only contains the influence of the mechanical system and near-surface factors; 4) Let the distortion factor of the mechanical and hydraulic system of the vibrator be recorded as p0(t), and the relevant factor near the surface be recorded as p1(t), and calculate the conjugate function of p0(t) and p1(t) respectively. 5) Combine the records including the mechanical system and near-surface factors obtained in step 3) with the records obtained in step 4) Perform convolution and then divide by the modulus of p0(t) and p1(t) to get the denoised mother record; 6) Convolve the denoised mother record with the vibrator excitation wavelet to obtain the vibrator related record and achieve "black triangle" noise attenuation.

2. The noise attenuation method according to claim 1, characterized in that: The calculation formula of the parent record in step 1) is as follows: S1(t)=ω0(t)*e(t)*p0(t)*p1(t)+n′(t) Formula (1); Among them, S1(t) is the mother record; ω0(t) represents the sub-wave excited by the controllable source, e(t) represents the formation pulse response, p0(t) is the distortion factor of the mechanical and hydraulic system of the controllable source, p1(t) is the correlation factor near the surface, and n′(t) is composed of the controllable source noise similar to the well gun excitation noise and the mechanical noise generated during the controllable source excitation.

3. The noise attenuation method according to claim 1, characterized in that: The calculation formula for the record including the influence of mechanical system and near-surface factors in step 3) is as follows: S2(t)=ω0(t)*e(t)*p0(t)*p1(t) Formula (2); Among them, S2(t) is the parent record that only includes the influence of mechanical system and near-surface factors.

4. The noise attenuation method according to claim 1, characterized in that: The distortion factor in step 4) is obtained by recording the vibration plate signal and the trigger signal of the controllable vibrator.

5. The noise attenuation method according to claim 4, characterized in that: The distortion factor in step 4) is obtained as follows: the trigger signal is the excitation wavelet ω0(t), the vibration plate signal of the controllable seismic source signal is the controllable seismic source force signal ω0′(t) recorded in the field, the conjugate function of the controllable seismic source force signal ω0′(t) recorded in the field is convolved with the controllable seismic source excitation wavelet ω0(t), and then divided by the modulus of ω0(t), to obtain the distortion factor p0(t).

6. The noise attenuation method according to claim 1, characterized in that: The near-surface correlation factors in step 4) are obtained by investigating the surface Q value.

7. The noise attenuation method according to claim 6, characterized in that: The near-surface correlation factor in step 4) is obtained by convolving the record ω0″(t) obtained from the surface Q value survey with the conjugate function of the controllable source force signal ω0′(t) recorded in the field, and then dividing it by the modulus of ω0′(t) to obtain the near-surface correlation factor p1(t).

8. The noise attenuation method according to claim 1, characterized in that: The calculation of the denoised mother record in step 5) includes the following steps: Combine formula (2) with Perform convolution and get: That is, ω0(t)*e(t)×[|p0(t)| 2 ×|p1(t)| 2 ] Formula (4); Dividing equation (4) by the modulus of p0(t) and p1(t) yields: S3(t) = ω0(t)*e(t) Equation (5): Among them, S3(t) is the mother record after denoising.

9. The noise attenuation method according to claim 1, characterized in that: The calculation formula for convolving the denoised mother record with the vibroseis excitation wavelet in step 6) is as follows: S4(t)=ω0(t)*e(t)*ω0(t) Formula (6); Among them, S4(t) is the relevant record of the controllable source after denoising.

10. The noise attenuation method according to claim 2, characterized in that: n′(t)=n′1(t)+n′2(t)+……n′ N (t)+E’(t); Among them, n′1(t)+n′2(t)+…n′ N (t) is the controllable source noise similar to the well gun excitation noise, and E'(t) is the mechanical noise generated by the controllable source excitation; where n'1(t)~n' N (t) includes one or more of surface wave noise, multiple wave noise, linear interference noise, random noise, and environmental noise, and N is a positive integer.

Images