Low-signal-to-noise-ratio seismic data Q factor extraction method, electronic equipment and storage medium

Abstract

The invention belongs to the technical field of seismic exploration, and particularly discloses a low-signal-to-noise-ratio seismic data Q factor extraction method, electronic equipment and a storage medium. The method comprises the following steps: firstly, preparing data, screening and classifying VSP seismic data into first-class to third-class quality data, preprocessing, calculating longitudinal wave layer velocities of a first-class quality factor and a second-class quality factor and a sound wave velocity of a third-class quality factor, then extracting a Q factor of the first-class quality factor, and calculating a Q factor of the second-class quality factor; and performing model training on the obtained Q factor to obtain a fitting formula, substituting the longitudinal wave layer velocity of the second type of quality factor into the fitting formula to obtain a fitted Q factor, and substituting the sound wave velocity of the third type of quality factor subjected to smoothing processing into the fitting formula to obtain a Q factor of the third type of quality factor. The method is used for extracting the Q factor, and the accuracy of the extracted Q factor can be improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of seismic exploration, and particularly relates to a method for extracting Q factor of seismic data with low signal-to-noise ratio, an electronic device and a storage medium. Background Art

[0002] In the analysis of vertical seismic profile (VSP) data, the Q factor is emphasized for two reasons: on the one hand, anelastic attenuation can be regarded as a kind of interference to seismic waves and must be eliminated by the inverse Q filtering method. Although a standardized inverse Q filtering method has been formed at present, which effectively improves the interpretation quality of seismic records, the Q factor estimation of the inverse filtering is not accurate, which is a long-existing and difficult problem to solve. On the other hand, people have realized that in the process of improving the interpretation quality of seismic records, the Q factor is an important physical parameter of seismic wave attenuation. Therefore, in addition to the velocity parameter, the Q factor is another seismic parameter that can be used for petrology and lithology interpretation.

[0003] Although there are many methods for calculating the Q factor in the prior art, there are many constraints for extracting the Q factor at present. Especially, it is extremely difficult to extract the Q factor from data with low signal-to-noise ratio, and the accuracy of the extracted Q factor is not high. Summary of the Invention

[0004] To solve the above deficiencies in the known technology, the first aspect of the present invention provides a method for extracting Q factor of seismic data with low signal-to-noise ratio, aiming to obtain a Q factor with higher accuracy.

[0005] To achieve the above object, the technical solutions adopted by the present invention are as follows:

[0006] A method for extracting Q factor of seismic data with low signal-to-noise ratio, the method includes the following steps carried out in sequence:

[0007] S1. Data preparation: Collect VSP seismic data and acoustic logging data in the research target area, screen and classify the VSP seismic data into the first-class quality data to the third-class quality data, perform data preprocessing on the first-class quality data to the third-class quality data to obtain the first-class quality factor to the third-class quality factor, and calculate the longitudinal wave layer velocity of the first-class quality factor and the second-class quality factor and the acoustic wave velocity of the third-class quality factor;

[0008] S2. Extraction of the first-class Q factor samples: Calculate the Q factor of the first-class quality factor by using the spectral ratio method;

[0009] S3. Q factor model training, including the following steps carried out in sequence:

[0010] a1) Establish a fitting relationship, one-to-one correspond the Q factor of the first type of quality factor with the longitudinal wave layer velocity of the same well, and obtain an exponential formula for the Q factor and the longitudinal wave layer velocity of the first type of quality factor by fitting. The exponential formula is denoted as:

[0011]

[0012] where i is the well number of the first type of quality factor, v is the longitudinal wave layer velocity of the first type of quality factor, a i and b i are fitting parameters;

[0013] a2) Optimization of fitting parameters, perform median processing on the fitting parameters obtained in step a1) to obtain optimized fitting parameters. The optimized fitting parameter a m is denoted as:

[0014] a m = median(a i )

[0015] where median() is the median function, and the optimized fitting parameter b m is denoted as:

[0016] b m = median(b i );

[0017] a3) Determine the fitting formula, use the optimized fitting parameters in step a2) to re-establish the fitting formula, and obtain the fitting formula trained with the Q factor of the first type of quality factor. The fitting formula is denoted as:

[0018]

[0019] where V is the velocity;

[0020] S4. Obtain the fitting of the second type of Q factor, substitute the longitudinal wave layer velocity of the second type of quality factor into the fitting formula to obtain the Q factor after fitting of the second type of quality factor;

[0021] Obtain the fitting of the third type of Q factor, perform multi-point smoothing on the acoustic wave velocity of the third type of quality data, and substitute the smoothed acoustic wave velocity of the third type of quality factor into the fitting formula to obtain the Q factor of the third type of quality factor.

[0022] As a limitation, in step S1, the screening and classification of VSP seismic data includes two screening conditions. The first screening condition is data with a signal-to-noise ratio greater than 10; the second screening condition is data with a signal-to-noise ratio greater than 2 and less than 10;

[0023] Data that simultaneously meet the first screening condition and the second screening condition are classified as the first type of quality data, data that meet the first screening condition but do not meet the second screening condition are classified as the second type of quality data, and data that do not meet both screening conditions are classified as the third type of quality data.

[0024] As a second limitation, in the step S1, the data preprocessing includes bad track elimination, three-component rotation, and wavefield separation, and does not include steps that damage the relative relationship of amplitudes.

[0025] As a third limitation, calculating the P-wave layer velocity of the first type of quality factor and the second type of quality factor in the step S1 includes:

[0026] Pick the first arrival of the first type of quality factor and separate the downward wavefield; pick the first arrival of the second type of quality factor;

[0027] Using the first arrivals of the first type of quality factor and the second type of quality factor, and combining the well-source distance and shot point depth information of data acquisition, calculate the P-wave layer velocities of the first type of quality factor and the second type of quality factor respectively;

[0028] Calculate the acoustic velocity of the third type of quality factor according to the acoustic logging data.

[0029] As a fourth limitation, the number of points n for multi-point smoothing processing in the step S4 satisfies the following conditions:

[0030]

[0031] where N vsp is the acquisition interval of VSP seismic data, N ac is the acquisition interval of acoustic logging data, and int() is the rounding function.

[0032] The second aspect of the present invention discloses an electronic device, including a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the steps of the above low signal-to-noise ratio Q factor extraction method are realized.

[0033] The third aspect of the present invention discloses a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above low signal-to-noise ratio Q factor extraction method are realized.

[0034] Due to the adoption of the above technical solutions, compared with the prior art, the beneficial effects obtained by the present invention are:

[0035] (1) The method of the present invention establishes a fitting formula through Q factor model training, and after fitting the Q factor, a more accurate Q factor is obtained, thereby obtaining stable and reliable formation rock absorption parameters, and at the same time providing relatively reliable seismic processing parameters for inverse Q filtering;

[0036] (2) The method of the present invention effectively improves the interpretation quality of seismic records, which is conducive to the popularization and application of seismic processing technology in low signal-to-noise ratio seismic exploration areas.

[0037] In summary, the present invention can further improve the accuracy of the extracted Q factor, effectively improve the interpretation quality of seismic records, obtain stable and reliable formation rock absorption parameters, and at the same time is conducive to the popularization and application of seismic processing technology in low signal-to-noise ratio seismic exploration areas. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0039] Figure 1 is a flowchart of an embodiment of the present invention;

[0040] Figure 2 is a sample diagram of the quality factor Q value extracted from the first type of quality factor in an embodiment of the present invention;

[0041] Figure 3 is the longitudinal wave layer velocity extracted from the first type of quality factor in an embodiment of the present invention;

[0042] Figure 4 is a sample diagram of the quality factor Q value extracted from the second type of quality factor in an embodiment of the present invention;

[0043] Figure 5 is the longitudinal wave layer velocity extracted from the second type of quality factor in an embodiment of the present invention;

[0044] Figure 6 is a sample diagram of the quality factor Q value obtained by fitting the second type of quality factor in an embodiment of the present invention;

[0045] Figure 7 is a sample diagram of the quality factor Q value obtained by fitting the third type of quality data in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0046] To better explain the present invention for easy understanding, the preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings through specific embodiments.

[0047] Embodiment 1 A method for extracting Q factor of low signal-to-noise ratio seismic data

[0048] This embodiment provides a method for extracting Q factor of low signal-to-noise ratio seismic data, including the following steps carried out in sequence:

[0049] S1. Data preparation: Collect VSP seismic data and acoustic logging data in the research target area. Screen and classify the VSP seismic data into the first-class quality data, the second-class quality data, and the third-class quality data. Perform data preprocessing on the first-class quality data, the second-class quality data, and the third-class quality data to obtain the first-class quality factor, the second-class quality factor, and the third-class quality factor, and calculate the P-wave layer velocity of the first-class quality factor and the second-class quality factor, as well as the acoustic velocity of the third-class quality factor.

[0050] Among them, the VSP seismic data refers to the original seismic data collected by the VSP observation method in the research target area, which is different from the surface seismic data. The screening and classification of the VSP seismic data include two screening conditions. The first screening condition is that the first arrival jumps simply, the first arrival wave is clear, and the signal-to-noise ratio is greater than 10. The second screening condition is that the data has a relatively high signal-to-noise ratio, the amplitude attenuation law of the first arrival wave, and the signal-to-noise ratio is greater than 2 and less than 10. The data that simultaneously meets the first screening condition and the second screening condition is classified as the first-class quality data; the data that meets the first screening condition but does not meet the second screening condition is classified as the second-class quality data; the data that does not meet both screening conditions is classified as the third-class quality data. Table 1 shows the three types of quality data and the corresponding classification criteria.

[0051] Table 1 Three types of quality data and the corresponding classification criteria

[0052]

[0053] In this embodiment, the data preprocessing of the VSP seismic data includes bad trace removal, three-component rotation, and wavefield separation, and steps that damage the relative amplitude relationship, such as amplitude compensation and automatic gain, are not included.

[0054] Before calculating the P-wave layer velocity of the first-class quality factor and the second-class quality factor, it is necessary to pick the first arrival of the first-class quality factor and separate the downward wavefield to make the subsequent Q-factor calculation more stable; pick the first arrival of the second-class quality factor. Using the first arrival of the first-class quality factor and the second-class quality factor, and combining the well-source distance and the shot depth information of the data acquisition, the P-wave layer velocities of the first-class quality factor and the second-class quality factor are calculated respectively through the Geoeast software. The well-source distance and the shot depth information are known when collecting the data in the research target area in step S1. The acoustic velocity of the third-class quality factor is calculated according to the acoustic logging data.

[0055] S2. Extraction of the first-class Q-factor samples: Calculate the Q-factor of the first-class quality factor by the spectral ratio method.

[0056] The spectral ratio method used to calculate the Q factor of the first type of quality factor is based on the method described in the article "The determination of the seismic quality factor Q from VSP data: a comparison of different computational methods [J]. Geophysical Prospecting, 1991, 39(1): 1-27."

[0057] Figure 2 The Q-value samples of the first type of quality factor are shown. The abscissa in the figure is the depth in meters, and the ordinate is the Q value, dimensionless; Figure 3 The P-wave layer velocity of the first type of quality factor is shown. The abscissa in the figure is the depth in meters, and the ordinate is the velocity in m / s; Figure 4 The Q-value samples of the second type of quality factor are shown. The abscissa in the figure is the depth in meters, and the ordinate is the Q value, dimensionless; Figure 5 The P-wave layer velocity of the second type of quality factor is shown. The abscissa in the figure is the depth in meters, and the ordinate is the velocity in m / s. Comparing Figure 2 and Figure 4 , it is concluded that the Q-factor error of the second type of quality factor is extremely large and cannot be applied.

[0058] S3. Q-factor model training, including the following steps carried out sequentially:

[0059] a1) Establish a fitting relationship, one-to-one correspond the Q factor of the first type of quality factor with the P-wave layer velocity of the same well, and fit to obtain the exponential formula of the Q factor and the P-wave layer velocity of the first type of quality factor. The exponential formula is denoted as:

[0060]

[0061] where i is the well number of the first type of quality factor, v is the P-wave layer velocity of the first type of quality factor, a i and b i are fitting parameters.

[0062] a2) Optimization of fitting parameters, perform median processing on the fitting parameters obtained in step a1) to obtain the optimized fitting parameters. The optimized fitting parameter a m is denoted as:

[0063] a m = median(a i )

[0064] where median() is the median function, and the optimized fitting parameter b m is denoted as:

[0065] b m = median(b i ).

[0066] a3) Determine the fitting formula, and use the optimized fitting parameters in step a2) to re - establish the fitting formula, obtaining the fitting formula trained with the Q - factor of the first - type quality factor. The fitting formula is denoted as:

[0067]

[0068] where V is the velocity.

[0069] S4. Obtain the fitting of the second - type Q - factor. Substitute the P - wave layer velocity of the second - type quality factor into the fitting formula to obtain the Q - factor after fitting the second - type quality factor; perform multi - point smoothing on the acoustic wave velocity of the third - type quality factor, and substitute the smoothed acoustic wave velocity of the third - type quality factor into the fitting formula to obtain the Q - factor of the third - type quality factor.

[0070] The number of points n for the multi - point smoothing process satisfies the following conditions:

[0071]

[0072] where N vsp is the acquisition interval of VSP seismic data, N ac is the acquisition interval of acoustic logging data, and int() is the rounding function.

[0073] As Figure 6 shown is the Q - value sample after fitting the second - type quality factor. Comparing Figure 4 with the Q - value sample before fitting, the error of the Q - value sample after fitting is reduced; Figure 7 shown is the Q - value sample after fitting the third - type quality factor. The error of the Q - value sample after fitting is reduced.

[0074] Embodiment 2 An electronic device and a storage medium for extracting the Q - factor of low - signal - to - noise seismic data

[0075] The electronic device in this embodiment uses a computer device, including a memory and a processor. A computer program is stored in the memory. When the processor executes the computer program, the steps of the method for extracting the Q - factor of low - signal - to - noise seismic data in Embodiment 1 can be implemented.

[0076] A computer-readable storage medium stores a computer program thereon. When the computer program is executed by a processor, the steps of the Q-factor extraction method for low signal-to-noise ratio seismic data in Embodiment 1 can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

Claims

1. A method for extracting the Q factor of seismic data with low signal-to-noise ratio, characterized in that, the method includes the following steps carried out in sequence: S1. Data preparation: Collect VSP seismic data and acoustic logging data in the research target area, screen and classify the VSP seismic data into first-class quality data to third-class quality data, perform data preprocessing on the first-class quality data to third-class quality data to obtain first-class quality factors to third-class quality factors, and calculate the longitudinal wave layer velocity of the first-class quality factor and the second-class quality factor and the acoustic wave velocity of the third-class quality factor; S2. Extraction of the first-class Q factor samples: Calculate the Q factor of the first-class quality factor by the spectral ratio method; S3. Q factor model training, including the following steps carried out in sequence: a1) Establish a fitting relationship: One-to-one correspond the Q factor of the first-class quality factor with the longitudinal wave layer velocity of the same well, and fit to obtain an exponential formula for the Q factor of the first-class quality factor and the longitudinal wave layer velocity, and the exponential formula is denoted as: where i is the well serial number of the first type of quality factor, v is the P-wave layer velocity of the first type of quality factor, and a i and b i are fitting parameters; a2) Fitting parameter optimization: perform median processing on the fitting parameters obtained in step a1) to obtain optimized fitting parameters, and the optimized fitting parameter is a m Denoted as: a m = median(a i ) Among them, median() is the median function, and the optimized fitting parameter b m is denoted as: b m = median(b i ); a3) Determine the fitting formula: Use the optimized fitting parameters in step a2) to re-establish the fitting formula to obtain the fitting formula trained with the Q factor of the first-class quality factor, and the fitting formula is denoted as: where V is the velocity; S4. Fitting and obtaining the second-class Q factor: Substitute the longitudinal wave layer velocity of the second-class quality factor into the fitting formula to obtain the Q factor of the second-class quality factor after fitting; Fitting and obtaining the third-class Q factor: Perform multi-point smoothing on the acoustic wave velocity of the third-class quality data, and substitute the acoustic wave velocity of the third-class quality factor after smoothing into the fitting formula to obtain the Q factor of the third-class quality factor.

2. The method for extracting the Q factor of seismic data with low signal-to-noise ratio according to claim 1, characterized in that, in the step S1, the screening and classification of the VSP seismic data includes two screening conditions. The first screening condition is data with a signal-to-noise ratio greater than 10; the second screening condition is data with a signal-to-noise ratio greater than 2 and less than 10; The data that simultaneously meet the first screening condition and the second screening condition are classified as first-class quality data, the data that meet the first screening condition but do not meet the second screening condition are classified as second-class quality data, and the data that do not meet both screening conditions are classified as third-class quality data.

3. The method for extracting the Q factor of seismic data with low signal-to-noise ratio according to claim 1 or 2, characterized in that, in the step S1, the data preprocessing includes bad trace rejection, three-component rotation and wave field separation, and does not include steps that damage the relative relationship of amplitudes.

4. The method for extracting the Q factor of seismic data with low signal-to-noise ratio according to claim 1 or 2, characterized in that, the calculation of the longitudinal wave layer velocity of the first-class quality factor and the second-class quality factor in the step S1 includes: Pick the first arrival of the first-class quality factor and separate the downward wave field; pick the first arrival of the second-class quality factor; Use the first arrivals of the first-class quality factor and the second-class quality factor, and combine the well-source distance and the excitation point depth information of data acquisition to calculate the longitudinal wave layer velocity of the first-class quality factor and the second-class quality factor respectively; Calculate the acoustic wave velocity of the third-class quality factor according to the acoustic logging data.

5. The method for extracting the Q factor of low signal-to-noise ratio seismic data according to claim 1 or 2, characterized in that, in step S4, the number of points n for multi-point smoothing processing satisfies the following conditions: where N vsp is the acquisition interval of VSP seismic data, and N ac is the acquisition interval of acoustic logging data, and int() is the rounding function.

6. An electronic 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 method described in any one of claims 1-5 are implemented.

7. 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 described in any one of claims 1-5 are implemented.

Images