A vibroseis related pre-wavelet sidelobe suppression method and system

Abstract

The application provides a controllable source related pre-subwave sidelobe suppression method and system, and belongs to the field of seismic data processing.The method obtains an ideal pulse signal according to the starting frequency and the ending frequency of a source scanning signal, obtains a deconvolution filter by using the source scanning signal and the ideal pulse signal, and obtains a cross-correlation record after subwave sidelobe suppression by using the deconvolution filter.The application improves the influence of the related subwave sidelobe by designing a deconvolution operator and performing deconvolution processing on the related pre-recordings.The test result shows that the precision is high and the effect is good after the seismic data is processed by using the method of the application, and the application can be used in the processing of the controllable source related pre-seismic record and can suppress the related subwave sidelobe.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of seismic data processing, and particularly relates to a controllable seismic source related pre-wavelet sidelobe suppression method and system. BACKGROUND

[0002] The controllable seismic source emits scanning signals to the underground in the form of continuous seismic waves, and then the frequency modulation signals are compressed into pulse signals with delay information through correlation technology, so that the detection of underground targets and the imaging of underground information are realized. The controllable seismic source correlation technology is strictly a cross-correlation processing, and the source wavelet after correlation is regarded as a Klauder wavelet. The controllable seismic source cross-correlation processing will bring the wavelet sidelobe problem, which has a certain influence on the resolution of seismic records. In view of the interference problem of the controllable seismic source correlation wavelet sidelobe, scholars at home and abroad have carried out research from different aspects.

[0003] Eugene C. Trantham and Ni proposed using the source characteristic signal to carry out deconvolution processing on the correlation record, so as to compress the wavelet. Zhang Hongle et al. proposed that the use of logarithmic segmented scanning signals can make the correlation wavelet have a minimum correlation side leaf. Shen Yuanyuan et al. carried out simulation analysis on the autocorrelation function and amplitude spectrum, and proposed a linear scanning signal parameter optimization design method, which improved the correlation characteristics of the wavelet to a certain extent. Cao Wuxiang proposed calculating the scanning signal under the action of stratum absorption attenuation by using the quality factor, and proposed Q correlation technology, that is, using the scanning signal with the attenuation factor added to carry out cross-correlation processing with the original record. Compared with the traditional correlation algorithm, the Q correlation algorithm has obvious advantages. In the correlation process, the attenuation of the earth is compensated, and the data resolution is greatly improved, especially for high-frequency weak signals. Ni Yudong proposed using two source arrays to improve the emission energy, and using the ground force signal to carry out deconvolution on the source record to obtain data with certain fidelity, avoid correlation noise, and improve the signal-to-noise ratio. Christine E. Krohn et al. proposed a vibration signal separation and vibration wavelet characteristic deconvolution method, the output of which is strictly minimum phase and matches the explosive data. The vibration signal is calculated from the source ground force signal, a plurality of sources and a plurality of vibration characteristic signals are used to design a separation filter, each source signal is optimally separated, and the designed impulse response is used to replace the cross-correlation scanning.

[0004] The Chinese patent publication CN107367758A discloses a method for improving the signal-to-noise ratio of a controllable seismic source weighted correlation, which comprises the following steps: obtaining the main frequency and the slope of the main frequency part of a controllable seismic source single shot; modifying the frequency spectrum of a reference signal to improve the energy of the main frequency part and obtain a weighted frequency spectrum curve; converting the weighted frequency spectrum curve into a weighted sweep signal; and correlating the weighted sweep signal with an original record to obtain a weighted correlation single shot record of the controllable seismic source. The method improves the energy of the main frequency part of the reference signal before correlation, converts it into a weighted sweep signal, and then correlates the weighted sweep signal with the original single shot record to obtain a single shot record. The signal-to-noise ratio of the single shot record obtained by the method is higher than that obtained by the conventional method. However, the method disclosed in the patent requires high quality of the original single shot record. When the original single shot record has strong noise and low signal-to-noise ratio, the slope of the main frequency part may have a large error, which reduces the accuracy of the weighted sweep signal and the reliability of the correlated record.

[0005] The Chinese patent publication CN105572723A discloses a design method for a controllable seismic source sweep signal based on autocorrelation wavelets, which comprises the following steps: selecting a wavelet with good form and wide frequency band according to the frequency requirement of the exploration area, and designing the start and end frequencies, sweep length, and start and end ramp length of the sweep signal; calculating the wavelet spectrum; distributing the sampling frequency according to the relationship that the scanning time of each frequency is proportional to the amplitude required by the frequency component, and calculating the time function t(f) corresponding to each sampling frequency; performing inverse transformation on the time function t(f) to calculate the time-frequency function f(t); and calculating the instantaneous phase by integrating the time-frequency function f(t), and then calculating the sine controllable seismic source sweep signal. The controllable seismic source sweep signal generated by the invention overcomes the problems of poor wavelet form of the conventional sweep signal and low energy in the high frequency band of the "rotating phase and logarithmic segmented" sweep signal, and can greatly improve the quality of controllable seismic data. However, the method disclosed in the patent is a design method for a sweep signal according to the frequency requirement of the exploration area. Unlike the conventional sweep signal, the sweep signal designed by the method is usually a nonlinear sweep signal, and the actual operability of the field seismic source instrument needs to be considered when using the sweep signal.

[0006] Chinese patent publication CN102692643A discloses a time-varying vibrator force signal deconvolution method for improving seismic data resolution, records the vibration data and force signal before correlation, constructs a time-varying force signal in combination with a quality factor, obtains a reflection coefficient series by deconvolution of the vibrator vibration record, multiplies the time-varying force signal with the vibrator record, obtains a formation quality factor correlation value, processes the time-varying force signal and the vibrator data, obtains a reflection wavelet formula, and obtains the reflection coefficient series by deconvolution. The application compensates the change of the Q value, improves the signal-to-noise ratio of the data, removes the influence of the force signal, and removes the correlation noise caused by the distortion of the force signal. However, the method disclosed in the patent uses the quality factor Q value to construct a time-varying force signal for cross-correlation processing, and the Q value needs to be obtained in advance. Whether the Q value is accurate or not is important for the construction of the time-varying force signal.

[0007] In conventional controlled source exploration, a scanning signal is usually cross-correlated with seismic data to obtain a compressed "approximate" pulse signal that can reflect the response of the formation. The existence of sidelobes of the correlation wavelet, especially the first sidelobe, directly affects the resolution of adjacent formations. SUMMARY

[0008] The present application aims to solve the problems existing in the prior art, and provides a controlled source correlation pre-wavelet sidelobe suppression method and system, which can significantly suppress the sidelobes of the wavelet and provide data resolution.

[0009] The present application is achieved by the following technical solutions:

[0010] In a first aspect of the present application, a controlled source correlation pre-wavelet sidelobe suppression method is provided. The method obtains an ideal pulse signal according to the start frequency and end frequency of the source scanning signal, obtains a deconvolution filter using the source scanning signal and the ideal pulse signal, and obtains a cross-correlation record after wavelet sidelobe suppression using the deconvolution filter.

[0011] Further improvements of the present application are as follows:

[0012] The method comprises the following steps:

[0013] Step one: obtaining an ideal pulse signal using the source scanning signal;

[0014] Step two: obtaining a deconvolution filter using the ideal pulse signal and the source scanning signal;

[0015] Step three: applying the deconvolution filter to the pre-correlation seismic record to obtain a pre-correlation record after deconvolution processing;

[0016] Step four: cross-correlating the pre-correlation record after deconvolution processing with the source scanning signal to obtain a correlation record after wavelet sidelobe suppression.

[0017] The further improvement of the present application is:

[0018] The source scanning signal in the step one is obtained by:

[0019] Before the field acquisition construction of the controllable source, the acquisition test of different source parameters is carried out to obtain the initial frequency fs and the terminal frequency fe of the source scanning signal.

[0020] The further improvement of the present application is:

[0021] The operation of the step one comprises:

[0022] The ideal pulse signal Fd is obtained by using the following formula:

[0023]

[0024] Wherein, f represents the signal frequency;

[0025] f3 is the left boundary of the attenuation cutoff: f3=fs+0.4×(fe-fs);

[0026] f4 is the right boundary of the attenuation cutoff: f4=fe-0.4×(fe-fs).

[0027] The further improvement of the present application is:

[0028] The operation of the step two comprises:

[0029] The Fourier transform is carried out on the source scanning signal to obtain the scanning signal spectrum Fw;

[0030] The signal power spectrum PowerW and the inverse phase operator phase are calculated and obtained;

[0031] The deconvolution filter is obtained by using the signal power spectrum PowerW and the inverse phase operator phase.

[0032] The further improvement of the present application is:

[0033] The operation of calculating and obtaining the signal power spectrum PowerW and the inverse phase operator phase comprises:

[0034] PowerW=FwxFw

[0035] Phase=exp(-i×theta)

[0036] Wherein, i is the imaginary unit;

[0037] Theta=atan(imag(Fw) / real(Fw))

[0038] imag(Fw) represents the imaginary part of the spectrum Fw;

[0039] real(Fw) represents the real part of the spectrum Fw;

[0040] atan is an inverse tangent operator.

[0041] A further improvement of the present application is that:

[0042] The operation of obtaining the deconvolution filter using the signal power spectrum PowerW and the inverse phase operator phase includes:

[0043] Ffilter = Fd / (sqrt(PowerW+noise) / phase)

[0044] Where noise is random noise.

[0045] A further improvement of the present application is that:

[0046] The operation of step three includes:

[0047] Obtain the seismic record received by the source scanning signal excitation and the detector, which is the correlation pre-seismic record data;

[0048] Apply the deconvolution filter to the correlation pre-seismic record data to obtain the correlation pre-record data_new after deconvolution processing:

[0049] data_new = data*Ffilter.

[0050] In the formula: * represents convolution.

[0051] The second aspect of the present application provides a controllable source correlation pre-wavelet sidelobe suppression system, the system comprises:

[0052] Pulse signal generation unit: obtain ideal pulse signal using source scanning signal;

[0053] Filter generation unit: connected with the pulse signal generation unit, for obtaining deconvolution filter using ideal pulse signal and source scanning signal;

[0054] Deconvolution unit: connected with the filter generation unit, for applying deconvolution filter to the correlation pre-seismic record to obtain the correlation pre-record after deconvolution processing;

[0055] Cross-correlation unit: connected with the deconvolution unit, for cross-correlation between the correlation pre-record after deconvolution processing and the source scanning signal to obtain the correlation record after wavelet sidelobe suppression.

[0056] In a third aspect, the application provides a computer readable storage medium storing at least one program executable by a computer, which when executed by the computer, causes the computer to perform the steps of the method for controlling the sidelobe of a pre-correlation wavelet of a vibroseis.

[0057] Compared with the prior art, the application has the following beneficial effects:

[0058] The application improves the influence of the sidelobe of the correlation wavelet by designing a deconvolution operator to perform deconvolution processing on the pre-correlation record. BRIEF DESCRIPTION OF DRAWINGS

[0059] Figure 1 The pre-correlation record of the vibroseis forward simulation;

[0060] Figure 2 The post-correlation record directly acted on by the scanning signal in the conventional method;

[0061] Figure 3 The cross-correlation record after the sidelobe suppression by the method of the application;

[0062] Figure 4 Comparison of the cross-correlation records before and after the sidelobe suppression;

[0063] Figure 5 The step block diagram of the method of the application. DETAILED DESCRIPTION

[0064] The application will be described in further detail below with reference to the accompanying drawings:

[0065] The application provides a sidelobe suppression technique for a pre-correlation wavelet of a vibroseis, which designs a deconvolution operator by using the start and end frequencies of the scanning signal, performs deconvolution processing on the pre-correlation record, and then performs cross-correlation with the scanning signal to obtain a post-correlation wavelet, thereby effectively improving the sidelobe problem caused by the conventional cross-correlation processing.

[0066] The basic idea of the application is to design an ideal pulse signal according to the start and end frequencies of the scanning signal of the source, to obtain a deconvolution filter by using the scanning signal of the source and the ideal pulse signal, to apply the deconvolution filter to the pre-correlation record to obtain a deconvolution pre-correlation record, and to perform cross-correlation processing on the scanning signal and the deconvolution pre-correlation record to obtain a cross-correlation record after the sidelobe suppression, so that the signal has higher fidelity.

[0067] The method of the present application is implemented as follows:

[0068] Example 1

[0069] As shown in Figure 5 , the method of the present application comprises:

[0070] Step 1: obtaining an ideal pulse signal using a seismic source scanning signal:

[0071] Before field acquisition construction of a controllable seismic source, different seismic source parameter acquisition tests need to be conducted to finally determine the starting frequency and the ending frequency of the seismic source scanning signal in production.

[0072] Suppose the starting frequency of the seismic source scanning signal is fs, and the ending frequency is fe. Define the left boundary of the frequency domain ideal pulse signal as fs, the left boundary of the decay cutoff as f3 = fs + 0.4 × (fe - fs), the right boundary as fe, and the right boundary of the decay cutoff as f4 = fe - 0.4 × (fe - fs).

[0073] The designed frequency domain ideal pulse signal Fd is:

[0074]

[0075] In the above formula, f represents the signal frequency. The "*" in the above formula represents multiplication.

[0076] Step 2: obtaining a deconvolution filter using the pulse signal and the seismic source scanning signal:

[0077] Perform Fourier transform on the seismic source scanning signal to obtain the scanning signal spectrum Fw.

[0078] According to the spectrum calculation, obtain the signal power spectrum PowerW and the inverse phase operator phase:

[0079] PowerW = Fw × Fw

[0080] phase = exp(-i × theta)

[0081] Where i is the imaginary unit, theta = atan(imag(Fw) / real(Fw)), where imag(Fw) represents the imaginary part of the spectrum Fw, real(Fw) represents the real part of the spectrum Fw, and atan is the inverse tangent operator.

[0082] The designed deconvolution filter is:

[0083] Ffilter = Fd / (sqrt(PowerW + noise) / phase), noise is random noise.

[0084] Step three: the seismic record received by the detector after the source scanning signal excitation is the correlated pre-seismic record data. The deconvolution filter is applied to the correlated pre-seismic record data to obtain the deconvolution processed correlated pre-record data_new:

[0085] data_new=data*Ffilter.

[0086] In the formula: * represents convolution.

[0087] Step four: the deconvolution processed correlated pre-record data_new is cross-correlated with the source scanning signal (the cross-correlation is processed by using the existing method, which will not be described here.), to obtain the correlated record after the sidelobe suppression of the wavelet.

[0088] The deconvolution filter is equivalent to a filter modification on the source scanning signal. The cross-correlation processing of the filter modified signal and the source scanning signal can suppress the sidelobe of the correlated wavelet, to obtain the correlated record after the sidelobe suppression of the wavelet.

[0089] The embodiments of the method of the application are as follows:

[0090]

Embodiment two

[0091] In order to verify the correctness and feasibility of the correlated pre-wavelet sidelobe suppression method, the embodiment uses simulation numerical to test and verify. A horizontal layered model is designed to perform controllable source acoustic wave equation forward modeling. Figure 1 For the correlated pre-seismic data obtained by forward modeling. The conventional method is to directly use the source scanning signal and the correlated pre-record to perform cross-correlation processing to obtain the correlated post-record, as shown in Figure 2 .

[0092] The correlated pre-record is deconvolution processed by using the wavelet sidelobe suppression method provided by the application, and then the cross-correlation is performed by using the source scanning signal, to obtain the correlated post-record, as shown in Figure 3 .

[0093] Comparison Figure 2 and Figure 3 It can be seen that the cross-correlation record after the cross-correlation pre-wavelet sidelobe suppression processing, the direct wave and the reflection wave same phase axis converge faster and the energy is more concentrated, which is visually represented as the same phase axis is thinner. Compared with the same phase axis before the sidelobe suppression, the energy of the direct wave and the reflection wave same phase axis is more divergent, and the convergence is not crisp enough, which is visually represented as the same phase axis is fat, as indicated by the arrows in Figure 2 and Figure 3 .

[0094] Figure 4In order to extract one of the records for comparison, the dark gray curve is the cross-correlation record obtained by directly cross-correlating the source scanning signal, and the light gray curve is the post-correlation record obtained by suppressing the sidelobes of the pre-correlation record and then cross-correlating the post-correlation record with the source scanning signal. It can be seen that the sidelobe energy of the cross-correlation record after sidelobe suppression processing is obviously suppressed, and the wavelet resolution is higher.

[0095] The application only uses the start frequency, stop frequency and other information of the source scanning signal to obtain a deconvolution filter, applies the deconvolution filter to the pre-correlation seismic record, and then cross-correlates the post-correlation record with the scanning signal to obtain a deconvolution post-correlation record. Therefore, the method has low dependence on data quality and other stratum information, is simple to operate, easy to implement, and can effectively improve the wavelet sidelobe problem caused by conventional cross-correlation processing.

[0096] The application also provides a controllable source pre-correlation wavelet sidelobe suppression system.

[0097]

Embodiment three

[0098] The system comprises:

[0099] A pulse signal generation unit is used to obtain an ideal pulse signal from the source scanning signal.

[0100] A filter generation unit is connected with the pulse signal generation unit and is used to obtain a deconvolution filter from the ideal pulse signal and the source scanning signal.

[0101] A deconvolution unit is connected with the filter generation unit and is used to apply the deconvolution filter to the pre-correlation seismic record to obtain a post-deconvolution pre-correlation record.

[0102] A cross-correlation unit is connected with the deconvolution unit and is used to cross-correlate the post-deconvolution pre-correlation record with the source scanning signal to obtain a post-sidelobe-suppression correlation record.

[0103] Specifically, the pulse signal generation unit performs the following processing:

[0104] Suppose that the start frequency of the source scanning signal is fs and the stop frequency is fe. The left boundary of the frequency domain ideal pulse signal is defined as fs, the left boundary of the attenuation cutoff is f3=fs+0.4×(fe-fs), the right boundary is fe, and the right boundary of the attenuation cutoff is f4=fe-0.4×(fe-fs).

[0105] The designed frequency domain ideal pulse signal Fd is:

[0106]

[0107] f in the formula above represents signal frequency, and * represents multiplication.

[0108] The filter generating unit performs the following processing:

[0109] The Fourier transform is performed on the source scanning signal to obtain a scanning signal spectrum Fw.

[0110] The signal power spectrum PowerW and the inverse phase operator phase are obtained according to the spectrum calculation:

[0111] PowerW = Fw * Fw

[0112] phase = exp(-i * theta)

[0113] where i is an imaginary unit, theta = atan(imag(Fw) / real(Fw)), where imag(Fw) represents the imaginary part of the spectrum Fw, real(Fw) represents the real part of the spectrum Fw, and atan is an inverse tangent operator.

[0114] The deconvolution filter is designed as:

[0115] Ffilter = Fd / (sqrt(PowerW + noise) / phase), and noise is random noise.

[0116] The deconvolution unit performs the following processing:

[0117] The deconvolution filter is applied to the correlation pre-seismic record data to obtain a deconvolution-processed correlation pre-record data_new:

[0118] data_new = data * Ffilter.

[0119] In the formula, * represents convolution.

[0120] The sidelobe energy of the correlation post-record processed by the present application is obviously suppressed, and the data resolution is improved. The present application performs correlation pre-sidelobe suppression on analog data, has good effect, and the method is stable in algorithm, and can be further applied to actual data.

[0121] In the description of the present application, it should be noted that, unless otherwise explicitly specified and limited, the terms "connected", "connected" should be understood in a broad sense, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0122] In the description of the present application, unless otherwise specified, the terms "upper", "lower", "left", "right", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application.

[0123] Finally, it should be noted that the above technical solutions are only one embodiment of the present application, and for those skilled in the art, on the basis of the application disclosed application method and principle, various types of improvements or modifications can be easily made, and are not limited to the methods described in the above specific embodiments of the present application, therefore the above described method is only preferred, and does not have the meaning of limitation.

Claims

1. A method of vibroseis related pre-subwave lobe sidelobe suppression, characterized by: The method obtains an ideal pulse signal according to a start frequency and an end frequency of a source scanning signal, obtains a deconvolution filter using the source scanning signal and the ideal pulse signal, and obtains a correlation record with a wavelet sidelobe suppressed using the deconvolution filter; The method comprises: Step 1: obtaining an ideal pulse signal using a source scanning signal; Step 2: obtaining a deconvolution filter using the ideal pulse signal and the source scanning signal; Step 3: applying the deconvolution filter to a pre-correlation seismic record to obtain a pre-correlation record after deconvolution processing; Step 4: correlating the pre-correlation record after deconvolution processing with the source scanning signal to obtain a correlation record with a wavelet sidelobe suppressed.

2. The controllable source related pre-subwavelet sidelobe abatement method of claim 1, wherein: The source scanning signal in the step 1 is obtained as follows: Before field acquisition of a controllable source, a test is performed for different source parameters to obtain a start frequency fs and an end frequency fe of the source scanning signal.

3. The controllable source related pre-subwavelet sidelobe canceling method of claim 2, wherein: The step 1 comprises: An ideal pulse signal Fd is obtained using the following formula: Wherein, f represents a signal frequency; f3 is a left boundary of a decay cutoff: f3 = fs + 0.4 × (fe-fs); f4 is a right boundary of the decay cutoff: f4 = fe-0.4 × (fe-fs).

4. The controllable source related pre-subwavelet sidelobe canceling method of claim 3, wherein: The step 2 comprises: Performing Fourier transform on the source scanning signal to obtain a scanning signal spectrum Fw; Calculating a signal power spectrum PowerW and an inverse phase operator phase; Obtaining the deconvolution filter using the signal power spectrum PowerW and the inverse phase operator phase.

5. The controllable source related pre-subwavelet sidelobe canceling method of claim 4, wherein: The calculation of the signal power spectrum PowerW and the inverse phase operator phase comprises: PowerW = Fw × Fw phase = exp(-i × theta) Wherein, i is an imaginary unit; theta = atan(imag(Fw) / real(Fw)) Wherein, imag(Fw) represents an imaginary part of the spectrum Fw; real(Fw) represents a real part of the spectrum Fw; atan is an inverse tangent operator.

6. The controllable source related pre-subwavelet sidelobe canceling method of claim 5, wherein: The operation of obtaining the deconvolution filter using the signal power spectrum PowerW and the inverse phase operator phase comprises: Ffilter = Fd / (sqrt(PowerW+noise) / phase) Wherein, noise is random noise.

7. The controllable source related pre-subwavelet sidelobe canceling method of claim 6, wherein: The step 3 comprises: Obtaining a seismic record excited by the source scanning signal and received by a receiver, which is a pre-correlation seismic record data; Applying the deconvolution filter to the pre-correlation seismic record data to obtain a pre-correlation record data_new after deconvolution processing: data_new = data * Ffilter, Wherein, * represents convolution.

8. A vibroseis related pre-subwavelet sidelobe suppression system, characterized by: The system comprises: A pulse signal generation unit for obtaining an ideal pulse signal using a source scanning signal; A filter generation unit connected with the pulse signal generation unit, configured to obtain a deconvolution filter using the ideal pulse signal and the source scanning signal; A deconvolution unit connected with the filter generation unit, configured to apply the deconvolution filter to a pre-correlation seismic record to obtain a pre-correlation record after deconvolution processing. A cross-correlation unit is connected with the deconvolution unit, configured to cross-correlate the deconvolution-processed correlation pre-recordings with the source scanning signals to obtain the correlation recordings after wavelet sidelobe suppression.

9. A computer-readable storage medium, characterized in that: The computer readable storage medium stores at least one program executable by the computer, and the at least one program is executed by the computer to make the computer execute the steps in the controllable source correlation pre-wavelet sidelobe suppression method according to any one of claims 1-7.

Images