CN103869364B - A Multiple Suppression Method Based on Bitermial Parabolic Radon Transform - Google Patents

Abstract

The invention discloses a multiple wave suppression method based on dual parabolic Radon transformation. The method includes the steps that a pre-stack CMP trace gather is processed, an objective function is established, positive Radon transformation is performed on the pre-stack CMP trace gather, one-time reflected waves and multiple waves are separated in the Radon area, and data of the one-time reflected waves are estimated. The method solves the problems that an existing multiple wave attenuation technology based on convectional Radon transformation can not simulate AVO features of seismic data, the false phenomenon is generated in the positive transformation, the smearing exists, the distortion of the amplitude of the one-time reflected waves happens and leakage of the amplitude is caused in the reverse transformation, transformation of the shape of the seismic same-phase axis and the amplitude can be achieved in the Radon area, the movement features and the AVO features of seismic data can be separated better in the Radon area, the false phenomenon generated in the positive transformation process is weakened, accurate restoring of energy of the one-time reflected waves is achieved, the AVO features of the seismic data are protected, and the subsequent seismic inversion and explanation reliability is improved.

Description

A Multiple Suppression Method Based on Bitermial Parabolic Radon Transform

technical field

The invention belongs to the field of exploration geophysics, and in particular relates to a method for suppressing multiple waves based on bi-term parabolic Radon transformation.

Background technique

Multiples are an important part of seismic data processing. There are many methods to suppress multiples. Among them, the parabolic Radon transform is a widely used method to effectively attenuate multiples. The theoretical basis for multiple wave attenuation based on the parabolic Radon transform is the speed difference between the primary reflected wave and the multiple wave. It can separate the multiple wave and the primary reflected wave in the Radon domain, and then obtain the primary reflected wave data through inverse transformation. . Its essence is the summation of the amplitudes of seismic events in different curvature directions. It can transform a point in the Ladong domain into a parabola in the time-offset domain, but it is only a shape conversion without considering the seismic event. change in amplitude. Because the conventional parabolic Radon transform cannot simulate the variation of amplitude with offset, that is, the AVO characteristic of seismic data, and the AVO phenomenon is an important feature for accurate description of underground target reservoirs.

At present, when the seismic data with AVO phenomenon is attenuated based on the conventional parabolic Radon transform, the AVO amplitude information of the primary reflection will be damaged. Since the parabolic Radon transform is not an orthogonal transform, a bow-tie-like divergence will appear during the forward transform process, which will cause a tailing effect, reduce the resolution of the transformed data volume, distort the reconstructed data after the inverse transform, and cause amplitude leakage. It leads to errors in AVO analysis and reduces the reliability of subsequent seismic inversion and interpretation.

Contents of the invention

The purpose of the present invention is to provide a multiple wave suppression method based on the bi-term parabolic Radon transform, aiming to solve the problem that the current multiple wave attenuation technology based on the conventional Radon transform cannot simulate the AVO characteristics of seismic data, and the Artifacts, there is a tailing effect, and the amplitude of the reflected wave is distorted during the inverse transformation, causing the problem of amplitude leakage.

The present invention is achieved in this way, a kind of multiple wave suppressing method step based on biterm parabolic Radon transformation is as follows:

Step 1. Process the pre-stack CMP gathers: perform dynamic correction and dynamic correction stretching and cutting on the pre-stack CMP gathers;

Step 2, establishment of the objective function: establish the objective function J for obtaining the conversion data body of the Radon domain according to the following two-term Radon transformation formula;

$\begin{array}{c}\mathrm{<span\; class="notranslate">\; J</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; t</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}}\mathrm{<span\; class="notranslate">\; W</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; )</span>{<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; d</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>\underset{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}<span\; class="notranslate">\; \{</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; A</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; pH</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; f</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; pH</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \}</span><span\; class="notranslate">\; |</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}\\ <span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&mu;</span>}\underset{\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}\mathrm{<span\; class="notranslate">\; R</span>}<span\; class="notranslate">\; [</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; A</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; /</span>{\mathrm{<span\; class="notranslate">\; the\; s</span>}}_{\mathrm{<span\; class="notranslate">\; A</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; /</span>{\mathrm{<span\; class="notranslate">\; the\; s</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}<span\; class="notranslate">\; ]</span>\end{array}$

where the regularization term R is the sparse constraint on the transformed data volumes m _A (τ,p) and m _B (τ,p), W(t,h) is the weighting factor of the data space, μ is the balance parameter, s _A and s _B are two calibration parameters;

Step 3. Radon forward transform processing of pre-stack CMP gathers: introduce the regularization term R, use the sparse constraint group to simultaneously constrain the converted data body in the Radon domain, and finally use the reweighted least square method to iteratively solve the objective function J, Realize the high-resolution inversion of the bi-term Radon transform operator, and obtain high-resolution transformed data volumes m _A and m _B representing the motion characteristics and AVO characteristics of primary reflection waves and multiple waves respectively in the Radon domain;

Step 4. Separation processing of primary reflection wave and multiple wave in Ladong domain: identify primary reflection wave and multiple wave in Ladong domain, and assign zero to the data volume representing primary reflection wave, so as to realize primary reflection wave and multiple wave The separation of multiple motion characteristics and AVO characteristics in the Ladong domain;

Step 5. Estimation of primary reflection data: Perform Radon inverse transformation on the multiples separated in the Radon domain to obtain multiple data in the time domain, and then use the initial pre-stack gather data to subtract the multiplicity obtained by the inverse transformation. The secondary wave obtains primary reflected wave data.

Effect summary

The present invention solves the problem that the current multiple wave attenuation technology based on the conventional Radon transformation cannot simulate the AVO characteristics of seismic data, and artifacts and tailing effects are generated during forward transformation, and amplitude distortion of primary reflection waves and amplitude leakage occur during inverse transformation. The problem is that the transformation of the shape and amplitude of the seismic event in the Ladong domain can be realized at the same time, so that the motion characteristics and AVO features of the seismic data are better separated in the Ladong domain, and the artifacts produced during the forward transformation process are weakened. Accurate recovery of primary reflected wave energy is realized, AVO characteristics of seismic data are protected, and the reliability of subsequent seismic inversion and interpretation is improved.

Description of drawings

Fig. 1 is the parameter description in the double term Radon transform provided by the embodiment of the present invention;

Fig. 2 is the synthesized CMP gather when Δτ ₀ and Δτ _hmax are both 0.03s provided by the embodiment of the present invention;

Fig. 3 is the synthesized CMP gather when Δτ ₀ and Δτ _hmax are both 0.01s provided by the embodiment of the present invention;

Fig. 4 is when Δτ ₀ is 0.03s provided by the embodiment of the present invention, the root mean square error that varies with different time differences Δτ _hmax at the far offset;

Fig. 5 is provided by the embodiment of the present invention when Δτ ₀ is 0.01s, the root mean square error that varies with different time differences Δτ _hmax at the far offset;

Figure 6 shows the difference between the actual primary reflection wave and the Radon transform inversion results when Δτ ₀ and Δτ _hmax provided by the embodiment of the present invention are both 0.03s

In the figure a) is Δτ ₀ , and when Δτ _hmax is both 0.03s, the difference between the actual primary reflection high resolution and the conventional Radon transform inversion results

In the figure b) is Δτ ₀ , and when Δτ _hmax is both 0.03s, the difference between the actual high-resolution primary reflection wave and the high-resolution dual-term Radon transform inversion results

Fig. 7 is the root mean square error of the variation of the amplitude with the offset when Δτ ₀ and Δτ _hmax are both 0.03s provided by the embodiment of the present invention;

Figure 8 shows the difference between the actual primary reflection wave and the Radon transform inversion results when Δτ ₀ and Δτ _hmax provided by the embodiment of the present invention are both 0.01s

In the figure a) is Δτ ₀ , when Δτ _hmax is 0.01s, the difference between the actual primary reflection wave and the high-resolution conventional Radon transform result

In the figure b) is Δτ ₀ , when Δτ _hmax is both 0.01s, the difference between the actual primary reflection wave and the high-resolution binomial Radon transform inversion result;

Fig. 9 is the root mean square error of the variation of the amplitude with the offset when Δτ ₀ and Δτ _hmax are both 0.01s provided by the embodiment of the present invention;

Fig. 10 is the primary reflection wave obtained by the actual seismic data provided by the embodiment of the present invention, the high-resolution conventional Radon transform and the high-resolution bi-term Radon amplitude-preserving inversion

In the figure, a) is the actual seismic data, b) is the primary reflection wave obtained by high-resolution conventional Radon transform, and c) is the primary reflection wave obtained by high-resolution dual-term Radon amplitude-preserving inversion;

Fig. 11 is the actual seismic data after the location of the target layer is amplified and the primary reflection waves inverted by the high-resolution conventional Radon transform and the high-resolution dual-term Radon transform provided by the embodiment of the present invention

In the figure: a) is the actual seismic data, b) is the high-resolution conventional Radon transform, and c) is the primary reflection wave inverted by the high-resolution dual-term Radon transform;

Fig. 12 is a flowchart of a method for suppressing multiple waves based on a bi-termal parabolic Radon transform provided by an embodiment of the present invention.

Detailed ways

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

The present invention is achieved in this way, a kind of multiple wave suppressing method step based on double parabolic Radon transform is as follows:

S101: Process the pre-stack CMP gathers: perform motion correction and motion correction stretching and resection processing on the pre-stack CMP gathers;

S102: Establishment of the objective function: establish the objective function J for obtaining the conversion data body of the Radon domain according to the following two-term Radon transformation formula;

$\begin{array}{c}\mathrm{<span\; class="notranslate">\; J</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; t</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}}\mathrm{<span\; class="notranslate">\; W</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; )</span>{<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; d</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>\underset{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}<span\; class="notranslate">\; \{</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; A</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; pH</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; f</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; pH</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \}</span><span\; class="notranslate">\; |</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}\\ <span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&mu;</span>}\underset{\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}\mathrm{<span\; class="notranslate">\; R</span>}<span\; class="notranslate">\; [</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; A</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; /</span>{\mathrm{<span\; class="notranslate">\; the\; s</span>}}_{\mathrm{<span\; class="notranslate">\; A</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; /</span>{\mathrm{<span\; class="notranslate">\; the\; s</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}<span\; class="notranslate">\; ]</span>\end{array}$

In the formula, the regularization term R is the sparse constraint on the transformed data volumes m _A (τ,p) and m _B (τ,p), and W(t,h) is the weighting factor of the data space. If only the inversion of m _A (τ,p), the algorithm becomes a high-resolution parabolic Radon transform in the conventional time domain, and the balance parameter μ is used to smooth the relationship between data matching and sparse constraints; s _A and s _B are two calibration parameters; The regularization term R based on sparse constraints can be expressed in the form of R(x,y)=log(1+x ² +y ² ), where x and y represent the required data body; here corresponds to the sparse constraint group, That is, m _A (τ, p) and m _B (τ, p) are regarded as a group, and sparse constraints are imposed on them at the same time;

S103: Radon forward transformation processing of pre-stack CMP gathers: introduce regularization item R, use sparse constraint group to constrain the transformation data body of Radon domain at the same time, and finally use reweighted least square method to iteratively solve the objective function J to realize The high-resolution inversion of the dual-term Radon transform operator obtains the high-resolution transformed data volumes m _A and m _B representing the motion characteristics and AVO characteristics of the primary reflection and multiple waves respectively in the Radon domain;

S104: Separation processing of the primary reflection wave and the multiple wave in the Ladong domain: identify the primary reflection wave and the multiple wave in the Ladong domain, and assign the data volume representing the primary reflection wave to zero to realize the primary reflection wave and the multiple wave The separation of the movement characteristics of the secondary wave and the AVO characteristics in the Ladong domain;

S105: Estimation of primary reflection data: perform Radon inverse transformation on the multiples separated in the Radon domain to obtain multiple data in the time domain, and then subtract the multiples obtained by the inverse transformation from the initial pre-stack gather data Wave to get the reflected wave data once.

Model analysis:

To test the feasibility and superiority of the method, the formation reflection coefficient generated by the third type of AVO model proposed by Ostrander was convolved with the 30Hz seismic wavelet to generate a seismic gather including primary reflections and multiples, Δτ _max and Δτ ₀ represents the dynamic correction difference between multiple waves and primary reflection waves at the maximum and minimum offsets, respectively, and the schematic diagram of its parameters is shown in Figure 1. Figure 2 and Figure 3 are the synthetic CMP gathers when Δτ ₀ and Δτ _hmax are both 0.03s and 0.01s, respectively. It can be seen from the figure that the amplitude of the primary reflected wave increases with the increase of the offset, and has obvious characteristics of the third type of AVO, while the AVO phenomenon of multiple waves is not obvious, which is consistent with the actual geological phenomenon. The reason is that the generation of multiple waves often requires multiple reflections, and the absorption of the formation reduces the energy of the multiple waves, and the characteristics of the amplitude changing with the offset are also weakened. Figure 4 and Figure 5 show the root mean square error of different time differences Δτ _hmax at relatively far offset distances when Δτ ₀ is 0.03s and 0.01s respectively Among them, the solid line represents the result obtained based on the bi-term Radon transformation, and the dotted line represents the result obtained based on the conventional Radon transformation. It can be seen from the figure that with the increase of Δτ _hmax , the error value gradually becomes smaller, but no matter whether Δτ ₀ is 0.03s or 0.01s, compared with the conventional Radon transform, the double-term Radon transform suppresses many times The wave makes the root mean square error of the result relatively small. Figure 6 and Figure 8 are Δτ ₀ , Δτ _hmax are both 0.03s and 0.01s, the actual primary reflection and a) based on high-resolution conventional Radon transform and b) based on high-resolution dual-term Radon transform inversion The difference between the results shows that the error obtained by conventional Radon transformation is relatively large. Figure 7 and Figure 9 show the root mean square error of the variation of the amplitude with the offset when Δτ ₀ and Δτ _hmax are both 0.03s and 0.01s. It can be seen from the comparison that, no matter at the near offset or the far offset, compared with the result obtained by the high-resolution conventional Radon transform, the amplitude of the primary reflection wave estimated based on the high-resolution bi-term Radon transform method is relatively small. More accurate. This is because when the conventional Radon transform is used to suppress the multiples, the tailing effect caused by the amplitude change cannot separate the multiples in the Radon domain from the primary reflection well, so that the recovered primary reflection The amplitude error is relatively large. Relatively speaking, because the bi-term Radon transform considers the AVO characteristics of the data, the tailing effect of the Radon domain data is reduced, and the multiple waves and primary reflection waves can be separated better, thus obtaining better Amplitude recovery.

Therefore, through the model test, it can be seen that the multiple wave suppression method based on the dual-term Radon transform proposed by the present invention can realize a single seismic event by converting the shape and amplitude of the seismic event separately, and under the constraints of the sparse regularization term. Accurate recovery of reflected wave data effectively protects the AVO phenomenon of seismic data.

Embodiment one

Figure 10 shows a) the seismic data with the same color scale range and the primary reflection recovered by b) high-resolution conventional Radon transform and c) high-resolution dual-term Radon transform method, and the selected time range is 1.6s By 4.8s, the target reservoir is around 2.7s. From the primary reflected wave data shown in Fig. 10b) and Fig. 10c), it can be seen that the multiple waves are basically eliminated, and the energy of the primary reflected wave is well recovered. At the position indicated by the arrow in the figure, there is a relatively obvious difference between the two.

Figure 11 enlarges the data in the dotted rectangular box in Figure 10. It can be seen from the figure that in the area indicated by the arrow in the black box, the event of the primary reflected wave obtained by the conventional Radon transform is obviously distorted, destroying the The AVO characteristics of the seismic data will be affected, which will affect the identification of the target reservoir, and there are obvious multiple wave residues in the area indicated by the black arrow; The primary reflected wave in the area indicated by the arrow is well restored, which effectively protects the AVO phenomenon of the seismic data and lays the foundation for the identification of the target reservoir, and the multiple waves in the area indicated by the black arrow are also well removed , which verifies the advantages of the binomial Radon transform in the processing of multiple wave attenuation and primary reflection wave recovery and the effectiveness of this research method.

The present invention solves the problem that the current multiple wave attenuation technology based on the conventional Radon transformation cannot simulate the AVO characteristics of seismic data, and artifacts and tailing effects are generated during forward transformation, and amplitude distortion of primary reflection waves and amplitude leakage occur during inverse transformation. The problem is that the transformation of the shape and amplitude of the seismic event in the Ladong domain can be realized at the same time, so that the motion characteristics and AVO features of the seismic data are better separated in the Ladong domain, and the artifacts produced during the forward transformation process are weakened. Accurate recovery of primary reflected wave energy is realized, AVO characteristics of seismic data are protected, and the reliability of subsequent seismic inversion and interpretation is improved.

Although the specific implementation of the present invention has been described above in conjunction with the accompanying drawings, it is not a limitation to the protection scope of the present invention. Those skilled in the art should understand that on the basis of the technical solution of the present invention, those skilled in the art do not need to pay any creative effort. Various modifications or deformations that can be made by labor are still within the protection scope of the present invention.

Claims (2)

1. a multiple wave suppressing method based on bi-term parabolic Radon transform, is characterized in that, described multiple wave suppressing method step based on bi-term parabolic Radon transform is as follows: Step 1. Process the pre-stack CMP gathers: perform dynamic correction and dynamic correction stretching and cutting on the pre-stack CMP gathers; Step 2, establishment of the objective function: establish the objective function for obtaining the conversion data body of the Radon domain according to the bi-term Radon transformation formula; Step 3. Radon forward transform processing of pre-stack CMP gathers: introduce the regularization term R, use the sparse constraint group to simultaneously constrain the converted data body in the Radon domain, and finally use the reweighted least square method to iteratively solve the objective function J, Realize the high-resolution inversion of the bi-term Radon transform operator, and obtain high-resolution transformed data volumes m _A and m _B representing the motion characteristics and AVO characteristics of primary reflection waves and multiple waves respectively in the Radon domain; Step 4. Separation processing of primary reflection wave and multiple wave in Ladong domain: identify primary reflection wave and multiple wave in Ladong domain, and assign zero to the data volume representing primary reflection wave, so as to realize primary reflection wave and multiple wave The separation of multiple motion characteristics and AVO characteristics in the Ladong domain; Step 5. Estimation of primary reflection data: Perform Radon inverse transformation on the multiples separated in the Radon domain to obtain multiple data in the time domain, and then use the initial pre-stack gather data to subtract the multiplicity obtained by the inverse transformation. The secondary wave obtains primary reflected wave data. 2. the method for suppressing multiple waves based on two-term parabolic Radon transform as claimed in claim 1, is characterized in that, step 2, the establishment of objective function: set up and seek Radon domain according to following two-term Radon transform formula The objective function J of the converted data body; $\begin{array}{c}\mathrm{<span\; class="notranslate">\; J</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; t</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}}\mathrm{<span\; class="notranslate">\; W</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; )</span>{<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; d</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>\underset{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}<span\; class="notranslate">\; \{</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; A</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; pH</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; f</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; pH</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \}</span><span\; class="notranslate">\; |</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}\\ <span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&mu;</span>}\underset{\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}\mathrm{<span\; class="notranslate">\; R</span>}<span\; class="notranslate">\; [</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; A</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; /</span>{\mathrm{<span\; class="notranslate">\; the\; s</span>}}_{\mathrm{<span\; class="notranslate">\; A</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; /</span>{\mathrm{<span\; class="notranslate">\; the\; s</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}<span\; class="notranslate">\; ]</span>\end{array}$ where the regularization term R is the sparse constraint on the transformed data volumes m _A (τ,p) and m _B (τ,p), W(t,h) is the weighting factor of the data space, μ is the balance parameter, s _A and s _B are two calibration parameters.