Seismic signal filtering method, medium and equipment

Abstract

The invention relates to the technical field of signal filtering, in particular to a seismic signal filtering method, a medium and equipment, and the method comprises the steps: carrying out the modal decomposition of a target signal, disassembling a complex mixed seismic signal into single-feature sub-modal signals, and carrying out the modal decomposition of the sub-modal signals; a clear sub-mode screening standard is established through the average instantaneous energy of the sub-mode signals in a preset time window, and a non-noise mode and a noise mode are accurately distinguished; dynamic control of an iteration process is realized through iteration updating and an iteration stopping condition based on a cross-correlation degree, excessive iteration or insufficient iteration caused by fixed iteration times is avoided, and the quality of non-noise modal sub-signals is guaranteed; all non-noise mode sub-mode signals are superposed, and effective signal components of different iteration levels are aggregated, so that the effective characteristics of each level of the output seismic effective signal are completely reserved, the noise interference is greatly reduced, and the purity of the seismic effective signal is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of signal filtering, in particular to a seismic signal filtering method, medium and equipment. BACKGROUND

[0002] Seismic signal filtering is a core link of seismic exploration data processing, and its goal is to separate surface interference, instrument noise, surface wave and other noises from reflection wave and refracted wave, etc. effective signals in noisy seismic signals, to provide high-quality data for subsequent reservoir prediction and structure interpretation.

[0003] The existing seismic signal filtering method usually relies on fixed parameter modal decomposition means, such as empirical mode decomposition and ensemble empirical mode decomposition. When facing shallow high main frequency seismic signals (main frequency > 50Hz), due to the fixed decomposition step, the effective signal is over-decomposed into many sub-modalities, resulting in the loss of part of the signal characteristics due to fragmentation, and reducing the accuracy of subsequent seismic signal analysis. For deep low signal-to-noise ratio seismic signals (signal-to-noise ratio < 3), the fixed initial decomposition layer cannot deeply excavate the weak effective signal embedded in the noise, resulting in the inability to identify in the subsequent screening link, which seriously affects the judgment of deep geological structure. In addition, the existing decomposition technology does not specially design an effective separation mechanism for the jump interference such as surface sudden vibration and instrument pulse in the seismic signal, so that these jump interferences are easily mixed into the effective sub-modalities, reducing the purity of the sub-modalities after decomposition, and interfering with the subsequent judgment of the effective seismic signal.

[0004] Therefore, how to adjust the modal decomposition method and parameters according to the dynamic characteristics of the seismic signal, avoid over-decomposition of the effective signal or burying of the weak effective signal, and simultaneously separate the jump interference to improve the purity of the sub-modalities after filtering has become a problem to be solved. SUMMARY

[0005] In order to solve the above technical problems, the technical scheme adopted by the present application is a seismic signal filtering method, which comprises the following steps: S1, initializing i=1, and determining the initial seismic signal as the i-th level target signal.

[0006] S2, for any i-th level target signal, performing modal decomposition processing on the current i-th level target signal to obtain a plurality of i+1-th level sub-modal signals corresponding to the current i-th level target signal, wherein the sub-modal signal is a signal component with amplitude modulation and frequency modulation characteristics.

[0007] S3, according to the average instantaneous energy of each i+1-th level sub-modal signal in each preset time window, obtaining the signal type corresponding to each i+1-th level sub-modal signal, wherein the signal type is a noise mode or a non-noise mode.

[0008] S4, each i+1th sub-modal signal of the signal type of the non-noise mode is determined as an i+1th target signal, and a cross-correlation degree between each i+1th target signal and an ith target signal generating the i+1th target signal is obtained.

[0009] S5, if the cross-correlation degrees corresponding to all i+1th target signals are greater than a preset degree threshold, all sub-modal signals of the signal type of the non-noise mode are obtained, and step S6 is executed, otherwise, i is updated as i+1, and step S2 is repeatedly executed.

[0010] S6, all sub-modal signals of the signal type of the non-noise mode are superimposed to obtain a filtered seismic effective signal.

[0011] The application further provides a non-transitory computer readable storage medium, and at least one instruction or at least one program is stored in the non-transitory computer readable storage medium, the at least one instruction or the at least one program is loaded and executed by a processor to implement the seismic signal filtering method.

[0012] The application further provides an electronic device, which comprises a processor and the non-transitory computer readable storage medium.

[0013] The application has at least the following beneficial effects: the i-th target signal is subjected to modal decomposition, the complex mixed seismic signal is disassembled into a single characteristic sub-modal signal, the problem of high difficulty in traditional mixed signal separation is broken through, the core amplitude modulation frequency modulation characteristics of the effective signal are retained, and the effective signal is prevented from being covered by noise due to disassembly; the explicit sub-modal screening standard is established by the average instantaneous energy of the sub-modal signal in the preset time window, the local energy analysis is realized by means of the preset time window, the weak effective signal misjudgment caused by the overall energy averaging is avoided, and the non-noise mode and the noise mode are accurately distinguished; the similarity of the upper and lower target signals is quantified by calculating the cross-correlation degree, data basis is provided for subsequent judgment of whether the decomposition is reasonable and whether the effective characteristics are retained, and the effective signal characteristics are prevented from being broken due to excessive decomposition; the iteration process is dynamically controlled by iteration updating and iteration stopping conditions based on the cross-correlation degree, excessive iteration or insufficient iteration caused by fixed iteration times is avoided, it is ensured that the non-noise modal sub-signals finally obtained all have stable effective characteristics, there is no insufficiently purified signal residue, and the subsequent reconstruction quality is ensured; the filtered seismic effective signal is obtained by superimposing all non-noise modal sub-modal signals, the effective signal components of different iteration levels are aggregated, the closed loop from sub-modal disassembly to complete effective signal restoration is realized, and the finally output seismic effective signal not only retains the effective characteristics of each level, but also greatly reduces the noise interference, and the purity of the seismic effective signal is improved. BRIEF DESCRIPTION OF DRAWINGS

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0015] Figure 1 This is a flowchart of a seismic signal filtering method provided in Embodiment 1 of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It is understood that, where appropriate, the terms used to distinguish similar objects can be interchanged so that the invention can also be implemented in other embodiments besides the illustrated or described embodiments. Furthermore, the terms "including," "having," and any variations are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products, or devices.

[0018] Example 1 This embodiment provides a method for filtering seismic signals, such as... Figure 1 As shown, the seismic signal filtering method includes the following steps: S1, initialize i=1, and determine the initial seismic signal as the i-th level target signal.

[0019] The initial seismic signal is the seismic signal collected by the seismic exploration equipment. Its waveform includes effective seismic signals such as reflected waves and refracted waves, as well as various noise and jump interference components such as surface interference, instrument noise, and surface waves. It is the original input data of the filtering method.

[0020] The seismic signal filtering needs to separate noise and effective signal step by step through multiple rounds of iteration, and the iteration process needs to clearly define the object of each round of decomposition, that is, the target signal. The initial seismic signal is defined as the first-level target signal, and in each subsequent iteration, the non-noise mode determined to be further decomposed will become the next-level target signal, forming a multi-level decomposition system to ensure that the iteration process is orderly and traceable.

[0021] S2, for any ith-level target signal, the current ith-level target signal is subjected to modal decomposition processing to obtain a plurality of (i+1)th-level sub-modal signals corresponding to the current ith-level target signal, wherein the sub-modal signal is a signal component with amplitude modulation and frequency modulation characteristics.

[0022] The modal decomposition processing is based on non-stationary signal analysis theory, and through a preset decomposition algorithm (such as a jump-enhanced amplitude modulation and frequency modulation modal decomposition), the complex ith-level target signal is decomposed into a plurality of sub-modal signals with independent signal characteristics such as frequency range and amplitude variation law. Each level of sub-modal signal is a denoising optimization of the previous level of target signal, and the effective signal proportion of the (i+1)th-level sub-modal signal is higher than that of the ith-level target signal. Through multiple rounds of decomposition iteration, the effective signal is finally highly purified.

[0023] The (i+1)th-level sub-modal signal is the next level of sub-modal signal generated after the modal decomposition of the ith-level target signal, and each sub-modal signal has independent signal characteristics such as frequency range and amplitude variation law. Each level of sub-modal signal is a denoising optimization of the previous level of target signal, and the effective signal proportion of the (i+1)th-level sub-modal signal is higher than that of the ith-level target signal. Through multiple rounds of decomposition iteration, the effective signal is finally highly purified.

[0024] The amplitude modulation and frequency modulation characteristics are the combination of the two core dynamic characteristics of amplitude modulation and frequency modulation of the seismic signal. The amplitude modulation refers to the periodic or regular change of the instantaneous amplitude of the signal with time, such as the periodic fluctuation of the amplitude of the seismic reflection wave with the reflection coefficient of the geological interface; the frequency modulation refers to the dynamic adjustment of the instantaneous frequency of the signal with time, such as the gradual decrease of the main frequency from 50Hz to 5Hz due to the density difference of the geological horizon from shallow to deep, showing a stable frequency decreasing trend; the noise has no stable amplitude modulation or frequency modulation rule, and the combination of the two forms the amplitude modulation and frequency modulation characteristics, which is a distinctive property of the effective seismic signal distinguishing from the noise. In modal decomposition, whether the signal has amplitude modulation and frequency modulation characteristics is used as the core screening basis to ensure that the (i+1)th-level sub-modal signal generated by decomposition is an effective signal candidate, reducing the mixing of noise from the source, and retaining the core waveform information of the effective signal. When stacking such sub-modal signals, the waveform characteristics of the initial seismic signal can be restored to the greatest extent, avoiding distortion of the reconstructed signal.

[0025] The above, by the i-th target signal execution modal decomposition processing, the complex non-stationary seismic signal is disassembled into several single characteristic sub-modal signals, so that the separation of effective signal and noise is changed from the whole mixed signal separation to the individual sub-modal separation, which greatly reduces the separation difficulty and improves the separation precision; by limiting the sub-modal signal to be a signal component with amplitude modulation and frequency modulation characteristics, the noise modal without amplitude modulation and frequency modulation characteristics is eliminated from the decomposition link, so that the generated i+1-th sub-modal signal is an effective signal candidate, which reduces the invalid operation in the subsequent screening link and improves the filtering efficiency.

[0026] In a specific embodiment, the modal decomposition processing is a jump-enhanced amplitude modulation and frequency modulation modal decomposition processing, and S2 includes the following steps: S21, extracting the key prior feature of the current i-th target signal, wherein the key prior feature includes a main frequency range and a signal-to-noise ratio.

[0027] S22, determining the iteration step and the initial decomposition layer number N in the jump-enhanced amplitude modulation and frequency modulation modal decomposition processing according to the key prior feature.

[0028] S23, performing the jump-enhanced amplitude modulation and frequency modulation modal decomposition processing according to the iteration step and the initial decomposition layer number N to obtain a plurality of i+1-th sub-modal signals and a jump interference component, wherein the determination basis of the jump interference component is that the peak amplitude to average amplitude ratio of the current i-th target signal is greater than a preset ratio threshold value, and the peak duration is less than a preset time threshold value.

[0029] The main frequency range is the main frequency distribution interval of the effective signal in the i-th target signal, which can be calculated by power spectrum analysis or wavelet transform, and is expressed as a range from the minimum main frequency to the maximum main frequency, for example, the main frequency range of the shallow signal is 50Hz to 80Hz, and the main frequency range of the deep signal is 8Hz to 20Hz, which is a core index reflecting the frequency characteristics of the effective signal. The main frequency range is used to guide the setting of the iteration step, and by extracting the main frequency range feature, the decomposition method can automatically adapt to the shallow and deep exploration scenes, improving the universality.

[0030] The signal-to-noise ratio is the ratio of the effective signal power to the noise signal power in the i-th target signal, which is usually expressed in decibels (dB), and the signal-to-noise ratio SNR=10lg(P valid / P noise ), wherein P valid is the effective signal power, P noise is the noise power, and reflects the proportion of the effective component in the signal. The jump interference in the low signal-to-noise ratio signal is easy to be confused with the weak effective signal, so the preset ratio threshold value needs to be reduced (such as from 5 to 4) to improve the interference recognition rate; the interference characteristics are more obvious in the high signal-to-noise ratio signal, so a higher ratio threshold value can be maintained to reduce the misjudgment of the effective signal.

[0031] The iteration step determines the fitting accuracy of the decomposed sub-modal signals. The smaller the iteration step, the more accurate the fitting, but the lower the efficiency. The initial decomposition layer number N determines the depth of the decomposition of the seismic signal. The more the initial decomposition layer number, the more the deep nested signal can be separated, but the higher the calculation cost. By establishing a quantitative mapping relationship between the main frequency range-iteration step and the signal-to-noise ratio-initial decomposition layer number, dynamic adaptation of the decomposition parameters can be achieved. The quantitative mapping relationship is a corresponding rule established by experts through experimental data and theoretical derivation, specifically: The higher the main frequency, the greater the frequency change rate of the effective signal, and the smaller the iteration step is needed to capture high-frequency details. For example, high main frequency signals (such as >50Hz) change rapidly in frequency and require a small iteration step (such as 0.005) to avoid over-decomposing the effective signal. Low main frequency signals (such as <20Hz) have a smooth frequency change and can appropriately increase the iteration step (such as 0.01) to improve the decomposition efficiency. The iteration step corresponding to the main frequency signal in the [20Hz, 50HZ] range can be set to 0.008 to balance the signal decomposition degree.

[0032] The lower the signal-to-noise ratio, the deeper the effective signal is masked by noise, and the more initial layers are needed to fully separate the noise. For example, in a low signal-to-noise ratio signal (SNR < 3dB), the effective signal is deeply masked by noise, and the initial decomposition layer number (such as 5 layers) needs to be increased to fully separate the nested weak effective signal. In a high signal-to-noise ratio signal (SNR > 10dB), the effective signal has a high proportion, and the initial decomposition layer number (such as 3 layers) can be reduced to avoid redundant calculation. The initial decomposition layer number corresponding to the signal-to-noise ratio signal in the [3dB, 10dB] range can be set to 4 to balance the decomposition degree and calculation efficiency.

[0033] Jump interference components are interference signal segments in seismic signals that are caused by sudden factors such as surface vehicle passing, instrument collision, do not have amplitude modulation and frequency modulation characteristics, and have short duration (usually <0.01 seconds) and high amplitude (usually 3-8 times the average amplitude of the signal). It is the main source of sudden interference that affects the quality of the effective signal and needs to be separated and removed.

[0034] Jump Plus AM-FM Mode Decomposition (JMD) is different from traditional AM-FM mode decomposition in that it simultaneously extracts effective AM-FM sub-modalities and separates jump interference. Jump interference in seismic signals has unique characteristics of short duration and high amplitude, which are in sharp contrast to the long and stable AM-FM characteristics of effective signals, and can be accurately identified through amplitude ratio + duration double threshold.

[0035] In the execution process of JMD, first, based on the iteration step and the initial decomposition layer number N, the amplitude modulation frequency modulation sub-modal fitting (such as fitting the instantaneous amplitude and frequency function by the least square method) of the i-th level target signal is performed to generate a preliminary sub-modal; then, for the peak value segment in the preliminary sub-modal, the jump interference component is screened out and separated through the double conditions of peak value amplitude / average amplitude> preset ratio threshold (determination of high amplitude) and peak value duration< preset time threshold (determination of short duration); finally, the pure effective amplitude modulation frequency modulation sub-modal (i.e. the i+1-th level sub-modal signal) and the separate jump interference component are output, realizing the complete separation of effective signal and interference classification.

[0036] The preset ratio threshold is an amplitude determination standard for determining whether the peak value is a jump interference, which is usually an integer in the range of 3-8, preferably 5. The calculation method is jump interference peak value amplitude / current i-th level target signal average amplitude. When the ratio exceeds the preset ratio threshold, it is determined that the peak value has the characteristic of high amplitude and is one of the necessary conditions of the jump interference.

[0037] The preset time threshold is a duration determination standard for determining whether the peak value is a jump interference, which is a short time value, usually in the range of 0.001-0.01 seconds, preferably 0.005 seconds. The calculation method is the time when the peak value amplitude lasts more than 1.5 times the average amplitude. When the time is less than the preset time threshold, it is determined that the peak value has the characteristic of short duration and is another necessary condition of the jump interference.

[0038] The above, by determining the iteration step through the main frequency range, the fitting accuracy of the decomposed sub-modal is matched with the frequency change characteristics of the effective signal, avoiding the fitting error caused by too large step length of high main frequency signal, and avoiding the efficiency waste caused by too small step length of low main frequency signal; by determining the initial decomposition layer number N through the signal-to-noise ratio, the decomposition depth is matched with the masking degree of the effective signal in the signal, the weak effective signal can be separated by a larger N in the low signal-to-noise ratio signal, and a smaller N can be used to control the cost in the high signal-to-noise ratio signal, realizing the decomposition on demand; by executing the decomposition through the iteration step and the initial decomposition layer number N, the fitting accuracy of the effective amplitude modulation frequency modulation sub-modal is adapted to the signal characteristics, avoiding over-decomposition or under-decomposition, and providing a high-quality signal basis for subsequent signal reconstruction.

[0039] In a specific embodiment, S23 includes the following steps: S231, initializing the core decomposition parameters according to the current i-th level target signal, wherein the core decomposition parameters include the decomposition window length, the modal convergence threshold and the jump interference detection window.

[0040] S232, performing N-layer cascaded decomposition on the current i-th level target signal according to the initial decomposition layer number N and the core decomposition parameters, obtaining one preliminary sub-modal signal of each layer decomposition output and a residual component of the N-th layer decomposition output.

[0041] S233, comparing the peak amplitude and the duration of the peak of each preliminary sub-modal signal and residual component with the preset ratio threshold and the preset time threshold, extracting the jump interference component from all preliminary sub-modal signals and residual components.

[0042] S234, removing the signal segment time-overlapped with the jump interference component from the preliminary sub-modal signal and the residual component, obtaining a plurality of i+1 level sub-modal signals corresponding to the current i level target signal.

[0043] Among them, the decomposition window length is the time window length used in decomposition to intercept the target signal and perform local amplitude modulation frequency modulation modal fitting, in the number of sampling points, which needs to be dynamically calculated according to the main frequency of the current i level target signal and the sampling rate, to ensure that the window can cover at least 2 complete effective signal periods, avoid signal fragmentation fitting error caused by too short window, or time resolution decline caused by too long window, to protect the local fitting precision. For example, when the main frequency is 50Hz and the sampling rate is 1000Hz, the window length is 40 sampling points, corresponding to 0.04 seconds, covering 2 50Hz signal periods.

[0044] The modal convergence threshold is a quantitative standard for determining whether the AM-FM sub-modal fitting has reached a stable state, which is represented by the root mean square error of the sub-modal signals generated by adjacent two iterations, and the value range is [10 -4 ,10 -3 ], preferably 10 -4 When the root mean square error is less than the modal convergence threshold, it is considered that the sub-modal fitting converges, and the iteration fitting of the current level is stopped.

[0045] The jump interference detection window is a time window for sliding detection of short duration and high amplitude jump interference in the signal, in seconds or sampling points, usually 0.001-0.01 seconds in length, preferably 0.005 seconds, with a sliding step of 1 sampling point, to ensure that the sudden interference is captured without omission, avoid mistaking the effective signal peak as interference due to too long window, or the interference is divided into multiple segments due to too short window, to accurately locate the time domain position of the interference and improve the efficiency of interference detection.

[0046] The current ith level target signal is a mixture of effective AM-FM signals and noise and jump interference, and single decomposition cannot completely separate the deeply nested effective components. Through N-level cascade decomposition, the signal components can be peeled off layer by layer in the order of frequency from high to low (or energy from strong to weak), and the highest frequency (or the strongest energy) AM-FM submode is extracted by the first layer of decomposition, the second layer extracts the second highest frequency (or the second strongest energy) AM-FM submode from the residual signal after the first layer of decomposition, and so on until the Nth layer, and finally N preliminary submodes and one residual component are formed. In this process, the initial decomposition layer number N determines the peeling depth, and the core decomposition parameter ensures the consistency of the accuracy of each layer of decomposition. Through hierarchical operation, the problem that the effective signal is covered by noise caused by single decomposition can be avoided, and the effective AM-FM submode is gradually separated from the mixed signal, providing a preliminary submode with more single characteristics for subsequent jump interference extraction.

[0047] The person skilled in the art knows that the N-level cascade decomposition of the target signal by the JMD method in the prior art falls within the protection scope of the present application, and will not be repeated here.

[0048] The above, by initializing the decomposition window length, the mode convergence threshold and the jump interference detection window, and serving the N-level cascade decomposition, the separation rate of the deeply nested low-frequency effective signal is improved, the weak effective signal missing extraction problem caused by high-frequency noise covering in single decomposition is avoided, and the fitting accuracy and efficiency of the sub-signal mode are improved; through the comparison of the amplitude and the duration of the double threshold, the false positive rate is reduced, the signal separation rate is improved, and high-purity effective submode is ensured, which provides high-quality data support for subsequent signal screening and reconstruction.

[0049] S3, according to the average instantaneous energy of each ith+1 level submode signal in each preset time window, the signal type corresponding to each ith+1 level submode signal is obtained, wherein the signal type is noise mode or non-noise mode.

[0050] Wherein, the preset time window is a fixed time segment set for analyzing the average instantaneous energy characteristics of the submode signal, avoiding the energy characteristic averaging caused by directly analyzing the whole submode signal, and the signal is divided into multiple local segments by windowing, and the energy characteristics of each segment are more easily distinguished.

[0051] The average instantaneous energy is the local energy value of the ith+1 level submode signal in the preset time window, reflecting the energy intensity of the submode signal in the local time segment, and is the core quantitative index for distinguishing noise and non-noise mode.

[0052] The noise mode is a submode unit in the ith+1 level submode signal, which presents irregular random fluctuation in average instantaneous energy and does not have the characteristics of AM-FM effective signal, mainly including surface wave interference, instrument background noise, surface low-frequency interference, etc.

[0053] Non-noise modal is a sub-modal unit in the i+1th sub-modal signal, which has a stable periodic change in average instantaneous energy and typical AM-FM effective signal characteristics, mainly including effective signal components such as seismic reflection wave, refracted wave, first wave, etc.

[0054] The above, by setting the preset time window adapted to the main frequency of the sub-modal, the average instantaneous energy analysis is focused on the local segment containing the complete effective signal period, avoiding the noise characteristics being covered due to the too long time window, or the incomplete effective signal energy analysis due to the too short time window; the average instantaneous energy characteristics corresponding to the preset time window can distinguish the noise modal and the non-noise modal, which can accurately identify the stable energy characteristics of the weak effective signal and reduce the misjudgment rate of the signal type.

[0055] In a specific embodiment, S3 includes the following steps: S31, for any i+1th sub-modal signal, performing Hilbert transform on the current i+1th sub-modal signal to obtain the Hilbert spectrum corresponding to the current i+1th sub-modal signal.

[0056] S32, according to the Hilbert spectrum corresponding to the current i+1th sub-modal signal, obtaining the average instantaneous energy of the current i+1th sub-modal signal in each preset time window, wherein the length of the preset time window is determined according to the main frequency of the current i+1th sub-modal signal.

[0057] S33, according to all the average instantaneous energies corresponding to the current i+1th sub-modal signal, obtaining the kurtosis coefficient, the instantaneous frequency standard deviation and the energy entropy value corresponding to the current i+1th sub-modal signal.

[0058] S34, obtaining the preset kurtosis coefficient threshold, the preset standard deviation threshold and the preset entropy value threshold corresponding to the current i+1th sub-modal signal.

[0059] S35, if the kurtosis coefficient of the current i+1th sub-modal signal is greater than or equal to the preset kurtosis coefficient threshold, the instantaneous frequency standard deviation is less than the preset standard deviation threshold, and the energy entropy value is less than the preset entropy value threshold, it is determined that the current i+1th sub-modal signal is a non-noise modal.

[0060] The Hilbert transform is a mathematical operation for converting real signals into analytic signals, which is used to break through the limitations of traditional frequency domain analysis and accurately extract the instantaneous time-frequency characteristics (such as the trend of instantaneous frequency changing with time) of non-stationary sub-modal signals, providing time-frequency data support for subsequent multi-dimensional judgment. The Hilbert spectrum is a three-dimensional spectrum with time as the horizontal axis, instantaneous frequency as the vertical axis and instantaneous amplitude as the amplitude, which visualizes the time-frequency characteristics of the sub-modal signal and avoids the one-sidedness of single time domain or frequency domain analysis.

[0061] The length of the preset time window can be dynamically determined according to the dominant frequency of the i+1-level sub-modal signal, and the formula is T win =k / f sub wherein, T win is the length of the preset time window, f sub is the dominant frequency of the corresponding sub-modal signal, and k is an empirical coefficient, usually 2-5, to ensure that the time window contains 2-5 complete effective signal periods. For example, for a sub-modal signal with a dominant frequency of 20 Hz, k=2, and the preset time window length can be 0.1 seconds, containing 2 20 Hz periods. When the sampling rate is 1000 Hz, the time window length is 100 sampling points.

[0062] First, the instantaneous amplitude at each time is extracted from the Hilbert spectrum, and then the instantaneous energy at each time is calculated according to the relationship that the instantaneous energy is equal to the square of the instantaneous amplitude. The average instantaneous energy of the sub-modal signal in each preset time window is the average of the instantaneous energy at all times in the time window.

[0063] The kurtosis coefficient reflects the steepness of the instantaneous energy distribution, the instantaneous frequency standard deviation reflects the stability of the instantaneous frequency, and the energy entropy value reflects the uniformity of the instantaneous energy distribution in the time window. By calculating the characteristics of these three dimensions, the characteristics of the sub-modal signal can be fully described from the energy distribution steepness, frequency stability, and energy distribution uniformity, avoiding the limitations of single instantaneous energy determination.

[0064] Those skilled in the art know that the calculation methods of the kurtosis coefficient, the instantaneous frequency standard deviation, and the energy entropy value in the prior art fall within the protection scope of the present application, and will not be described here.

[0065] Since only the kurtosis coefficient meeting the standard may be energy-smooth noise (such as low-frequency background noise), only the frequency meeting the standard may be frequency-stable interference (such as power frequency interference), and only the entropy value meeting the standard may be energy-concentrated burst noise. Non-noise modalities need to have the characteristics of energy distribution smoothness (kurtosis coefficient ≥ preset kurtosis coefficient threshold), frequency stability (instantaneous frequency standard deviation < preset standard deviation threshold), and energy concentration (energy entropy value < preset entropy threshold), and none of the three can be missing, so as to ensure that the sub-modal signal has the complete characteristics of AM-FM effective signal, and finally output the non-noise modalities. If any condition is not met, it is determined as a noise modality, which needs to be removed in the subsequent process.

[0066] The above, by calculating the kurtosis coefficient, the instantaneous frequency standard deviation, and the energy entropy value, the determination dimension of the sub-modal signal is expanded from single instantaneous energy to three-dimensional characteristics, and through the determination logic that the three conditions are met at the same time, the determination accuracy of the non-noise modalities is improved.

[0067] In a specific embodiment, S34 includes the following steps: S341, collect a known seismic effective signal sample set and a noise signal sample set of the same region and the same type as the initial seismic signal.

[0068] S342, calculate a first kurtosis coefficient of each effective signal sample and a second kurtosis coefficient of each noise signal sample respectively, and obtain a minimum value of the first kurtosis coefficient and a maximum value of the second kurtosis coefficient.

[0069] S343, statistically analyze the kurtosis coefficients of all the i+1 level sub-modal signals to obtain an i+1 level kurtosis coefficient distribution.

[0070] S344, according to the i+1 level kurtosis coefficient distribution, the minimum value of the first kurtosis coefficient and the maximum value of the second kurtosis coefficient, obtain a preset kurtosis coefficient threshold corresponding to the current i+1 level sub-modal signal.

[0071] The known seismic effective signal sample set contains at least 50 groups of pure effective signal segments verified by geology, and the noise signal sample set contains at least 50 groups of pure noise signal segments labeled, each group of samples has the same time length as the i+1 level sub-modal signal and the same sampling rate as the current signal.

[0072] The effective component and the noise component of the seismic signal have significant regional dependence. The kurtosis coefficient range of the effective signal and the kurtosis coefficient range of the noise signal of the seismic signal of the same region and the same type of exploration scene are relatively stable, while the signal characteristics of different regions are quite different. By collecting known samples of the same region and the same type as the initial seismic signal, the feature distribution of the samples can be ensured to be highly consistent with the feature distribution of the current processed signal, providing a regionally adaptive basic anchor point for subsequent threshold calculation, and avoiding threshold invalidation caused by the disconnection between the samples and the current signal.

[0073] The preset standard deviation threshold and the preset entropy value threshold can also refer to the method of obtaining the preset kurtosis coefficient threshold, and are obtained by statistical analysis of the standard deviation features and the entropy value features in the known seismic effective signal sample set and the noise signal sample set, which will not be described here.

[0074] The above-mentioned progressive operation of sample set collection, feature statistics, current signal analysis and threshold determination refines the generation process of the preset kurtosis coefficient threshold, solves the poor adaptability of the traditional fixed threshold, ensures that the threshold is based on historical experience and fits the current sub-modal signal characteristics, thereby improving the accuracy of non-noise modal determination.

[0075] In a specific embodiment, S344 includes the following steps: S3441, determine the distribution range according to the minimum value of the first kurtosis coefficient.

[0076] S3442, obtaining an overlap ratio of the peak state coefficient distribution of the (i+1)th submodality signal and the distribution range.

[0077] S3443, if the overlap ratio is greater than a preset ratio threshold, determining a product of an average of a minimum value of the first peak state coefficient and a maximum value of the second peak state coefficient and the first preset coefficient as the preset peak state coefficient threshold.

[0078] S3444, if the overlap ratio is less than or equal to the preset ratio threshold, determining a product of the average of the minimum value of the first peak state coefficient and the maximum value of the second peak state coefficient and the second preset coefficient as the preset peak state coefficient threshold.

[0079] In an embodiment, the second preset coefficient is greater than the first preset coefficient.

[0080] wherein the peak state coefficient distribution range of the effective signal needs to satisfy covering more than 90% of the known effective signal samples and having no overlap with the noise signal range, and the average of the minimum value C min of the first peak state coefficient and the maximum value C min of the second peak state coefficient is set as [C min +2×σ], wherein σ is a standard deviation of the peak state coefficient of the effective signal sample set, and 2×σ is an empirical value based on a 95% confidence interval of a normal distribution, which ensures that the range can cover most of the effective signals.

[0081] The overlap ratio of the peak state coefficient distribution of the (i+1)th submodality signal and the effective signal distribution range is essentially the proportion of the number of submodalities in the current submodality whose peak state coefficient meets the characteristics of the effective signal. A high overlap ratio indicates that the coefficients of most submodalities in the current submodality are close to the characteristics of the effective signal, and thus the threshold is reduced by the average value x the first preset coefficient (the first preset coefficient <1, such as 0.8) to ensure that weak effective signals can be correctly identified as non-noise modalities, while strong noise with coefficients much higher than the threshold can still be removed. A low overlap ratio indicates that the coefficients of most submodalities in the current submodality deviate from the characteristics of the effective signal, and thus the threshold is increased by the average value x the second preset coefficient (the second preset coefficient >1, such as 1.2) to strengthen the noise removal capability, so that only submodalities with coefficients close to the characteristics of the effective signal can be retained, reducing the accumulation of noise in subsequent iterations.

[0082] The specific value of the preset ratio threshold can be set by the implementer according to the actual situation, for example, 60%.

[0083] The above calculates the overlap range proportion, converts the correlation degree of the current sub-modal and the effective signal to a quantitative value, provides a clear quantitative basis for subsequent threshold adjustment, avoids adjustment deviation caused by subjective judgment, realizes accurate adaptation of the two scenes of effective signal as the main and noise as the main through the differentiated setting of the second preset coefficient and the first preset coefficient, solves the problem that the traditional fixed threshold or one-way adjusted threshold cannot adapt to the two scenes, and improves the judgment accuracy of the non-noise modal.

[0084] S4, each i+1 level sub-modal signal of the signal type of non-noise modal is determined as the i+1 level target signal, and the cross-correlation degree between each i+1 level target signal and the i level target signal for generating the i+1 level target signal is obtained.

[0085] Among them, the determined non-noise modal is explicitly the i+1 level target signal of the next level decomposition, ensuring that the iteration process can continue, that is, only the non-noise modal with high purity has the value of further decomposition, and the weak effective signal can be separated from it, and the noise modal is eliminated to avoid the accumulation of noise in the iteration.

[0086] The i+1 level target signal is generated by the i level target signal decomposition, and both of them should retain the core characteristics of the effective signal, such as the main frequency and the phase, but excessive decomposition may cause the effective signal characteristics to be broken, so that the i+1 level target signal and the i level target signal are too different, losing the effective signal properties.

[0087] The cross-correlation degree is a quantitative index of the similarity between two signals, and the value range is [-1, 1]. By calculating the cross-correlation degree, it can be quantitatively evaluated whether the i+1 level target signal inherits the effective characteristics of the i level target signal: the cross-correlation degree is high, which means that the effective characteristics of the two are consistent, and the decomposition is reasonable; the cross-correlation degree is low, which means that the decomposition may introduce false signals or lose effective characteristics, and the subsequent iteration termination logic needs to be judged whether to stop the decomposition.

[0088] S5, if all the cross-correlation degrees corresponding to the i+1 level target signals are greater than the preset degree threshold, all the sub-modal signals of the signal type of non-noise modal are obtained, and step S6 is executed, otherwise, i=i+1 is updated, and step S2 is repeatedly executed.

[0089] Among them, the current iteration level sequence number i is updated to i+1, so that the processing object of the next round of iteration is changed from the i level target signal to the i+1 level target signal, ensuring that the iteration process progresses in the level of 1→2→3→…→n.

[0090] The termination criterion is that the cross-correlation degree of all the i+1th target signals is greater than a preset degree threshold. If the cross-correlation degree of any i+1th target signal is less than or equal to the preset degree threshold, it is determined that the signal still contains noise that can be stripped, and iteration needs to be continued. If the cross-correlation degree of all the i+1th target signals is greater than the preset degree threshold, it is determined that the effective signal characteristics of all non-noise modalities have been stabilized, and there is no more noise that can be stripped. Therefore, it is meaningless to continue iteration, and iteration is terminated. All non-noise modality submodality signals are collected to provide high-purity materials for subsequent reconstruction.

[0091] The specific value of the preset degree threshold can be set by the implementer according to the actual situation, for example, based on a large amount of experimental data, and the value range is 0.7-0.9, preferably 0.8, to meet the requirements of ensuring that the effective signal purity meets the standard and avoiding excessive iteration.

[0092] All submodality signals of the non-noise modality are the collection of submodality signals determined as non-noise modalities from the first iteration to the termination iteration.

[0093] Through the iteration update of i=i+1, hierarchical progressive purification is realized, so that the effective signal purity of the non-noise modality is gradually improved with the iteration round. Through the termination condition that the cross-correlation degree of all the i+1th target signals is greater than the preset degree threshold, excessive iteration or insufficient iteration caused by fixed iteration times is avoided, and the decomposition rate and decomposition accuracy of the effective signal are improved.

[0094] S6, superimposing all the submodality signals of the non-noise modality to obtain the filtered seismic effective signal.

[0095] All submodality signals of the non-noise modality are high-purity effective signals, and each corresponds to different frequencies and different energy effective components in the original seismic signal. These submodality signals are essentially characteristic fragments of the original effective signal, and their phase and frequency characteristics are consistent with those of the original effective signal. Through linear superposition, they can be re-aggregated into complete effective signal waveforms, and finally output high-fidelity and low-noise filtered seismic effective signals to provide core data support for subsequent geological interpretation tasks such as reservoir prediction and fault identification.

[0096] In a specific embodiment, S6 includes the following steps: S61, for each non-noise modal sub-modal signal, according to the peak state coefficient, the instantaneous frequency standard deviation and the energy entropy value corresponding to the current sub-modal signal, an effective signal confidence corresponding to the current sub-modal signal is obtained, wherein the value range of the effective signal confidence is [0, 1], the absolute value of the difference between the peak state coefficient and the mean of the first peak state coefficient, the instantaneous frequency standard deviation and the energy entropy value are all negatively correlated with the effective signal confidence.

[0097] S62, according to the effective signal confidences corresponding to the sub-modal signals of all non-noise modalities, the superposition weight of each non-noise modal sub-modal signal is determined, wherein the superposition weight is positively correlated with the effective signal confidence, and the sum of the superposition weights corresponding to the sub-modal signals of all non-noise modalities is 1.

[0098] S63, according to the superposition weights of the sub-modal signals of all non-noise modalities, the sub-modal signals of all non-noise modalities are superimposed to obtain a filtered seismic effective signal.

[0099] Wherein, although the non-noise modalities all satisfy the conditions that the peak state coefficient is greater than or equal to the preset peak state coefficient threshold, the instantaneous frequency standard deviation is less than the preset standard deviation threshold and the energy entropy value is less than the preset entropy value threshold, the effective signal purity of different sub-modalities still differs, such as some sub-modalities are close to the threshold boundary and contain a small amount of noise, and some sub-modalities are far from the threshold and have extremely high purity.

[0100] Through the peak state coefficient, the instantaneous frequency standard deviation and the energy entropy value, a negative correlation mapping of the feature deviation and the effective signal confidence can be established. Correspondingly, the smaller the absolute value of the difference between the peak state coefficient and the mean of the peak state coefficient of the effective signal sample set, the closer the sub-modal peak state feature is to the effective signal, and the higher the effective purity is; the smaller the instantaneous frequency standard deviation, the more stable the sub-modal frequency is, and the higher the effective purity is; the smaller the energy entropy value, the more concentrated the sub-modal energy distribution is, and the higher the effective purity is.

[0101] Then, according to the absolute value of the difference between the peak state coefficient and the mean of the first peak state coefficient, the instantaneous frequency standard deviation and the energy entropy value, the peak state coefficient deviation, the instantaneous frequency standard deviation deviation and the energy entropy value deviation are respectively normalized, and then the three deviations are converted into the effective signal confidence by using weighted average inverse operation. The weights corresponding to the absolute value of the difference between the peak state coefficient and the mean of the first peak state coefficient, the instantaneous frequency standard deviation and the energy entropy value can be set to 0.4, 0.3 and 0.3 respectively.

[0102] Further, when superimposing all non-noise modal sub-modal signals, since each sub-modal signal comes from different iteration levels of decomposition, there may be a slight time offset. Taking the time axis of the initial seismic signal as the benchmark, adjust the time starting point of all sub-modal signals, align each sub-modal signal with the initial seismic signal by the maximum cross-correlation method, such as calculating the cross-correlation coefficient of the sub-modal signal and the initial signal, finding the time offset corresponding to the maximum value of the coefficient, shifting the sub-modal signal to the same time starting point, and ensuring that the effective signal components at the same time are accurately corresponding when superimposed.

[0103] Due to the large difference in amplitude range of different sub-modal signals, such as shallow sub-modal amplitude 0.5V-2V and deep sub-modal amplitude 0.1V-0.5V, direct superposition will cause the deep weak effective signal to be covered by the shallow strong signal, therefore, the amplitude normalization processing can be used to uniformly map the amplitude range of all sub-modal signals to [0, 2V], ensuring that sub-modal signals of different amplitudes can contribute effective features when superimposed.

[0104] Through the progressive operation of effective signal confidence quantification, dynamic weight distribution, and weighted superposition reconstruction, the confidence-driven precise weighted superposition solves the pain point of equal contribution of low-purity sub-modal and high-purity sub-modal in traditional average superposition, and improves the fidelity and signal-to-noise ratio of the filtered effective signal.

[0105] Through the above-mentioned modal decomposition of the i-th level target signal, the complex mixed seismic signal is decomposed into single-feature sub-modal signals, breaking through the problem of high difficulty in traditional mixed signal separation, while retaining the core amplitude modulation and frequency modulation features of the effective signal, avoiding the effective signal being covered by noise due to not being decomposed; By establishing an explicit sub-modal screening standard through the average instantaneous energy of the sub-modal signal in the preset time window, local energy analysis is realized by means of the preset time window, avoiding the misjudgment of weak effective signal caused by overall energy averaging, and accurately distinguishing between non-noise modal and noise modal; By calculating the correlation degree to quantify the similarity of the upper and lower level target signals, data basis is provided for subsequent judgment of whether the decomposition is reasonable and whether the effective features are retained, preventing the effective signal features from being broken due to excessive decomposition; Through iterative updating and iterative stop condition based on correlation degree, the dynamic control of the iteration process is realized, avoiding excessive iteration or insufficient iteration caused by fixed iteration times, ensuring that the final obtained non-noise modal sub-signal has stable and effective features, without residual signals that are not fully purified, and ensuring the quality of subsequent reconstruction; By superimposing all non-noise modal sub-modal signals, the filtered seismic effective signal is obtained, the effective signal components of different iteration levels are aggregated, the closed loop from sub-modal decomposition to complete effective signal restoration is realized, and the final output seismic effective signal not only retains the effective features of each level, but also greatly reduces the noise interference, improving the purity of the seismic effective signal.

[0106] Embodiment two Embodiment two of the present application provides a non-transitory computer readable storage medium, which can be arranged in an electronic device to save at least one instruction or at least one program related to a method in the method embodiment, and the at least one instruction or the at least one program is loaded and executed by the processor to realize the seismic signal filtering method provided in the above embodiment.

[0107] Embodiment three Embodiment three of the present application provides an electronic device, which includes a processor and the non-transitory computer readable storage medium in embodiment two of the present application.

[0108] The above is only the preferred embodiment of the present application, and does not limit the present application in any form. Although the present application has been disclosed as above with the preferred embodiment, it is not intended to limit the present application. Any person skilled in the art can make some changes or modifications to the above disclosed technical content without departing from the technical solution of the present application, and any simple modification, equivalent change and modification made on the basis of the technical essence of the present application to the above embodiment are still within the scope of the technical solution of the present application.

Claims

1. A method of filtering seismic signals, characterized by, The seismic signal filtering method comprises the following steps: S1, initializing i=1, and determining an initial seismic signal as an ith-level target signal; S2, for any ith-level target signal, performing modal decomposition processing on the current ith-level target signal to obtain a plurality of (i+1)th-level sub-modal signals corresponding to the current ith-level target signal, wherein the sub-modal signal is a signal component with amplitude modulation and frequency modulation characteristics; S3, according to the average instantaneous energy of each (i+1)th-level sub-modal signal in each preset time window, obtaining a signal type corresponding to each (i+1)th-level sub-modal signal, wherein the signal type is a noise mode or a non-noise mode; S4, determining each (i+1)th-level sub-modal signal with the non-noise mode as an (i+1)th-level target signal, and obtaining a cross-correlation degree between each (i+1)th-level target signal and the ith-level target signal for generating the (i+1)th-level target signal; S5, if the cross-correlation degrees corresponding to all (i+1)th-level target signals are greater than a preset degree threshold, obtaining all sub-modal signals with the non-noise mode, and performing step S6, otherwise, updating i=i+1, and repeating step S2; S6, superimposing all sub-modal signals with the non-noise mode to obtain a filtered seismic effective signal.

2. The seismic signal filtering method of claim 1, wherein, The modal decomposition processing is a jump-enhanced amplitude modulation and frequency modulation modal decomposition processing, and S2 comprises the following steps: S21, extracting key prior features of the current ith-level target signal, wherein the key prior features comprise a main frequency range and a signal-to-noise ratio; S22, determining an iteration step and an initial decomposition layer number N in the jump-enhanced amplitude modulation and frequency modulation modal decomposition processing according to the key prior features; S23, performing the jump-enhanced amplitude modulation and frequency modulation modal decomposition processing according to the iteration step and the initial decomposition layer number N to obtain a plurality of (i+1)th-level sub-modal signals and a jump interference component, wherein the jump interference component is determined according to that a peak amplitude to average amplitude ratio of the current ith-level target signal is greater than a preset ratio threshold, and a peak duration is less than a preset time threshold.

3. The seismic signal filtering method of claim 2, wherein, S23 comprises the following steps: S231, initializing core decomposition parameters according to the current ith-level target signal, wherein the core decomposition parameters comprise a decomposition window length, a modal convergence threshold, and a jump interference detection window; S232, performing N-layer cascaded decomposition on the current ith-level target signal according to the initial decomposition layer number N and the core decomposition parameters to obtain one preliminary sub-modal signal output by each layer of decomposition and a residual component output by the Nth layer of decomposition; S233, comparing a peak amplitude and a peak duration of each preliminary sub-modal signal and the residual component with the preset ratio threshold and the preset time threshold to extract the jump interference component from all preliminary sub-modal signals and the residual component; S234, removing signal segments time-overlapped with the jump interference component from the preliminary sub-modal signals and the residual component to obtain a plurality of (i+1)th-level sub-modal signals corresponding to the current ith-level target signal.

4. The seismic signal filtering method of claim 1, wherein, S3 comprises the following steps: S31, for any i+1th sub-modal signal, performing Hilbert transform on the current i+1th sub-modal signal to obtain a Hilbert spectrum corresponding to the current i+1th sub-modal signal; S32, according to the Hilbert spectrum corresponding to the current i+1th sub-modal signal, obtaining an average instantaneous energy of the current i+1th sub-modal signal in each preset time window, wherein the length of the preset time window is determined according to the dominant frequency of the current i+1th sub-modal signal; S33, according to all the average instantaneous energies corresponding to the current i+1th sub-modal signal, obtaining a kurtosis coefficient, an instantaneous frequency standard deviation and an energy entropy value corresponding to the current i+1th sub-modal signal; S34, obtaining a preset kurtosis coefficient threshold, a preset standard deviation threshold and a preset entropy threshold corresponding to the current i+1th sub-modal signal; S35, if the kurtosis coefficient of the current i+1th sub-modal signal is greater than or equal to the preset kurtosis coefficient threshold, the instantaneous frequency standard deviation is less than the preset standard deviation threshold, and the energy entropy value is less than the preset entropy threshold, determining that the current i+1th sub-modal signal is a non-noise mode.

5. The seismic signal filtering method of claim 4, wherein, S34 includes the following steps: S341, collecting a known seismic effective signal sample set and a noise signal sample set of the same region and type as the initial seismic signal; S342, calculating a first kurtosis coefficient of each effective signal sample and a second kurtosis coefficient of each noise signal sample respectively, and obtaining a minimum value of the first kurtosis coefficient and a maximum value of the second kurtosis coefficient; S343, statistically analyzing the kurtosis coefficients of all i+1th sub-modal signals to obtain an i+1th kurtosis coefficient distribution; S344, according to the i+1th kurtosis coefficient distribution, the minimum value of the first kurtosis coefficient and the maximum value of the second kurtosis coefficient, obtaining a preset kurtosis coefficient threshold corresponding to the current i+1th sub-modal signal.

6. The seismic signal filtering method of claim 5, wherein, S344 includes the following steps: S3441, determining a distribution range according to the minimum value of the first kurtosis coefficient; S3442, obtaining an overlap range proportion of the kurtosis coefficient distribution of the i+1th sub-modal signal and the distribution range; S3443, if the overlap range proportion is greater than a preset proportion threshold, determining a product of an average value of the minimum value of the first kurtosis coefficient and the maximum value of the second kurtosis coefficient and a first preset coefficient as the preset kurtosis coefficient threshold; S3444, if the overlap range proportion is less than or equal to the preset proportion threshold, determining a product of the average value of the minimum value of the first kurtosis coefficient and the maximum value of the second kurtosis coefficient and a second preset coefficient as the preset kurtosis coefficient threshold.

7. The seismic signal filtering method of claim 6, wherein, The second preset coefficient is greater than the first preset coefficient.

8. The method of filtering seismic signals according to claim 5, wherein, S6 includes the following steps: S61, for each non-noise mode sub-modal signal, according to the kurtosis coefficient, the instantaneous frequency standard deviation and the energy entropy value corresponding to the current sub-modal signal, obtaining an effective signal confidence corresponding to the current sub-modal signal, wherein the effective signal confidence has a value range of [0, 1], the absolute value of the difference between the kurtosis coefficient and the average value of the first kurtosis coefficient, the instantaneous frequency standard deviation and the energy entropy value are all negatively correlated with the effective signal confidence; S62, determining a superposition weight of each submodal signal of the non-noise modal according to the effective signal confidence corresponding to the submodal signal of the non-noise modal, wherein the superposition weight is in a positive correlation with the effective signal confidence, and a sum of the superposition weights corresponding to all the submodal signals of the non-noise modal is 1; S63, superimposing all the submodal signals of the non-noise modal according to the superposition weights of the submodal signals of the non-noise modal to obtain a filtered seismic effective signal. 9.A non-transitory computer-readable storage medium having stored therein at least one instruction or at least one piece of program, characterized in that, The at least one instruction or the at least one program is loaded and executed by the processor to implement the seismic signal filtering method as claimed in any one of claims 1-8.

10. An electronic device, comprising: The non-transitory computer readable storage medium as claimed in claim 9 is included. The non-transitory computer readable storage medium as claimed in claim 9 is included.

Images