Harmonic wavelet domain fitting denoising method and system

Abstract

The application provides a harmonic wavelet domain fitting denoising method and system, and the method comprises the following steps: obtaining a seismic record signal; simulating single-frequency noise in the seismic record signal by using a sine function; solving each parameter in the sine function to obtain a sine simulation function; and removing the single-frequency noise in the seismic record signal by using the sine simulation function to obtain a denoised seismic record signal. The single-frequency noise in the seismic record signal is simulated by using the sine function, the accuracy of fitting the single-frequency noise is ensured, and the signal-to-noise ratio of the denoised seismic record signal is greatly improved.

Description

Technical Field

[0001] This invention relates to the field of seismic signal denoising technology, and more specifically, to a harmonic wavelet domain fitting denoising method and system. Background Technology

[0002] Noise is a significant factor affecting the accuracy of seismic data, and one of the main tasks of seismic data processing is to eliminate various noises in seismic records and maximize the signal-to-noise ratio (SNR) of the seismic data. High-density seismic exploration employs single-point reception, small trace spacing, high trace count, non-combination, and large dynamic range seismic acquisition methods, which, while fully sampling the seismic wavefield, also record a large amount of noise. During field seismic data acquisition, seismic work areas are often located in densely networked oilfield well areas with complex surface conditions, dense buildings, crisscrossing roads, and numerous oil wells, joint stations, oil and gas pipelines, and high-voltage lines. Due to the unique characteristics of these mining and urban environments, various environmental noises are inevitably present in single-shot seismic records. These noises are high in energy and extremely complex in type, severely affecting the quality of seismic data, reducing the SNR of seismic records, impacting the entire data processing process, and limiting further improvements in the quality of seismic profiles. Summary of the Invention

[0003] In order to overcome the shortcomings of the prior art, the purpose of this invention is to provide a harmonic wavelet domain fitting denoising method and system.

[0004] A harmonic wavelet domain fitting denoising method includes:

[0005] Step 1: Acquire seismic record signals;

[0006] Step 2: Use a sine function to simulate the single-frequency noise in the seismic record signal;

[0007] Step 3: Solve for each parameter in the sine function to obtain the sine analog function;

[0008] Step 4: Use the sinusoidal simulation function to remove single-frequency noise from the seismic record signal to obtain the denoised seismic record signal.

[0009] Preferably, in step 2, the sine function is:

[0010]

[0011] Where A is the amplitude. f is the phase, and f is the frequency.

[0012] Preferably, step 3: solving for each parameter in the sine function to obtain the sine analog function includes:

[0013] Step 3.1: Calculate the amplitude of the sine function based on the peak values ​​of each peak point in the seismic record signal;

[0014] Step 3.2: Determine the phase of the sine function based on the phase of the seismic record signal;

[0015] Step 3.3: Determine the frequency of the sine function based on the power spectrum of the earthquake record signal.

[0016] Preferably, step 3.1: calculating the amplitude of the sine function based on the peak values ​​of each peak point in the seismic record signal includes:

[0017] Formula used:

[0018]

[0019] The amplitude of the sinusoidal function is obtained; where AM i Let M be the peak value of the i-th peak point in the seismic record signal, and M be the number of peak points.

[0020] Preferably, step 3.2: determining the phase of the sine function based on the phase of the seismic record signal includes:

[0021] The phase of the seismic record signal is used as the phase of the sine function.

[0022] Preferably, step 3.3: solving for the frequency of the sine function based on the power spectrum of the seismic record signal includes:

[0023] Step 3.3.1: Use the maximum search method to find the local power spectrum maxima;

[0024] Step 3.3.2: Use the formula:

[0025]

[0026] The frequency of the sine function is obtained; where P(k) is the local power spectrum maximum, k is the frequency, N is the length of the local region, λ is the threshold value, K is the wave number, and P(k1) is the power spectrum corresponding to the k1th amplitude.

[0027] The present invention also provides a harmonic wavelet domain fitting denoising system, comprising:

[0028] Seismic recording signal acquisition module, used to acquire seismic recording signals;

[0029] A single-frequency noise simulation module is used to simulate single-frequency noise in the seismic record signal using a sine function;

[0030] The solver module is used to solve for each parameter in the sine function to obtain the sine analog function;

[0031] The denoising module is used to remove single-frequency noise from the seismic record signal using the sinusoidal simulation function to obtain the denoised seismic record signal.

[0032] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps in the above-described harmonic wavelet domain fitting denoising method.

[0033] The beneficial effects of the harmonic wavelet domain fitting denoising method and system provided by the present invention are as follows: Compared with the prior art, the present invention uses a sine function to simulate single-frequency noise in seismic record signals, which can ensure the accuracy of fitting single-frequency noise and greatly improve the signal-to-noise ratio of the denoised seismic record signals.

[0034] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 The flowchart of a harmonic wavelet domain fitting denoising method provided by an embodiment of the present invention is shown;

[0037] Figure 2 The original single-shot record of actual seismic data provided in the embodiments of the present invention is shown;

[0038] Figure 3 This illustrates the single-shot record after conventional six-segment denoising provided in an embodiment of the present invention;

[0039] Figure 4 This shows the single-shot record after harmonic wavelet domain fitting and denoising provided in an embodiment of the present invention;

[0040] Figure 5 The original stacked profile of actual seismic data provided in the embodiments of the present invention is shown;

[0041] Figure 6 The superimposed profile after conventional six-part denoising provided in the embodiment of the present invention is shown;

[0042] Figure 7The superimposed profile after harmonic wavelet domain fitting and denoising provided in the embodiment of the present invention is shown;

[0043] Figure 8 The signal-to-noise ratio (SNR) of the original superimposed profile provided in the embodiments of the present invention is shown (SNR > 2 accounts for 9%, SNR > 3 accounts for 0%).

[0044] Figure 9 The signal-to-noise ratios after conventional six-way denoising provided in this embodiment of the invention are shown (signal-to-noise ratio >2 accounts for 70%, signal-to-noise ratio >3 accounts for 35%).

[0045] Figure 10 The signal-to-noise ratios after harmonic wavelet domain fitting and denoising provided in the embodiments of the present invention are shown (signal-to-noise ratio >2 accounts for 93%, signal-to-noise ratio >3 accounts for 65%). Detailed Implementation

[0046] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0047] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0048] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0049] Please refer to the following: Figure 1-10 A harmonic wavelet domain fitting denoising method, comprising:

[0050] Step 1: Acquire seismic record signals;

[0051] Step 2: Simulate the single-frequency noise in the seismic record signal using a sine function; the sine function is:

[0052]

[0053] Where A is the amplitude. f is the phase, and f is the frequency.

[0054] Step 3: Solve for each parameter in the sine function to obtain the sine analog function;

[0055] Furthermore, step 3 includes:

[0056] Step 3.1: Calculate the amplitude of the sine function based on the peak values ​​of each peak point in the seismic record signal; specifically, this invention can use the following formula:

[0057]

[0058] The amplitude of the sinusoidal function is obtained; where AM i Let M be the peak value of the i-th peak point in the seismic record signal, and M be the number of peak points.

[0059] Step 3.2: Determine the phase of the sine function based on the phase of the seismic record signal; In this embodiment of the invention, the phase of the seismic record signal can be used as the phase of the sine function.

[0060] Step 3.3: Determine the frequency of the sine function based on the power spectrum of the earthquake record signal.

[0061] Furthermore, step 3.3 includes:

[0062] Step 3.3.1: Use the maximum search method to find the local power spectrum maxima;

[0063] Step 3.3.2: Use the formula:

[0064]

[0065] The frequency of the sine function is obtained; where P(k) is the local power spectrum maximum, k is the frequency, N is the length of the local region, λ is the threshold value, K is the wave number, and P(k1) is the power spectrum corresponding to the k1th amplitude.

[0066] Step 4: Use the sinusoidal simulation function to remove single-frequency noise from the seismic record signal to obtain the denoised seismic record signal.

[0067] The noise reduction principle of the present invention will be explained below with reference to specific embodiments:

[0068] Assuming the harmonic wavelet is a complex wavelet, its frequency domain expression is:

[0069]

[0070] Its time-domain representation is:

[0071]

[0072] Harmonic wavelets have a box-shaped characteristic, which makes them less prone to spectral leakage during frequency domain transformation. However, they can exhibit the Gipps phenomenon in the time domain. To overcome this drawback, the edge-finding method is generally used in practice.

[0073] Extending formula (1) gives:

[0074]

[0075] The corresponding time-domain representation is:

[0076] w m,n (x)=(e i2πnx -e i2mπx ) / (i2(nm)πx) (4)

[0077] In the formula, m and n are the lower and upper limits of the harmonic wavelet transform. m and n are real numbers, and nm>0. When n=2 and m=1, the above formula becomes formula (2).

[0078] Given the displacement step size k / (nm) of the harmonic wavelet, formula (4) becomes:

[0079]

[0080] Equation (5) is the general expression for harmonic wavelets with a bandwidth of (nm) and an analysis center at x = k / (nm). Clearly, equation (5) is formally identical to that of the binary wavelet. The widespread application of harmonic wavelets stems not only from their numerous advantages but also from their fast algorithm.

[0081] The single-frequency noise on the seismic profile represents the impact of oilfield facilities such as high-voltage lines, oil wells, and pumping stations operating normally near the seismic acquisition site on the seismic acquisition process. These facilities have relatively stable transmission frequencies and amplitudes during transmission, a reality that provides the basis for this invention to simulate using a sine function. With the sine simulation function, this information can be subtracted from the seismic record, resulting in a seismic record free of single-frequency noise industrial interference. However, any denoising process inevitably damages the effective wave to varying degrees. To minimize the impact on the effective signal during single-frequency noise removal, the denoising process should be performed within a finite frequency band. Due to the arbitrary frequency division capability and frequency leakage-free advantage of harmonic wavelet transform, this invention uses harmonic wavelet packet transform to divide the signal into a finite frequency band with single-frequency noise, placing the sine function simulation and subtraction operation within this finite frequency band. This approach reduces the impact on the effective wave during the denoising process.

[0082] 1. Sine function to simulate single-frequency noise

[0083] For seismic records containing single-frequency noise, they can be represented as:

[0084] x(t)=v(t)+s(t), t=Δt,2Δt,…,Δt·N, (6)

[0085] In formula (6), Δt is the sampling interval, N is the channel length, v(t) is the single-frequency noise, and s(t) is the signal. s(t) may also contain other interferences; for ease of discussion, these other interferences are set to zero. According to the propagation law of single-frequency noise, v(t) can be written in the form of a sine function.

[0086]

[0087] In formula (7), A is the amplitude of the sine function. Given its phase, the following question is how to determine the amplitude A and the phase.

[0088] Let AM i These are the peak values ​​of each peak point in the seismic record (signal). These peak values ​​contain the contribution of v(t), AMV. i It also contains the contribution of signal s(t) in AMS. i To find its mean, we have:

[0089]

[0090] In formula (8), M represents the number of peak points. Considering the function formed by the peaks of s(t) along the vertical direction as a random function with zero mean, we have:

[0091]

[0092] A represents the amplitude of the single-frequency noise to be determined.

[0093] Phase of a simulated sine function A simple method is used to obtain this value. Since single-frequency noise dominates in seismic records with single-frequency noise, its phase can be considered as the phase of the single-frequency noise. Therefore:

[0094]

[0095] N is the same as equation (7). The sine function with zero phase is v'(t)=A·sin(2πf·j·Δt). Let the time positions of the first peak points of these two functions be i·Δt and j·Δt respectively, then we should have:

[0096]

[0097]

[0098] This yields the required phase. Amplitude A and phase Substituting into equation (7), we obtain the simulation function v(t).

[0099] 2. Frequency determination of single-frequency noise

[0100] The most significant characteristic of single-frequency noise is that its power spectrum in the frequency domain is significantly higher than that of other frequencies within a certain local region. Based on this characteristic, the frequency of single-frequency noise can be determined very conveniently.

[0101] The input record s(t) is transformed to the frequency domain using FFT to obtain S(f), and its power spectrum P(f) is calculated. Local power spectrum maxima are found using a maximum search method, and the frequency of the single-frequency noise can be determined by applying a ratio method among these maxima. Specifically:

[0102]

[0103] Let P(k) be the local power spectral maximum, k be the frequency, N be the length of the local region, and λ be a threshold value. Then k is the frequency of a single-frequency noise.

[0104] This invention provides a harmonic wavelet domain fitting denoising method, comprising: analyzing the characteristics of different noises within the frequency range; simulating correlated noise using a sine function; performing wavelet transform on the seismic record signal; fitting various noises in different wavelet domains; and subtracting or suppressing the noise in the corresponding wavelet domain to obtain the denoised seismic record signal. The advantage of this method is that it leverages the excellent time-frequency characteristics of wavelet transform for frequency division. After transforming the original data to the wavelet domain, the characteristics of various noises become more prominent, allowing for a more accurate fitting of various environmental noises. Compared to conventional shot domain methods, wavelet domain methods remove hyperbolic noise while minimizing damage to the effective signal.

[0105] Figure 2 This is a typical single-shot unit within the study area, and various types of interference are evident. Figure 3 This is the result after using the conventional six-segment denoising method. As can be seen from the comparison, most of the noise is suppressed to a certain extent, and the effective waves submerged in the noise are revealed, but some noise still remains on the denoised single-shot. Figure 4 The single-shot recordings are based on conventional six-part denoising and then enhanced with wavelet frequency division denoising. This demonstrates further noise reduction and a significantly improved signal-to-noise ratio. This is also evident in the comparison of the superimposed profiles before and after denoising using the old and new methods. Figures 5-7 ).

[0106] Figures 8-10 Analysis using different denoising signal-to-noise ratios shows that the signal-to-noise ratio of the data is significantly improved after comprehensive denoising. This effectively solves the special interference problems of oilfield facilities and other factors in high-density seismic data in densely networked oilfield well areas. Compared with previous processing methods, the signal-to-noise ratio is further improved, and the signal-to-noise ratio of the processed data basically reaches above 2.

[0107] The present invention also provides a harmonic wavelet domain fitting denoising system, comprising:

[0108] Seismic recording signal acquisition module, used to acquire seismic recording signals;

[0109] A single-frequency noise simulation module is used to simulate single-frequency noise in the seismic record signal using a sine function;

[0110] The solver module is used to solve for each parameter in the sine function to obtain the sine analog function;

[0111] The denoising module is used to remove single-frequency noise from the seismic record signal using the sinusoidal simulation function to obtain the denoised seismic record signal.

[0112] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps in the above-described harmonic wavelet domain fitting denoising method.

[0113] The present invention also provides a computer-readable storage medium storing a computer program thereon, characterized in that, when the computer program is executed by a processor, it implements the steps in the aforementioned harmonic wavelet domain fitting denoising method. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present invention are the same as the beneficial effects of the harmonic wavelet domain fitting denoising method described above, and will not be repeated here.

[0114] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for fitting denoising in harmonic wavelet domain, characterized in that, The method comprises the following steps: Step 1: obtaining a seismic record signal; Step 2: simulating single-frequency noise in the seismic record signal by using a sine function; Step 3: solving each parameter in the sine function to obtain a sine simulation function; Step 4: removing the single-frequency noise in the seismic record signal by using the sine simulation function to obtain a denoised seismic record signal; In the step 2, the sine function is as follows: wherein A is the amplitude of the sinusoidal function, is the phase of the sinusoidal function, f is the frequency of the sinusoidal function, and t is the time of the recorded signal at a certain time. The step 3: solving each parameter in the sine function to obtain a sine simulation function, comprises the following steps: Step 3.1: solving the amplitude of the sine function according to the peak value of each peak point in the seismic record signal; Step 3.2: solving the phase of the sine function according to the phase of the seismic record signal; Step 3.3: solving the frequency of the sine function according to the power spectrum of the seismic record signal; The step 3.1: solving the amplitude of the sine function according to the peak value of each peak point in the seismic record signal, comprises the following steps: The formula is as follows: obtaining an amplitude of the sinusoidal function; wherein AM i is a peak value of the i th peak point in the seismic record signal, and M is the number of peak points. The step 3.2: solving the phase of the sine function according to the phase of the seismic record signal, comprises the following steps: The phase of the seismic record signal is taken as the phase of the sine function; The step 3.3: solving the frequency of the sine function according to the power spectrum of the seismic record signal, comprises the following steps: Step 3.3.1: using a maximum value search method to solve the local power spectrum maximum value; Step 3.3.2: using the formula: a frequency of the sinusoidal function is obtained; wherein P(k) is a local power spectrum maximum, k is a frequency, N is a length of a local region, λ is a threshold value, K is a wave number, P (k1) is a power spectrum corresponding to the k1th amplitude.

2. A denoising system based on the harmonic wavelet domain fitting denoising method of claim 1, characterized in that, The method comprises the following steps: A seismic record signal obtaining module is configured to obtain a seismic record signal; A single-frequency noise simulating module is configured to simulate single-frequency noise in the seismic record signal by using a sine function; A solving module is configured to solve each parameter in the sine function to obtain a sine simulation function; A denoising module is configured to remove the single-frequency noise in the seismic record signal by using the sine simulation function to obtain a denoised seismic record signal.

3. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by a processor to realize the steps in the harmonic wavelet domain fitting denoising method according to claim 1.

Images