Full-band automatic identification and suppression method and device for seismic traces containing single-frequency interference

Through the full-band automatic identification and suppression method, Fourier transform and least squares optimal approximation technology are used to solve the seismic channel identification and suppression problems of single-frequency interference of multiple frequency components, improving the quality of seismic data and assisting in fine oil and gas exploration.

CN119165527BActive Publication Date: 2025-09-02DAQING OILFIELD CO LTD +1
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202310738389.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-06-20
Publication Date
2025-09-02
Estimated Expiration
2043-06-20

AI Technical Summary

Technical Problem

The prior art cannot effectively identify and suppress single-frequency interfering seismic channels containing multiple frequency components and unfixed frequency positions in seismic data, resulting in limited pre-stack seismic pretreatment work.

Method used

The full-band automatic identification and suppression method is adopted, and the Fourier transform, smoothing processing, least squares optimal approximation and final single-frequency interference model channel is constructed, and the frequency components are suppressed, and the multi-threaded processing technology is used to achieve fully automatic identification and adaptive suppression.

Benefits of technology

It improves the quality of earthquake data, improves the precision of oil and gas exploration, effectively removes single-frequency interference, and has significant fidelity and amplitude protection treatment effect.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119165527B_ABST
    Figure CN119165527B_ABST
Patent Text Reader

Abstract

A method and device for automatic full-band identification and suppression of seismic channels containing single-frequency interference. It mainly solves the problem that existing methods cannot effectively identify and suppress single-frequency interference seismic channels containing multiple frequency components and whose frequency positions are not fixed in any seismic channel. It includes: obtaining the total number of seismic channels of seismic data; determining the total number of task processing according to the number of node threads; continuously reading several seismic channels from the seismic data and putting them into the node memory according to the thread priority preset by the node; reading a seismic channel, identifying the single-frequency interference frequency in the seismic channel, constructing a single-frequency interference model channel, and suppressing the single-frequency interference frequency component; completing the processing task of a group of seismic channels in any thread; identifying and suppressing seismic channels containing multiple single-frequency interference in all seismic channels. This method solves the problem of fully automatic identification and adaptive suppression of single-frequency interference seismic channels containing multiple frequency components and whose frequency positions are not fixed in any seismic channel in the entire region.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of seismic exploration data processing, in particular to a method and device for automatically identifying and suppressing full-band single-frequency interference seismic traces containing multiple frequency components. Background Art

[0002] In areas mature for oil and gas exploration, densely populated well networks, vibroseis data acquisition areas, or other areas with periodic noise sources, seismic signals received by geophones are often superimposed with single-frequency interference of one or more frequency components, known as industrial electrical interference. The amplitude, phase, and frequency of each single-frequency interference component remain essentially constant from shallow to deep layers. Each seismic trace may also contain single-frequency interference of one or more frequency components, overlapping with the dominant frequency band of the seismic data. These frequency components are unevenly distributed within the effective frequency band of the seismic data, exhibiting no regularity. Automatically identifying and adaptively suppressing this uneven distribution of single-frequency interference within the effective frequency band while preserving fidelity and amplitude is crucial for improving the fidelity of pre-stack seismic data preprocessing and a crucial factor in enabling precise oil and gas exploration. To this end, geophysicists have proposed methods for identifying seismic traces containing heavy single-frequency interference, such as the energy statistics identification method of seismic traces containing industrial single interference based on common detection point gathers and bandpass filtering processing, and the automatic identification and suppression method of industrial electrical interference by multiple iterative approximation of fixed frequency components. At the same time, they have proposed a variety of single-frequency interference suppression methods in the time domain or frequency domain, such as the frequency domain limited frequency filtering method, the industrial electrical interference suppression method based on Wiener filtering, the linear spectrum modulation method, and the suppression method of industrial electrical interference by approximation of sine and cosine functions.

[0003] The aforementioned literature or patent methods can only effectively identify and suppress single-frequency interference with amplitude preservation for a given seismic trace containing one or more fixed frequency components. This requires manual identification of the frequency components, resulting in a complex process and low identification accuracy. Furthermore, existing methods and technologies are unable to effectively address situations where a single seismic trace's effective frequency band contains multiple single-frequency interference components with unstable frequency distribution, or where the spatial distribution of single-frequency interference within the seismic area is uneven and the frequency distribution is unstable, thus restricting pre-stack seismic preprocessing for amplitude preservation. Summary of the Invention

[0004] In order to overcome the problem that existing methods cannot effectively identify and suppress single-frequency interference seismic channels containing multiple frequency components and with unfixed frequency positions in any seismic channel, the present invention provides a full-band automatic identification and suppression method for single-frequency interference seismic channels. This method can efficiently solve the problem of fully automatic identification and adaptive suppression of single-frequency interference seismic channels containing multiple frequency components and with unfixed frequency positions in any seismic channel in the entire region.

[0005] The technical solution of the present invention is: a full-band automatic identification and suppression method containing single-frequency interference seismic traces, comprising the following steps:

[0006] S1, the total number of seismic channels T obtained from seismic data 总 , as the total task pool;

[0007] S2, based on the number of nodes and the total number of threads N o , determine the total number of tasks to be processed N;

[0008] S3, according to the thread priority preset by the node, each thread reads T from the seismic data of step S1 in sequence n The seismic traces are put into the node memory. At the same time, for the first completed T n The thread for processing the seismic traces reads T data from the seismic data of step S1 again. n Each seismic trace is put into the node memory;

[0009] S4. Sequentially read a seismic trace from the data in step S3, automatically identify the initial frequency and quantity of the single-frequency interference in the seismic trace on its spectrum, accurately construct a single-frequency interference model trace for each frequency component through least squares optimal approximation, and suppress the single-frequency interference frequency component by frequency component;

[0010] S5. Complete a set of T in any thread n The processing task of each seismic channel is to store the seismic channels after the single-frequency interference is suppressed, and the suppressed single-frequency interference seismic channels are selected and output according to the needs;

[0011] S6. Repeat steps S3-S5 to automatically identify and suppress seismic traces containing multiple single-frequency interferences in all seismic traces.

[0012] Furthermore, in step 1, the observation system definition is performed on the collected original seismic data, invalid channels or abnormal channels are removed, the original seismic data to be processed are obtained, and the total number of seismic channels T of the seismic data is obtained. 总 , managed uniformly as a task pool.

[0013] Furthermore, in step S2, when T 总 / T n When the value of is an integer, N=T 总 / T n When T 总 / T n When the value of is a decimal, N=(int)[T 总 / T]+1, where T n The number of seismic traces processed at a time by each thread.

[0014] Furthermore, in step 4, the following steps are performed:

[0015] S4.1. Performing Fourier transform on the seismic trace to obtain a frequency spectrum of the seismic trace;

[0016] S4.2. Smoothing the spectrum to obtain a smoothed spectrum;

[0017] S4.3. Divide the spectrum obtained in step S4.1 by the spectrum obtained in step S4.2 to obtain a ratio spectrum;

[0018] S4.4. Preset the threshold value of the ratio spectrum to obtain the frequency number N of the single-frequency interference contained in the seismic trace f and initial frequency;

[0019] S4.5. For any single-frequency interference initial frequency in step S4.4, first construct an initial single-frequency interference model trace. Optimally approximate the seismic trace and the initial single-frequency interference model trace using the least squares method to construct a final single-frequency interference model trace corresponding to the initial frequency. Subtract the final single-frequency interference model trace from the seismic trace to obtain a single-frequency interference trace that suppresses the initial frequency.

[0020] S4.6. Complete the suppression processing of all single-frequency interference frequency components of the seismic channel.

[0021] Furthermore, in step S5, interference seismic traces containing multiple single-frequency interference frequency components are obtained and output from the calculation memory to the disk for storage.

[0022] Furthermore, in step S6, for any thread that completes step 5, continue to read the seismic channel data and repeat steps S3-S5 until all threads complete all tasks in the task pool, identify and suppress all seismic channels containing multiple single-frequency interference seismic channels, and store them.

[0023] Furthermore, the smoothing method in step S4.2 is: mean filtering or median filtering.

[0024] Furthermore, in step S4.3, the ratio spectrum is not less than 1, and the frequency range of the ratio spectrum is the effective frequency band range of the seismic data, which is obtained by analyzing the average spectrum of the work area or by analyzing the spectrum of a single seismic trace.

[0025] Furthermore, in step S4.4, the threshold value of the ratio spectrum is set such that the same threshold value is used for all seismic traces in the same seismic work area.

[0026] An automatic identification and suppression device for single-frequency interference seismic traces containing multiple frequency components, comprising a seismic data acquisition unit for obtaining the total number of seismic traces T of seismic data 总 ;

[0027] Obtain the total task processing number unit, which is used to determine the total task processing number N;

[0028] Reading unit, continuously reads T from seismic data n Each seismic trace is put into the node memory;

[0029] An identification unit, which identifies the initial frequency and quantity of single-frequency interference in the seismic trace;

[0030] The suppression unit suppresses the single-frequency interference frequency components one by one.

[0031] The present invention has the following beneficial effects: due to the adoption of the above-mentioned scheme, it can effectively solve the influence of the strong energy single-frequency interference on the quality of seismic data in mature oil and gas exploration areas, dense well network areas, controllable source data acquisition areas or other periodic noise source areas, and at the same time efficiently solve the problem of fully automatic identification and adaptive suppression of single-frequency interference seismic channels containing multiple frequency components and non-fixed frequency positions in any seismic channel in the entire area, thereby greatly improving the quality of seismic data in these exploration and development blocks, and contributing to the effectiveness of fine oil and gas exploration and development. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 is a flow chart of the present invention;

[0033] Figure 2 It is vibroseis data with single-frequency interference containing multiple frequency components;

[0034] Figure 3 It is a spectrum of vibrator with single-frequency interference containing multiple frequency components;

[0035] Figure 4 It is the vibroseis data after identifying and suppressing multiple single-frequency interference seismic traces;

[0036] Figure 5 It is the spectrum of the vibroseis source after identifying and suppressing multiple single-frequency interference seismic traces;

[0037] Figure 6 It is the suppressed single-frequency interference result data;

[0038] Figure 7 It is the suppressed single-frequency interference spectrum. DETAILED DESCRIPTION

[0039] The present invention will be further described below in conjunction with the accompanying drawings:

[0040] Depend on Figure 1 As shown, a full-band automatic identification and suppression method containing single-frequency interference seismic traces includes the following steps:

[0041] S1. Define the observation system for the collected original seismic data and remove invalid or abnormal channels. This step does not perform processing such as amplitude compensation, deconvolution, and multi-channel denoising. In this way, the original seismic data for industrial electrical interference identification and suppression is obtained, and the total number of seismic channels T of the seismic data is obtained. 总 , managed uniformly as the general task pool.

[0042] S2. Get the number of CPU cores of the current computing node and preset the number of threads N of the node o , requires the number of threads N o Not greater than the number of cores in the node, and the number of seismic channels processed by each thread at a time is preset T n , determine the total number of tasks to be processed N.

[0043] When T 总 / T n When the value of is an integer, N=T 总 / T n ,

[0044] When T 总 / T n When the value of is a decimal, N=(int)[T 总 / T]+1, where (int)[·] represents the rounding down operation.

[0045] Where T n The number of seismic traces processed at a time by each thread.

[0046] S3, for any thread in step S2, continuously read T from the seismic data in step S1 according to the thread priority preset by the node n The seismic traces are placed in the node memory as the first set of processing tasks for the thread in the task pool.

[0047] S4. Sequentially read a seismic trace from the data in step S3 and process it according to the following steps:

[0048] S4.1. Performing Fourier transform on the seismic trace to obtain a frequency spectrum of the seismic trace;

[0049] S4.2. Smoothing the spectrum to obtain a smoothed spectrum. Smoothing methods include mean filtering, median filtering, etc.

[0050] S4.3. Divide the spectrum obtained in step S4.1 by the spectrum obtained in step S4.2 to obtain a ratio spectrum. The ratio spectrum is greater than or equal to 1, and the frequency range of the ratio spectrum is mainly the effective frequency band range of the seismic data. The ratio spectrum is obtained by analyzing the average spectrum of the work area or by analyzing the spectrum of a single seismic trace.

[0051] S4.4. Preset the threshold value for the ratio spectrum based on actual work experience. The threshold value is typically the same for all seismic traces in the same seismic area. Different threshold values ​​can also be used for different frequency components within the same trace. For single-frequency interference with a single frequency component, a larger threshold value indicates a stronger amplitude energy for that frequency component in the original trace. Conversely, a smaller threshold value indicates a weaker amplitude energy.

[0052] The frequency positions where the ratio spectrum value is greater than or equal to the threshold value are identified as the single-frequency interference frequencies contained in the seismic trace; for the frequency positions where the ratio spectrum value is less than the threshold value, the frequency positions are considered to contain no single-frequency interference frequency components. The number of single-frequency interference frequencies contained in the seismic trace and the initial frequency of each interference frequency are obtained. When the initial frequency is 0, the seismic trace is considered to contain no single-frequency interference and no processing is performed;

[0053] S4.5. For any single-frequency interference initial frequency in step S4.4, first construct an initial single-frequency interference model trace in the form of sine or cosine. Optimally approximate the seismic trace and the initial single-frequency interference model trace using the least squares method to obtain the single-frequency interference frequency, amplitude, and phase with the minimum square error. From this, construct a final single-frequency interference model trace corresponding to the initial frequency, and subtract the final single-frequency interference model trace from the seismic trace to obtain the single-frequency interference corresponding to the initial frequency that has been suppressed.

[0054] S4.6. Complete the single-frequency interference suppression processing of all frequencies of the seismic channel according to step S4.5, thereby completing the suppression processing of all single-frequency interference frequency components of the seismic channel.

[0055] S5, any thread completes a set of T n The processing task of each seismic channel is to output the seismic channel after the single-frequency interference is suppressed from the computing memory to the disk for storage according to the preset data format; at the same time, the interference seismic channel containing multiple single-frequency interference frequency components can be obtained by subtracting the seismic channel after the single-frequency interference is suppressed from the original seismic channel as needed, and it can also be output from the computing memory to the disk for storage.

[0056] S6. For any thread that completes step 5, continue to read seismic data from the task pool, and complete the group of processing tasks in accordance with steps 3, 4, and 5 in sequence until all threads complete all tasks in the task pool. In this way, the fully automatic identification and adaptive suppression processing of all seismic channels in the entire work area, including multiple single-frequency interference seismic channels, are completed and stored.

[0057] An automatic identification and suppression device for single-frequency interference seismic traces containing multiple frequency components, comprising a seismic data acquisition unit for obtaining the total number of seismic traces T of seismic data 总 ;

[0058] Obtain the total task processing number unit, which is used to determine the total task processing number N;

[0059] Reading unit, continuously reads T from seismic data n Each seismic trace is put into the node memory;

[0060] An identification unit, which identifies the initial frequency and quantity of single-frequency interference in the seismic trace;

[0061] The suppression unit suppresses the single-frequency interference frequency components one by one.

[0062] In some embodiments, the functions or modules and units included in the device provided by the embodiment of the present invention can be used to execute the method described in the above method embodiment. Its specific implementation can refer to the description of the above method embodiment. For the sake of brevity, it will not be repeated here.

[0063] Example: Taking the data collected by a certain vibrator in Sichuan Basin as an example, the present invention is further described in detail with reference to the accompanying drawings.

[0064] S1. Perform observation system processing and remove invalid or abnormal channels on a beam of controllable seismic data collected in the field to obtain the total number of seismic channels T of the seismic data. 总 =18400000, one of the gun heads is Figure 1 As shown in the figure, single-frequency interference seriously reduces the quality of vibroseis data, and obvious single-frequency interference energy can be clearly seen in the spectrum. Its frequency distribution basically overlaps with the effective signal band, and its single-frequency energy is abnormally prominent in the spectrum.

[0065] S2. The number of CPU cores of the current computing node is 16. The number of threads of the node is preset to be equal to the number of cores of the node, that is, N o =16, and the number of seismic channels processed by each thread at a time is preset n =1000, thus obtaining the total number of task processing times N=T 总 / T n =18400.

[0066] S3, N in step S2 o = 16, reads T continuously from the seismic data in step 1 according to the thread priority preset by the node n = 1000 seismic traces are placed in the node memory as the first set of processing tasks for this thread in the task pool.

[0067] S4. Sequentially read one seismic trace from the first set of processed data in step S3 and process it according to the following steps:

[0068] S4.1. Performing Fourier transform on the seismic trace to obtain a frequency spectrum of the seismic trace;

[0069] S4.2. Filter the spectrum mean to obtain a smoothed spectrum;

[0070] S4.3. Divide the spectrum obtained in step S4.1 by the spectrum obtained in step S4.2 to obtain a ratio spectrum;

[0071] S4.4. Preset a ratio spectrum threshold of 2.0 and identify frequency locations with ratio spectrum values ​​greater than or equal to 2.0 as single-frequency interference frequencies contained in the seismic trace. For frequency locations with ratio spectrum values ​​less than 2.0, consider that the frequency locations do not contain single-frequency interference frequency components. Obtain the number and initial frequency of the single-frequency interference frequencies contained in the seismic trace.

[0072] S4.5. For any single-frequency interference initial frequency in step S4.4, first construct a sinusoidal initial single-frequency interference model trace. Optimally approximate the seismic trace and the initial single-frequency interference model trace using the least squares method to construct a final single-frequency interference model trace corresponding to the initial frequency. Subtract the final single-frequency interference model trace from the seismic trace to obtain the single-frequency interference energy that suppresses the initial frequency.

[0073] S4.6. Complete the single-frequency interference suppression processing of all frequencies of the seismic channel according to step S4.5, thereby completing the suppression processing of all single-frequency interference frequency components of the seismic channel.

[0074] S5. Complete a set of T in any thread n The processing task of each seismic channel is to output the seismic channel after the single-frequency interference is suppressed from the computing memory to the disk for storage; at the same time, the interference seismic channel containing multiple single-frequency interference frequency components is obtained and output from the computing memory to the disk for storage.

[0075] S6. For any thread that completes step S5, continue to read seismic data from the task pool, and complete the group of processing tasks in sequence according to steps 3, 4, and 5 until all threads complete all tasks in the task pool, thereby completing the fully automatic identification and adaptive suppression processing of all seismic channels in the entire work area, including multiple single-frequency interference seismic channels.

[0076] Figure 2 、 Figure 3 The vibroseis data and spectrum contain single frequency interference with multiple frequency components. Figure 4 、 Figure 5 After applying this method, the vibroseis data and spectrum after fully automatic identification and adaptive suppression of multiple single-frequency interference seismic traces are obtained. The energy of single-frequency interference is greatly attenuated, and the signal-to-noise ratio of the vibroseis data is significantly improved. At the same time, it can be clearly seen from the spectrum that the effective reflection signal has been effectively restored. Figure 6 、 Figure 7The single-frequency interference and spectrum containing multiple frequency components suppressed by this method are not shown in the figure. This shows that the patented method of removing single-frequency interference has good amplitude preservation and fidelity, and has broad prospects for promotion and application.

Claims

1. A full-band automatic identification and suppression method for seismic traces containing single-frequency interference, characterized in that The following steps are involved: S1, the total number of seismic channels T obtained from seismic data 总 , as the total task pool; S2, based on the number of nodes and the total number of threads N o , determine the total number of tasks to be processed N; S3, according to the thread priority preset by the node, each thread reads T from the seismic data of step S1 in sequence n The seismic traces are put into the node memory. At the same time, for the first completed T n The thread for processing the seismic traces reads T data from the seismic data of step S1 again. n Each seismic trace is put into the node memory; S4. Sequentially read a seismic trace from the data in step S3, automatically identify the initial frequency and quantity of the single-frequency interference in the seismic trace on its spectrum, accurately construct a single-frequency interference model trace for each frequency component through the least squares optimal approximation method, and suppress the single-frequency interference frequency component by frequency component; Follow these steps to process: S4.

1. Performing Fourier transform on the seismic trace to obtain a frequency spectrum of the seismic trace; S4.

2. Smoothing the spectrum to obtain a smoothed spectrum; S4.

3. Divide the spectrum obtained in step S4.1 by the spectrum obtained in step S4.2 to obtain a ratio spectrum; S4.

4. Preset the threshold value of the ratio spectrum to obtain the frequency number N of the single-frequency interference contained in the seismic trace f and initial frequency; S4.

5. For any single-frequency interference initial frequency in step S4.4, first construct an initial single-frequency interference model trace. Optimally approximate the seismic trace and the initial single-frequency interference model trace using the least squares method to construct a final single-frequency interference model trace corresponding to the initial frequency. Subtract the final single-frequency interference model trace from the seismic trace to obtain a single-frequency interference trace that suppresses the initial frequency. S4.

6. Complete the suppression of all single-frequency interference frequency components of the seismic channel; S5. Complete a set of T in any thread n The processing task of each seismic channel is to store the seismic channels after the single-frequency interference is suppressed, and the suppressed single-frequency interference seismic channels are selected and output according to the needs; S6. Repeat steps S3-S5 to automatically identify and suppress seismic traces containing multiple single-frequency interferences in all seismic traces.

2. The full-band automatic identification and suppression method for seismic traces containing single-frequency interference according to claim 1 is characterized in that: In step 1, the observation system definition is performed on the collected original seismic data, invalid or abnormal channels are removed, the original seismic data to be processed is obtained, and the total number of seismic channels T of the seismic data is obtained. 总 , managed uniformly as a task pool.

3. The method for automatic identification and suppression of full-band seismic traces containing single-frequency interference according to claim 1, characterized in that: In step S2, when T 总 / T n When the value of is an integer, N=T 总 / T n When T 总 / T n When the value of is a decimal, N=(int)[T 总 / T]+1, where T n The number of seismic traces processed at a time by each thread.

4. The method for automatic identification and suppression of full-band seismic traces containing single-frequency interference according to claim 1, characterized in that: In step S5, interference seismic traces containing multiple single-frequency interference frequency components are obtained and output from the calculation memory to the disk for storage.

5. The method for automatic identification and suppression of full-band seismic traces containing single-frequency interference according to claim 1, characterized in that: In step S6, for any thread that completes step 5, continue to read the seismic channel data and repeat steps S3-S5 until all threads complete all tasks in the task pool. All seismic channels containing multiple single-frequency interference seismic channels are identified and suppressed and stored.

6. The method for automatic identification and suppression of full-band seismic traces containing single-frequency interference according to claim 1, characterized in that: The smoothing method in step S4.2 is: mean filtering or median filtering.

7. The method for automatic identification and suppression of full-band seismic traces containing single-frequency interference according to claim 1, characterized in that: In step S4.3, the ratio spectrum is not less than 1, and the frequency range of the ratio spectrum is the effective frequency band range of the seismic data, which is obtained by analyzing the average spectrum of the work area or by analyzing the spectrum of a single seismic trace.

8. The full-band automatic identification and suppression method for seismic traces containing single-frequency interference according to claim 1 is characterized in that: In step S4.4, the threshold value of the ratio spectrum is set such that all seismic traces in the same seismic area use the same threshold value.

9. A device using the automatic identification and suppression method according to any one of claims 1 to 8, characterized in that: Including the seismic data acquisition unit, the total number of seismic channels used to obtain seismic data T 总 ; Obtain the total task processing number unit, which is used to determine the total task processing number N; Reading unit, continuously reads T from seismic data n Each seismic trace is put into the node memory; An identification unit, which identifies the initial frequency and quantity of single-frequency interference in the seismic trace; The suppression unit suppresses the single-frequency interference frequency components one by one.

Citation Information

Patent Citations

  • Method for adaptively recognizing and eliminating seismic exploration single-frequency interference

    CN101551465A

  • Data domain oriented model channel reconstruction multi-focusing imaging method

    CN106338762A