A method for extracting earthquake source wavelet based on empirical mode method

Through empirical modal decomposition and feature evaluation of seismic data, the problem of insufficient adaptability and multi-component signal processing in complex time-varying signal processing is solved, and the fineness of the source wavelet is achieved Identification and separation improves the accuracy and flexibility of signal decomposition.

CN118566980BActive Publication Date: 2024-11-08INST OF OCEANOLOGY - CHINESE ACAD OF SCI
View PDF 10 Cites 0 Cited by

Patent Information

Application Number
CN202411022252.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-29
Publication Date
2024-11-08
Estimated Expiration
2044-07-29

AI Technical Summary

Technical Problem

Traditional seismic wavelet extraction methods lack adaptability and limited multi-component signal processing capabilities when processing complex time-varying signals, especially when the seismic wavelet is highly mixed with other wave components, it is difficult to accurately separate.

Method used

The seismic data is decomposed by empirical modal decomposition (EMD) technology, and the inherent modal function (IMF) is extracted, and candidate wavelets are selected through feature similarity evaluation, denoising and correction are performed, and the extracted wavelet pattern is finally verified. .

Benefits of technology

Adaptive processing of complex seismic signals is realized, and the source wavelet can be more precisely identified and separated, especially when the wavelet contains multiple frequency components, improving the accuracy and flexibility of signal decomposition.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118566980B_ABST
    Figure CN118566980B_ABST
Patent Text Reader

Abstract

The present invention discloses a method for extracting source wavelets based on the empirical mode method, and belongs to the technical field of seismic data processing. The method comprises the following steps: signal decomposition; selection of a suitable IMF; denoising and smoothing; wavelet correction and synthesis and verification and application. The present invention is applicable to nonlinear and non-stationary signals, does not require pre-setting of basis functions, identifies and decomposes signals through local averages and local extreme points, and can automatically adapt to the intrinsic characteristics and change trends of signals, making it more flexible and effective when processing complex geological structures and different types of seismic signals. The original seismic signal can be decomposed into a series of intrinsic mode functions, each IMF representing the information of a specific frequency band of the signal, which helps to more finely identify and separate the source wavelets. The leakage and aliasing effects caused by the common window function in spectrum analysis are avoided, thereby better maintaining the time locality and frequency characteristics of the signal.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of seismic data processing, and in particular to a method for extracting source wavelets based on an empirical mode method. Background Art

[0002] Empirical Mode Decomposition (EMD) is an adaptive data analysis method proposed by Norden E. Huang et al. in 1998. It is suitable for processing nonlinear and non-stationary signals. In the field of seismic data processing, EMD is widely used because it can effectively extract multi-scale features from signals.

[0003] Seismic wavelet is a signal with a certain starting time, limited energy and a certain duration. It is the basic unit in seismic records. It is generally believed that the seismic wave generated by the earthquake source is only a sharp pulse with a very short duration. As the sharp pulse propagates in the viscoelastic medium, the high-frequency component of the sharp pulse decays quickly, and the waveform grows accordingly, forming a seismic wavelet. A seismic wavelet generally has a duration of 2 to 3 phases, about 90ms, and then propagates underground in the form of a seismic wavelet.

[0004] At present, traditional source wavelet extraction methods often lack sufficient adaptability to capture the various time scale characteristics within the signal when processing signals with complex time-varying characteristics. EMD can adaptively decompose according to the changes in the signal itself, better retaining the nonlinear and non-stationary characteristics of the signal. In addition, seismic signals often contain multiple frequency components and modes. Traditional methods may not be as accurate and effective as EMD in separating these signal components of different physical mechanisms, especially when the source wavelet is highly mixed with other wave components.

[0005] It can be seen that the traditional source wavelet extraction method still has the problems of lack of adaptability and limited multi-component signal processing capabilities. Summary of the invention

[0006] The purpose of the present invention is to provide a method for extracting earthquake source wavelets based on the empirical mode method, which solves the problems of lack of adaptability and limited multi-component signal processing capability.

[0007] To achieve the above object, the present invention provides a method for extracting source wavelets based on the empirical mode method, comprising the following steps:

[0008] S1, signal decomposition, the seismic data is decomposed into multiple layers through the empirical mode decomposition technique to obtain a series of intrinsic mode functions IMF, each of which represents the local fluctuation characteristics of the original signal at different time scales;

[0009] S2, select the appropriate IMF, among all the obtained IMFs, extract the features of each IMF, evaluate the similarity between each IMF and the source wavelet, and select the IMF with the closest features to the source wavelet as the candidate wavelet;

[0010] S3, denoising and smoothing, performing appropriate denoising on the selected IMF to reduce the impact of noise on the source wavelet;

[0011] S4, wavelet correction and synthesis, make necessary corrections to the selected IMF, and appropriately combine the corrected IMF with other IMFs to restore the complete source wavelet shape;

[0012] S5. Verification and Application. Compare the extracted source wavelet with the original data to verify its validity.

[0013] Preferably, in step S1, the specific steps of signal decomposition are as follows:

[0014] S11. Empirical Mode Decomposition , select a section of seismic data as the analysis signal and find all local maximum and minimum points in the signal;

[0015] S12. Use spline interpolation to connect all local maximum and minimum points to form an upper envelope and lower envelope ;

[0016] S13, calculate the upper envelope and lower envelope The average , get the intermediate signal, subtract the intermediate signal from the original signal to get the differential signal ;

[0017] S14, check differential signal Whether the IMF condition is met, if not, the differential signal is used as a new signal to repeat steps S11 to S14 to obtain until Until the conditions are met, define ;

[0018] S15, when the differential signal When the IMF conditions are met, it will be used as Save and subtract from the original signal , obtain the residual signal, and use the residual signal as the new signal, repeating steps S11 to S15 until all IMFs are decomposed or the residual signal meets the stop condition;

[0019] S16. Reconstruct the original seismic data signal by adding all separated IMFs to the last residual signal, and analyze each IMF to obtain instantaneous frequency and amplitude information by performing Hilbert transform.

[0020] Preferably, in step S11, the empirical mode decomposition The expression is as follows:

[0021] ;

[0022] In the formula, is the second intrinsic mode function IMF, is the second intrinsic mode function IMF, is the Nth intrinsic mode function IMF, is a monotonic residual.

[0023] Preferably, in step S15, the stopping condition is that the remaining signal becomes monotonous or its energy is lower than a set threshold.

[0024] Preferably, in step S3, wavelet transform is used to perform denoising on the selected IMF.

[0025] Preferably, in step S4, Hilbert transform is used to perform spectrum analysis on the IMF to check whether there is energy leakage or mode confusion, and then correction is performed by reallocating high-frequency or low-frequency components.

[0026] Preferably, in step S5, a deconvolution operation is used to verify the source wavelet. The specific formula is as follows: ,in, is the deconvolved seismic record, is the original earthquake record, It is the source wavelet.

[0027] Therefore, the present invention adopts a source wavelet extraction method based on the empirical mode method with the above structure, which has the following beneficial effects:

[0028] (1) The present invention does not require pre-setting of basis functions. It identifies and decomposes signals through local averages and local extreme points, and can automatically adapt to the intrinsic characteristics and changing trends of signals, which makes it more flexible and effective in processing complex geological structures and different types of seismic signals.

[0029] (2) The present invention can decompose the original seismic signal into a series of intrinsic mode functions (IMFs), each of which represents the information of a specific frequency band of the signal. This multi-level decomposition helps to more finely identify and separate the source wavelet, especially when the wavelet contains multiple frequency components.

[0030] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 The present invention is a flowchart of a method for extracting source wavelets based on the empirical mode method. DETAILED DESCRIPTION

[0032] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0033] Unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0034] Example

[0035] like Figure 1 As shown, the present invention provides a source wavelet extraction method based on the empirical mode method, comprising the following steps:

[0036] S1, signal decomposition, the seismic data is decomposed into multiple layers through the empirical mode decomposition technique to obtain a series of intrinsic mode functions IMF, each of which represents the local fluctuation characteristics of the original signal at different time scales;

[0037] S2, select the appropriate IMF, among all the obtained IMFs, extract the features of each IMF, evaluate the similarity between each IMF and the source wavelet, and select the IMF with the closest features to the source wavelet as the candidate wavelet;

[0038] S3, denoising and smoothing, performing appropriate denoising on the selected IMF to reduce the impact of noise on the source wavelet;

[0039] S4, wavelet correction and synthesis, make necessary corrections to the selected IMF, and appropriately combine the corrected IMF with other IMFs to restore the complete source wavelet shape;

[0040] S5. Verification and Application. Compare the extracted source wavelet with the original data to verify its validity.

[0041] Preferably, in step S1, the specific steps of signal decomposition are as follows:

[0042] S11. Empirical Mode Decomposition , select a section of seismic data as the analysis signal and find all local maximum and minimum points in the signal;

[0043] S12. Use spline interpolation to connect all local maximum and minimum points to form an upper envelope and lower envelope ;

[0044] S13, calculate the upper envelope and lower envelope The average , get the intermediate signal, subtract the intermediate signal from the original signal to get the differential signal ;

[0045] S14, check differential signal Whether the IMF condition is met, if not, the differential signal is used as a new signal to repeat steps S11 to S14 to obtain until Until the conditions are met, define ;

[0046] S15, when the differential signal When the IMF conditions are met, it will be used as Save and subtract from the original signal , obtain the residual signal, and use the residual signal as the new signal, repeating steps S11 to S15 until all IMFs are decomposed or the residual signal meets the stop condition;

[0047] S16. Reconstruct the original seismic data signal by adding all separated IMFs to the last residual signal, and analyze each IMF to obtain instantaneous frequency and amplitude information by performing Hilbert transform.

[0048] In step S11, the empirical mode decomposition The expression is as follows:

[0049] ;

[0050] In the formula, is the second intrinsic mode function IMF, is the second intrinsic mode function IMF, is the Nth intrinsic mode function IMF, is a monotonic residual.

[0051] In step S15, the stopping condition is that the residual signal becomes monotonous or its energy is lower than a set threshold.

[0052] In step S3, wavelet transform is used to perform denoising on the selected IMF.

[0053] In step S4, the IMF is spectrally analyzed using Hilbert transform to check whether there is energy leakage or mode confusion, and then corrected by redistributing high-frequency or low-frequency components.

[0054] In step S5, the deconvolution operation is used to verify the source wavelet. The specific formula is as follows: ,in, is the deconvolved seismic record, is the original earthquake record, It is the source wavelet.

[0055] The wavelet transform is used to denoise the selected IMF, including the following steps:

[0056] Wavelet decomposition: Perform multi-layer wavelet decomposition on the selected IMF to obtain wavelet coefficients of different scales.

[0057] Thresholding: Apply hard thresholding or soft thresholding to the wavelet coefficients to remove noise. The choice of threshold is critical and can be a fixed threshold, a threshold based on noise estimation, or an adaptive threshold.

[0058] Wavelet reconstruction: Perform inverse wavelet transform on the processed coefficients to reconstruct the denoised IMF.

[0059] Reconstruct the signal: Recombine the denoised IMF with the unprocessed IMF and reconstruct the entire signal through the inverse process of EMD (i.e. direct addition). It should be noted that the denoised IMF should replace the corresponding original IMF, while other IMFs remain unchanged.

[0060] Evaluate denoising effect: Evaluate denoising effect by comparing signal-to-noise ratio, spectral characteristics and other indicators before and after denoising. If necessary, you may need to adjust the threshold strategy or wavelet basis function to achieve the best denoising effect.

[0061] Iterative optimization: For particularly complex noise situations, it may be necessary to iterate the denoising process multiple times and gradually optimize the denoising parameters until you are satisfied.

[0062] Through the above steps, the selected IMF can be effectively denoised by wavelet transform, thereby reducing the impact of noise on the source wavelet and improving the signal quality and analysis accuracy.

[0063] Therefore, the present invention adopts the above-mentioned source wavelet extraction method based on the empirical mode method, which does not need to pre-set basis functions, identifies and decomposes signals through local averages and local extreme points, and can automatically adapt to the intrinsic characteristics and changing trends of signals, which makes it more flexible and effective when processing complex geological structures and different types of seismic signals. And the original seismic signal can be decomposed into a series of intrinsic mode functions, each IMF represents the information of a specific frequency band of the signal. This multi-level decomposition helps to more finely identify and separate the source wavelet, especially when the wavelet contains multiple frequency components.

[0064] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for extracting source wavelets based on the empirical mode method, characterized in that: The following steps are involved: S1, signal decomposition, the seismic data is decomposed into multiple layers through the empirical mode decomposition technique to obtain a series of intrinsic mode functions IMF, each of which represents the local fluctuation characteristics of the original signal at different time scales; In step S1, the specific steps of signal decomposition are as follows: S11. Empirical Mode Decomposition , select a section of seismic data as the analysis signal and find all local maximum and minimum points in the signal; S12. Use spline interpolation to connect all local maximum and minimum points to form an upper envelope and lower envelope ; S13, calculate the upper envelope and lower envelope The average , get the intermediate signal, subtract the intermediate signal from the original signal to get the differential signal ; S14, check differential signal Whether the IMF condition is met, if not, the differential signal is used as a new signal to repeat steps S11 to S14 to obtain until Until the conditions are met, define ; S15, when the differential signal When the IMF conditions are met, it will be used as Save and subtract from the original signal , obtain the residual signal, and use the residual signal as the new signal, repeating steps S11 to S15 until all IMFs are decomposed or the residual signal meets the stop condition; S16, reconstructing the original seismic data signal by adding all separated IMFs to the last residual signal, and analyzing each IMF to obtain instantaneous frequency and amplitude information by performing Hilbert transform; S2, select the appropriate IMF, among all the obtained IMFs, extract the features of each IMF, evaluate the similarity between each IMF and the source wavelet, and select the IMF with the closest features to the source wavelet as the candidate wavelet; S3, denoising and smoothing, performing appropriate denoising on the selected IMF to reduce the impact of noise on the source wavelet; in step S3, wavelet transform is used to perform denoising on the selected IMF; S4, wavelet correction and synthesis, make necessary corrections to the selected IMF, and appropriately combine the corrected IMF with other IMFs to restore the complete source wavelet morphology; in step S4, use Hilbert transform to perform spectrum analysis on the IMF to check whether there is energy leakage or mode confusion, and then correct it by redistributing high-frequency or low-frequency components; S5. Verification and application. Compare the extracted source wavelet with the original data to verify its validity. Use deconvolution operation to verify the source wavelet. The specific formula is as follows [Z(t)=X(t) / S(t)], where Z(t) is the deconvolved earthquake record, X(t) is the original earthquake record, and S(t) is the source wavelet.

2. The method for extracting source wavelets based on the empirical mode method according to claim 1, characterized in that: In step S11, the empirical mode decomposition The expression is as follows: ; In the formula, is the first intrinsic mode function IMF, is the second intrinsic mode function IMF, is the Nth intrinsic mode function IMF, is a monotonic residual.

3. The method for extracting source wavelets based on the empirical mode method according to claim 1, characterized in that: In step S15, the stopping condition is that the residual signal becomes monotonous or its energy is lower than a set threshold.

Citation Information

Patent Citations

  • EEMD-based automatic balancing machine vibration signal processing method

    CN104931197A

  • Time-varying wavelet extraction method based on local similarity and evaluation feedback

    CN105005073A

  • First-arrival travel time picking method and device for seismic signal comprising noise

    CN107272066A

  • AMD-HHT-based time-frequency domain fluid identification method

    CN108051856A

  • Seismic data denoising method combining CEEMD (Complete Ensemble Empirical Mode Decomposition) and generalized S transform

    CN108983286A