CN107678062A - The integrated forecasting deconvolution of hyperbolic Radon domains and feedback loop methodology multiple suppression model building method - Google Patents
The integrated forecasting deconvolution of hyperbolic Radon domains and feedback loop methodology multiple suppression model building method Download PDFInfo
- Publication number
- CN107678062A CN107678062A CN201710843785.2A CN201710843785A CN107678062A CN 107678062 A CN107678062 A CN 107678062A CN 201710843785 A CN201710843785 A CN 201710843785A CN 107678062 A CN107678062 A CN 107678062A
- Authority
- CN
- China
- Prior art keywords
- mrow
- deconvolution
- msub
- tau
- radon
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
- G01V1/28—Processing seismic data, e.g. analysis, for interpretation, for correction
- G01V1/36—Effecting static or dynamic corrections on records, e.g. correcting spread; Correlating seismic signals; Eliminating effects of unwanted energy
- G01V1/364—Seismic filtering
- G01V1/368—Inverse filtering
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V2210/00—Details of seismic processing or analysis
- G01V2210/20—Trace signal pre-filtering to select, remove or transform specific events or signal components, i.e. trace-in/trace-out
- G01V2210/24—Multi-trace filtering
- G01V2210/244—Radon transform
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V2210/00—Details of seismic processing or analysis
- G01V2210/30—Noise handling
- G01V2210/32—Noise reduction
- G01V2210/324—Filtering
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V2210/00—Details of seismic processing or analysis
- G01V2210/30—Noise handling
- G01V2210/36—Noise recycling, i.e. retrieving non-seismic information from noise
Abstract
The invention provides the integrated forecasting deconvolution of hyperbolic Radon domains and the model building method of feedback cycle method multiple suppression.Read in it is original it is shallow cut open data after, be utilized respectively predictive deconvolution and the processing of feedback cycle method, and carry out Hyperbola Radon Transform；Sef-adapting filter is built, as its output valve FmWhen (τ, h) is approximately 0, it is believed that effective reflection also be present, then need to return to iterative processing again, until Fm(τ, h) is approximately equal to 1, terminates iterative processing；Anti- Radon conversion is carried out to multiple wave energy model in hyperbolic Radon domains；Multiple wave pattern data are subtracted with original shallow seismic profile data, finally obtain the high reliability shallow seismic profile data after multiple wave pressure system.The present invention can the effectively more subwaves in stranglehold interface and interbed multiple, significant wave can not be damaged again, realize efficiently separating for more subwaves and primary wave, understand more reliable foundation is provided for the seabed underground medium in later stage.
Description
Technical field
The present invention relates to Marine Geology engineering field, the more particularly to integrated forecasting deconvolution of hyperbolic Radon domains and feedback is followed
Ring method suppresses the shallow multiple wave pattern construction method in stratum.
Background technology
Due to being influenceed by the complexity of the shallow sedimentation thing structure of metering system and seabed, sub-bottom profiler is (shallow
Cut open) transducer receive battle array inevitably receive some multiple reflections from sea and shallow stratum interlayer, to shallow
Layer effectively primary wave separation brings difficulty, has also severely impacted the authenticity and reliability of shallow formation imaging, and then mislead
The shallow stratum substrate in later stage is explained.The method of multiple suppression mainly has two classes at present：One kind is based on significant wave and repeatedly
The filtering method of difference between wave characteristic and property, it is such including (Foster such as predictive deconvolution, FK conversion, Radon conversion
and Mosher,1992；Yilmaz and Taner,1994)；Another kind of is the prediction subtractive method based on wave theory, such
Including wave field extrapolation method, feedback cycle method, backscattering progression method etc. (Morley and Claerbout, 1983；Weglein et
al.,1997).Predictive deconvolution principle be according to more subwaves have periodically and primary wave does not possess periodic differential separation
Out, suitable for the medium with larger positive velocity gradient, there is the characteristics of algorithm is simple, calculating cost is low.Taner is proposed
Linear Radon domains predictive deconvolution method, the ability of the more subwaves of remote offset distance in compacting is improved, it is again that single track prediction is anti-afterwards
Convolution develops into two dimension, predicts primary wave from the convolution of different filtering factors using adjacent seismic channel, adds horizontal to stratum
The adaptability (Song family text et al., 2014) of change.Lokshtanov confirms that linear Radon domains predictive deconvolution also can
The more subwaves of effective attenuation (Lokshtanov, 1995；1999).Zhao Chang bases etc. analyze the parameter of linear Radon predictive deconvolutions
Choose with applicable sex chromosome mosaicism, and the compacting of South China Sea Deep Water Multiple Attenuation in Seismic Data is obtained certain effect (Zhao Chang builds et al.,
2013).Verschuur and Berkhout proposes the feedback cycle method of multiple suppression, and whole prediction process need not know macroscopic view
Velocity field, so as to add the adaptability of Forecasting Methodology.Subsequent two people is again based on the theoretical proposition 2D series expansion pressures of feedback model
The more subwave algorithms of control surface, behind development improve theory, without to multiple prediction terms summation in the case of, based on feedback
Model theory propose the more subwave algorithms of iteration pressed surface, improve to a certain extent computational accuracy (Verschuur, 1992；
Verschuur et al.,1995；Berkhout and Verschuur,1997；Verschuur and Berkhout,
1997；Berkhout,1999).Based on the thought, Kelamis, Wang, Van Groenestjin etc. are successively to based on feedback
Circulation law theory and application carried out it is beneficial improve (Kelamis and Verschuur, 2000；Wang,2004；
2007；Van Groenestijn and Verschuur,2009a；Van Groenestijn and Verschuur,
2009b).Radon conversion METHOD OF SUPPRESSION OF MULTIPLE WAVESs derive from the pie slice concept that Ryu is proposed, when more subwaves and primary wave are deposited
In speed difference, multiple wave energy and primary wave energy imaging to the different velocity spaces, can be by primary waves in the velocity space
With multiple wavelength-division from.Xue Yaru proposes that a kind of multi-direction orthogonal polynomial based on Radon conversion and orthogonal polynomial transformation becomes
METHOD OF SUPPRESSION OF MULTIPLE WAVES is changed, this method only can describe lineups residual move out time parameter with a curvature parameters, significantly improve
The remaining time resolution ratio of primary wave and more subwaves (Xue Ya eat et al., 2012).Gong Xiangbo is directed to different and long in speed
In the case of offset distance, geological data is not Radon domains self-energy restrains still the problem of, it is proposed that anisotropy Radon conversion compactings
Multiple wave method, derived by offset distance, slowness, the state modulator of non-ellipticity three the positive reconstructed formula of integral curve, avoid
Big matrix operation in time-domain Sparse Pulse iterative inversion, maintains higher precision, and improve computational efficiency (consolidate to
Rich et al., 2014).
However, as described above, there is following defect in prior art：
(1) when the SVEL on shallow stratum reverses or lateral velocity conversion is violent, predictive deconvolution fails effectively to press
More subwaves are made, or even significant wave can be damaged；
(2) feedback cycle method needs to know the wavelet and underground medium knot of known transmitting transducer in Multiple attenuation
Structure, and the often more difficult accurate acquisition of underground medium structure, so as to have influence on whether precisely prediction simultaneously Multiple attenuation；And
(3) Radon conversion is not suitable for multiple wave pressure system caused by non-horizontal homogeneous layered medium, and is bent repeatedly to micro-
Ripple pressing result is poor.
The content of the invention
The technical problem to be solved in the present invention is in analysis predictive deconvolution method comprehensively, feedback cycle method and Radon changes
On the basis of method advantage and disadvantage, with reference to the shallow seismic profile data characteristicses of acquisition, there is provided a kind of hyperbolic Radon domains integrated forecasting is anti-
The model building method of convolution and feedback cycle method multiple suppression, realizes efficiently separating for more subwaves and primary wave, is the later stage
Seabed underground medium understand more reliable data be provided.
For achieving the above object, the technical problem to be solved in the present invention is included following aspects：
(1) more subwaves are predicted with feedback cycle method and predictive deconvolution method while respectively, then both are predicted multiple
Wave component and prediction error carry out Hyperbola Radon Transform.
(2) for including effective reflection and the long-period multiple do not suppressed thoroughly in predictive deconvolution prediction error,
Sef-adapting filter is designed, by effective reflection Regional resection, retains multiple wave energy.
(3) as sef-adapting filter output valve FmWhen (τ, h) is approximately 0, it is believed that effective reflection also be present, then need to return
Iterative processing again, until Fm(τ, h) is approximately equal to 1, terminates iterative processing.Afterwards, the multiple number of channels obtained with reference to two methods
According to model, the unified multiple ripple data model in hyperbolic Radon domains is obtained, then by anti-Hyperbola Radon Transform, finally from original shallow
Cut open data and subtract multiple wave number according to the synthesis pressing process that can complete more subwaves.
Therefore, the mould of hyperbolic Radon domains integrated forecasting deconvolution provided by the invention and feedback cycle method multiple suppression
Type construction method, comprises the following steps：
(1) original shallow seismic profile data are read in, it is more with predictive deconvolution and feedback loop methodology prediction respectively first
Secondary wave component；
(2) Hyperbola Radon Transform is carried out to more subwaves of both the above method prediction；
(3) for still having long-period multiple in predictive deconvolution prediction error, sef-adapting filter is built, is filtered out
Significant wave, and then obtain with the multiple wave energy model in hyperbolic Radon domains after two methods；
(4) as sef-adapting filter output valve FmWhen (τ, h) is approximately 0, it is believed that effective reflection also be present, then need to return
Iterative processing again, until Fm(τ, h) is approximately equal to 1, terminates iterative processing.
(5) anti-Radon conversion is carried out to multiple wave energy model in hyperbolic Radon domains；
(6) multiple wave pattern data are subtracted with original shallow seismic profile data, the height finally obtained after multiple wave pressure system can
By property shallow seismic profile data.
In one embodiment of the invention, the mathematical modeling of predictive deconvolution is：
In one embodiment of the invention, the mathematical modeling of feedback cycle is：
In one embodiment of the invention, the mathematical modeling of Hyperbola Radon Transform is：
In one embodiment of the invention, the mathematical modeling of sef-adapting filter is：
Pass through above-mentioned technical proposal, the beneficial effects of the invention are as follows：
(1) the shallow seismic profile data of neritic area are directed to, due to more by the shallow stratum interlayer of sea free wave and seabed
Secondary wave action is more serious, the comprehensive analysis advantage and disadvantage of predictive deconvolution method and feedback cycle method, using both mutual supplement with each other's advantages,
It is proposed the model building method of a kind of hyperbolic Radon domains integrated forecasting deconvolution and feedback cycle method multiple suppression.
(2) a kind of sef-adapting filter is designed, predictive deconvolution being predicted to, the significant wave in error is cut off, and then obtain
The higher multiple wave energy model of composition, recycle initial data to subtract multiple wave energy model data, be finally reached effective pressure
Make the purpose of more subwaves.
Brief description of the drawings
, below will be to required in embodiment or description of the prior art in order to illustrate more clearly of the technical characteristic of the present invention
The accompanying drawing used is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the present invention, right
For those of ordinary skill in the art, on the premise of not paying creative work, it can also be obtained according to these accompanying drawings
His relevant drawings.
Fig. 1 is the flow chart of predictive deconvolution multiple suppression；
Fig. 2 is predictive deconvolution and feedback cycle synthesis multiple suppression flow.
Embodiment
In order that the technical means, the inventive features, the objects and the advantages of the present invention are easy to understand, tie below
Conjunction is specifically illustrating, and the present invention is expanded on further.
First, the present invention relates to following technical term：
Predictive deconvolution method
Predictive deconvolution can be described as a kind of Least square deconvolution of broad sense in some sense, and it is had based on more subwaves
There is periodic feature, and the process of the composition periodically occurred in initial data is eliminated by designing prediction error filter
(Wang Zhaoqi, 2007).Predictive deconvolution can be applied to prestack records data, can also be applied to poststack record data.
Feedback cycle method
Feedback cycle method is to predict more subwaves based on free interface and stratum boundary surface model, is believed without shallow stratum media
Breath, even if more subwaves and the time difference of effective reflection are little, it can also predict more subwaves (Shen is grasped, 2003).It is generally used for
Prestack records data.
Hyperbolic Radon domains convert
The Hyperbolic Feature that the conversion of hyperbolic Radon domains records using earthquake common-source point or CMP, along hyperbolic circuit
FootpathData are carried out first time shaft Square Transformation (under time shaft square coordinate, time shaft square with
Intercept time square), it the operator of frequency domain can be used to carry out the positive and negative conversion of Radon, result is entered again after conversion
The conversion of row time shaft evolution obtains Hyperbola Radon Transform result (consolidate to rich et al., 2016).
The mould of hyperbolic Radon domains integrated forecasting deconvolution provided by the invention and feedback loop methodology multiple suppression is built
Method comprises the following steps：
(1) multiple wave component is predicted with predictive deconvolution and feedback loop methodology respectively；
(2) Hyperbola Radon Transform is carried out to more subwaves of both the above method prediction；
(3) for still having long-period multiple in predictive deconvolution prediction error, sef-adapting filter is built, is filtered out
Significant wave, and then obtain with the multiple wave energy model in hyperbolic Radon domains after two methods；
(4) as sef-adapting filter output valve FmWhen (τ, h) is approximately 0, it is believed that effective reflection also be present, then need to return
Iterative processing again, until Fm(τ, h) is approximately equal to 1, terminates iterative processing.
(5) anti-Radon conversion is carried out to multiple wave energy model in hyperbolic Radon domains；
(6) multiple wave pattern data are subtracted with original shallow seismic profile data, the height finally obtained after multiple wave pressure system can
By property shallow seismic profile data.
Referring to shown in Fig. 1 to Fig. 2, the specific embodiment of the present invention is as follows：
(1) overall technology path
Fig. 2 is the integrated forecasting deconvolution of hyperbolic Radon domains and feedback loop methodology multiple suppression model building method
Schematic diagram, first by it is original it is shallow cut open data and pre-processed respectively by predictive deconvolution and feedback cycle method, followed using feedback
It is around-France to predict the positions of more subwaves in data roughly in original shallow cut open, as multiple radio frequency channel data model, predict come it is multiple
Radio frequency channel data model without strictly matching completely with more subwaves of initial data.Meanwhile with reference to the result of predictive deconvolution,
Due to being mainly distributed on tool comprising effective reflection and the long-period multiple do not suppressed thoroughly, multiple wave energy in prediction error
Have on the road of greater curvature parameter (Zeng Zhongyu, 2013).Secondly, the multiple wave component and prediction error both predicted are carried out
Hyperbolic Radon is converted.Then the sef-adapting filter designed by the present invention, effective reflection Regional resection retains multiple
Wave energy.Work as FmWhen (τ, h) is approximately 0, if effective reflection also be present, need to return to iterative processing again, until Fm(τ,h)
It is approximately equal to 1, terminates iterative processing.Afterwards, the multiple radio frequency channel data model obtained with reference to two methods, hyperbolic Radon domains are obtained
Unified multiple ripple data model, then converted by anti-hyperbolic Radon, filtered repeatedly ripple data model can be obtained, i.e.,
Remain and shallow existing for more subwave lineups cut open data.Finally multiple wave number is subtracted according to can complete repeatedly from original shallow data of cuing open
The synthesis pressing process of ripple.
(2) predictive deconvolution is theoretical
Predictive deconvolution is that have periodic feature based on more subwaves, and eliminates original by designing prediction error filter
The process (Wang Zhaoqi, 2007) of the composition periodically occurred in beginning data.This method at zero-offset to having the preferable cycle
The multiple wave pressure significant effect of property, and to periodically multiple wave pressure system is ineffective at remote offset distance in, so this side
Method is difficult to preferably suppress more subwaves in the range of whole offset distance.
Predictive deconvolution general principle is, if shallow cut open that prediction step is l is recorded as shown in formula (9), second
Section 1 represents prediction error (being coincide with primary reflection) on the right of equal sign, and Section 2 isRepresent prediction
More subwaves (Liu Jun, 2008).
D (t) is set again cuts open signal currency to be shallow, d (t-i), i=1,2,3 ... signal past value is cutd open to be shallow, is utilized
It is shallow to cut open signal currency and past value integrated forecasting its future valueI.e.：
T represents conversion in above formula, and linear time-invariant system, predicted value can be represented with convolution model, i.e.,：
Wherein, c (t)={ c (0), c (1) ... } is linear predictor.For the t+l moment, prediction error is：
In above formula d (t+l) be the t+l moment it is shallow cut open signal desired value,For real output value, as more subwaves.Selection
Suitable predictive factor predicts more subwaves, and using it is original it is shallow cut open data and subtract multiple wave component and can obtain effectively reflect one
Subwave, if the prediction error e (t+l) in formula (12) is effective primary wave.The flow of predictive deconvolution method multiple suppression is such as
Shown in Fig. 1, window data when inputting shallow seismic profile data record and parameter first, and taking k-th；Next solves Toeplitz
Matrix；Then solution system of linear equations seeks filtering factor c (m)；Finally ask deconvolution operator with when window in per trace record convolution,
Exported after all track datas have been handled and shallow cut open record data.
(3) feedback cycle law theory
Feedback cycle method is to predict more subwaves based on free interface and stratum boundary surface model, is believed without shallow stratum media
Breath, even if more subwaves and the time difference of effective reflection are little, can also predict more subwaves.It is more being suppressed using feedback cycle method
Two conditions that need to meet during subwave are：1. the near migration range track data for needing repairing to lack；2. need to estimate sound source wavelet.
The principle of feedback cycle method is, first inverting acoustic wavefield, and the more subwaves relevant with free interface are predicted to come,
Then the more subwaves (Shi Ying et al., 2011) predicted are subtracted from original wave field.Feedback cycle method is free by forward modeling
Surface-related multiple wave field is weakened, and is derived from the iterative algorithm suppressed by surface-related multiple herein.
By D (z0)=D0(z0)+D0(z0)A(z0)D(z0) can obtain：
D0(z0)=D (z0)-D0(z0)A(z0)D(z0) (13)
D is combined by formula (13)(0)=D0Obtaining iteration form is：
Formula (14) is the iterative inversion formula of Free Surface more subwaves filtering, and n is iterations, D (z0) it is original shallow cut open
Data (include more subwaves), D0 (n)(z0)、D0 (n+1)(z0) it is respectively except the shallow of multiple elimination cuts open after n-th and (n+1)th iteration
Number can use different free interface operator A it is estimated that each iterative process(n+1)(z0)。
(4) comprehensive drawing method
1) constant Hyperbola Radon Transform during frequency domain
Because the shallow back wave cutd open is curve and needs to ensure High Resolution, therefore it is contemplated that and uses hyperbolic
Radon is converted.Furthermore taken very much due to carrying out Hyperbola Radon Transform in time domain, and practicality is not strong, therefore be contemplated that
Changed on frequency domain.When constant Hyperbola Radon Transform can be represented by the formula (Jia Chun plum et al., 2016)：
In formula (15), d (x, t), which is that time domain is shallow, cuts open data, and h is ray parameter, and τ is the intercept time, zrefFor reference depth, m
(τ, h) is Hyperbola Radon Transform normal cross-section.
Fourier transformation is carried out respectively at formula (15) both ends, can be obtained：
Wherein θ is represented by：
Single-frequency components are cutd open for shallow, the positive inverse transformations of hyperbolic Radon can be represented with following two formulas：
M=LD (18)
D=LHM (19)
In above formula, LHFor L associate matrix, L is represented by following formula：
For overdetermined equation, formula (18) is represented by following least square form：
M=(LLH+λI)-1LD (21)
When ray parameter is equal interval sampling, LLHFor Toeplitz matrixes, carried out using Levinson recurrence methods
Solve, there is higher computational efficiency.
2) sef-adapting filter
Due to predictive deconvolution for it is original it is shallow to cut open macrocyclic more subwave filtration results in data not notable, by prediction
After deconvolution processing, most long-period multiple is still left in effective reflection, it is therefore desirable to design one adaptively
Wave filter, real effectively primary reflection is realized with long-period multiple and separated, effective wave energy is removed, retains more subwaves
Energy.Sef-adapting filter is represented by：
In formula (22), D (τ, h) represents to cut open data, M (τ, h) table comprising the shallow of more subwaves after predictive deconvolution is handled
Show the Hyperbola Radon Transform of multiple wave pattern data, η represents the rejecting parameter of wave filter, generally takes the exhausted of bottom reflection coefficient
To value, m represents the smooth control parameter of filtering, generally takes the even number within 10.Work as FmWhen (τ, h) is approximately 0, it is believed that be not present
More subwaves, and work as FmWhen (τ, h) is approximately 1, it is believed that more subwaves be present.
3) comprehensive drawing method flow
Hyperbolic Radon domain integrated forecasting Deconvolutions and the flow of feedback cycle method multiple suppression are as shown in Figure 2.First
By it is original it is shallow cut open data and pre-processed respectively by predictive deconvolution and feedback cycle method, using feedback cycle method original shallow
The position for predicting more subwaves in data roughly is cutd open, as multiple radio frequency channel data model, predicts the multiple radio frequency channel data model come
Without strictly being matched completely with more subwaves of initial data.Meanwhile with reference to the result of predictive deconvolution, due to predicting error
In comprising effective reflection and the long-period multiple do not suppressed thoroughly, multiple wave energy is mainly distributed on greater curvature ginseng
On several roads (Zeng Zhongyu, 2013).Secondly, the multiple wave component and prediction error both predicted carry out hyperbolic Radon changes
Change.Then by the sef-adapting filter of design, by effective reflection Regional resection, multiple wave energy is retained.Work as Fm(τ,h)
When being approximately 0, it is believed that effective reflection also be present, then need to return to iterative processing again, until Fm(τ, h) is approximately equal to 1, terminates to change
Generation processing.Afterwards, the multiple radio frequency channel data model obtained with reference to two methods, the unified multiple wave number in hyperbolic Radon domains is obtained
According to model, then by anti-Hyperbola Radon Transform, filtered repeatedly ripple data model can be obtained, that is, it is same to remain more subwaves
It is shallow existing for phase axle to cut open data.Finally from it is original it is shallow cut open data and subtract multiple wave number suppressed according to the synthesis that can complete more subwaves
Journey.
As can be seen here, solves following technical problem in the present invention：
(1) more serious phenomenon, total score are influenceed by sea free wave and shallow stratum interbed multiple for neritic area
The advantage and disadvantage of predictive deconvolution method and feedback cycle method have been analysed, using both mutual supplement with each other's advantages, have proposed that both comprehensive compactings are multiple
The model construction scheme of ripple.
(2) for including effective reflection and the long-period multiple do not suppressed thoroughly in predictive deconvolution prediction error,
From design sef-adapting filter, by effective reflection Regional resection, retain multiple wave energy, finally subtracted from initial data more
Subwave energy datum obtains the shallow seismic profile data of high reliability.
In addition, the technical characterstic of the present invention is as follows：
(1) in fully analysis neritic area shallow seismic profile data characteristicses, and predictive deconvolution method and feedback cycle method
On the basis of advantage and disadvantage, integrated forecasting Deconvolution and the model of feedback cycle method multiple suppression are built.
(2) predicted due to predictive deconvolution and effective reflection and the long-period multiple do not suppressed thoroughly included in error,
Angle is set out from the negative, and effective reflection is thoroughly eliminated by successive ignition process, and then obtains higher multiple of material
Wave energy data, finally subtract multiple wave energy data from initial data and obtain the shallow seismic profile data of high reliability.
General principle, principal character and the advantages of the present invention of the present invention has been shown and described above.The technology of the industry
Personnel are it should be appreciated that the present invention is not limited to the above embodiments, and the simply explanation described in above-described embodiment and specification is originally
The principle of invention, various changes and modifications of the present invention are possible without departing from the spirit and scope of the present invention, these changes
Change and improvement all fall within the protetion scope of the claimed invention.The claimed scope of the invention by appended claims and its
Equivalent defines.
Bibliography：
A.J.Berkhout,1999.Multiple removal based on the feedback
model.Leading Edge.18,127-131.
A.J.Berkhout,D.J.Verschuur,1997.Estimation of multiple scattering by
iterative inversion,Part I:Theoretical considerations.Geophysics.62,1586-
1595.
D.J.Foster,C.C.Mosher,1992.Suppression of multiple reflections using
the Radon transform.Geophysics.57, 386-395.
P.G.Kelamis,D.J.Verschuur,2000.Surface-related multiple elimination
on land seismic data—Strategies via case studies.Geophysics.65,719-734.
D.Lokshtanov,1995.Multiple suppression by single channel and
multichannel deconvolution in the tau-p domain.Seg Technical Program Expanded
Abstracts.14,1482-1485.
D.Lokshtanov,1999.Multiple suppression by data‐consistent
deconvolution.Seg Technical Program Expanded Abstracts.18,115-119.
L.Morley,J.Claerbout,1983.Predictive deconvolution in shot‐receiver
space.Geophysics.48,515-531. G.J.A.Van Groenestijn,D.J.Verschuur,
2009a.Estimating primaries by sparse inversion and application to near-offset
data reconstruction.Geophysics.74,A23-A28.
G.J.A.Van Groenestijn,D.J.Verschuur,2009b.Estimation of primaries and
near-offset reconstruction by sparse inversion:Marine data
applications.Geophysics.74,R119-R128.
D.J.Verschuur,1992.Adaptive surface-related multiple
elimination.Geophysics.57,1166-1177.
D.J.Verschuur,A.J.Berkhout,1997.Estimation of multiple scattering by
iterative inversion,part II:practical aspects and examples.Geophysics.62,
1596-1611.
D.J.Verschuur,A.J.Berkhout,P.G.Kelmis,1995.Estimation of multiple
scattering by iterative inversion,Part II: Examples on marine and land
data.Seg Technical Program Expanded Abstracts.14,1470-1473.
Y.H.Wang,2004.Multiple prediction through inversion:Afully data-
driven concept for surface-related multiple attenuation.Geophysics.69,547-
553.
Y.H.Wang,2007.Multiple prediction through inversion:Theoretical
advancements and real data application. Geophysics.72,V33-V39.
A.B.Weglein,F.A.Gasparotto,P.M.Carvalho,1997.An inverse-scattering
series method for attenuating multiples in seismic reflection
data.Geophysics.62,1975-1989.
O.Yilmaz,M.T.Taner,1994.Discrete plane-wave decomposition by least-
mean-square-error method. Geophysics.59,973-982.
Zeng Zhongyu, the compressional wave multi -components NMO NMO velocities research of 2013. complex dielectrics three-dimensional trace gather models, China ground
Zhen Ju Geophysics Institutes consolidates to rich, Han Liguo, Li Hongjian, 2014. anisotropy Radon conversion and its in multiple wave pressure
Application Chinese Journal of Geophysics .57,2928-2936. in system
Gong Xiangbo, Han Liguo, Wang Shengchao, 2016. hybrid domain high-resolution Hyperbola Radon Transform and its in multiple wave pressure system
In application petroleum exploration 55,711-718.
Jia Chunmei, Jiang Guoqing, Liu Zhicheng, Xu Lu, 2016. frequency-domain sparse Hyperbola Radon Transform denoising method physical prospectings are with changing
Visit .40,527-533.
Liu Jun, 2008. are based on the maximized predictive deconvolution technology of non-Gaussian system, Tsing-Hua University
Shen is grasped, the 2003. Surface-related multiple compactings based on wave equation, The Chinese Geology Univ. (Beijing)
Shi Ying, Liu Hong, go through beautiful English, the multiple ripple drawing method Jilin University journal of 2011. reverse data field surfaces:Earth section
Learn version .41,271-276.
Song Jiawen, D.J.Verschuur, Chen little Hong, SONGJia-wen, D.J.Verschuur, CHENXiao-hong,
The present Research of more than 2014. wave pressure and progress Advances in Geophysics .29,240-247.
Wang Zhaoqi, 2007. predictive deconvolutions and its Analysis on Issues Related, Chang An University
The multi-direction orthogonal polynomial transformation multiple suppression geophysics of Xue Yaru, Chen little Hong, Ma Jitao, 2012.
Report .55,3450-3458.
Zhao Chang builds, Ye Yueming, Yao Genshun, Hu Bing, Zhuan Xijin, and 2013. linear La Dong domains predictive deconvolutions are multiple in ocean
Application Advances in Geophysics .28,1026-1032. in ripple removal
Claims (5)
1. the integrated forecasting deconvolution of hyperbolic Radon domains and feedback cycle method multiple suppression model building method, it is characterised in that
Comprise the following steps：
(1) original shallow seismic profile data are read in, predict more subwaves with predictive deconvolution and feedback loop methodology respectively first
Composition；
(2) Hyperbola Radon Transform is carried out to the multiple wave component of both the above method prediction；
(3) for still having long-period multiple in predictive deconvolution prediction error, sef-adapting filter is built, is filtered out once
Significant wave, and then obtain with the multiple wave energy model in hyperbolic Radon domains after two methods；
(4) as sef-adapting filter output valve FmWhen (τ, h) is approximately 0, it is believed that effective reflection also be present, then need return to change again
Generation processing, until Fm(τ, h) is approximately equal to 1, terminates iterative processing；
(5) anti-Radon conversion is carried out to multiple wave energy model in hyperbolic Radon domains；
(6) multiple wave pattern data are subtracted with original shallow seismic profile data, finally obtains the high reliability after multiple wave pressure system
Shallow seismic profile data.
2. hyperbolic Radon domains integrated forecasting deconvolution as claimed in claim 1 and feedback cycle method multiple suppression model structure
Construction method, it is characterised in that the mathematical modeling of predictive deconvolution is：
<mrow>
<mtable>
<mtr>
<mtd>
<mrow>
<mi>d</mi>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>+</mo>
<mi>l</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>&tau;</mi>
<mo>=</mo>
<mn>0</mn>
</mrow>
<mi>&infin;</mi>
</munderover>
<mi>b</mi>
<mrow>
<mo>(</mo>
<mi>&tau;</mi>
<mo>)</mo>
</mrow>
<mi>&xi;</mi>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>+</mo>
<mi>l</mi>
<mo>-</mo>
<mi>&tau;</mi>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>&tau;</mi>
<mo>=</mo>
<mn>0</mn>
</mrow>
<mrow>
<mi>l</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</munderover>
<mi>b</mi>
<mrow>
<mo>(</mo>
<mi>&tau;</mi>
<mo>)</mo>
</mrow>
<mi>&xi;</mi>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>+</mo>
<mi>l</mi>
<mo>-</mo>
<mi>&tau;</mi>
<mo>)</mo>
</mrow>
<mo>+</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>&tau;</mi>
<mo>=</mo>
<mi>l</mi>
</mrow>
<mi>&infin;</mi>
</munderover>
<mi>b</mi>
<mrow>
<mo>(</mo>
<mi>&tau;</mi>
<mo>)</mo>
</mrow>
<mi>&xi;</mi>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>+</mo>
<mi>l</mi>
<mo>-</mo>
<mi>&tau;</mi>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
</mtable>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</mrow>
3. hyperbolic Radon domains integrated forecasting deconvolution as claimed in claim 1 and feedback cycle method multiple suppression model structure
Construction method, it is characterised in that the mathematical modeling of feedback cycle is：
<mrow>
<msubsup>
<mi>D</mi>
<mn>0</mn>
<mrow>
<mo>(</mo>
<mi>n</mi>
<mo>+</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</msubsup>
<mrow>
<mo>(</mo>
<msub>
<mi>z</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mi>D</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>z</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
</mrow>
<mo>-</mo>
<msubsup>
<mi>D</mi>
<mn>0</mn>
<mrow>
<mo>(</mo>
<mi>n</mi>
<mo>)</mo>
</mrow>
</msubsup>
<mrow>
<mo>(</mo>
<msub>
<mi>z</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
</mrow>
<msup>
<mi>A</mi>
<mrow>
<mo>(</mo>
<mi>n</mi>
<mo>+</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</msup>
<mrow>
<mo>(</mo>
<msub>
<mi>z</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
</mrow>
<mi>D</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>z</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>2</mn>
<mo>)</mo>
</mrow>
</mrow>
4. hyperbolic Radon domains integrated forecasting deconvolution as claimed in claim 1 and feedback cycle method multiple suppression model structure
Construction method, it is characterised in that the mathematical modeling of Hyperbola Radon Transform is：
<mrow>
<mi>m</mi>
<mrow>
<mo>(</mo>
<mi>&tau;</mi>
<mo>,</mo>
<mi>h</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munder>
<mo>&Sigma;</mo>
<mi>x</mi>
</munder>
<mi>d</mi>
<mo>&lsqb;</mo>
<mi>x</mi>
<mo>,</mo>
<mi>t</mi>
<mo>=</mo>
<mi>&tau;</mi>
<mo>+</mo>
<mi>h</mi>
<mrow>
<mo>(</mo>
<msqrt>
<mrow>
<msup>
<mi>x</mi>
<mn>2</mn>
</msup>
<mo>+</mo>
<msubsup>
<mi>z</mi>
<mrow>
<mi>r</mi>
<mi>e</mi>
<mi>f</mi>
</mrow>
<mn>2</mn>
</msubsup>
</mrow>
</msqrt>
<mo>-</mo>
<msub>
<mi>z</mi>
<mrow>
<mi>r</mi>
<mi>e</mi>
<mi>f</mi>
</mrow>
</msub>
<mo>)</mo>
</mrow>
<mo>&rsqb;</mo>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>3</mn>
<mo>)</mo>
</mrow>
</mrow>
5. hyperbolic Radon domains integrated forecasting deconvolution as claimed in claim 1 and feedback cycle method multiple suppression model structure
Construction method, it is characterised in that the mathematical modeling of sef-adapting filter is：
<mrow>
<msub>
<mi>F</mi>
<mi>m</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>&tau;</mi>
<mo>,</mo>
<mi>h</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfrac>
<mn>1</mn>
<msqrt>
<mrow>
<mn>1</mn>
<mo>+</mo>
<msup>
<mrow>
<mo>(</mo>
<mfrac>
<mi>&eta;</mi>
<mrow>
<mi>M</mi>
<mrow>
<mo>(</mo>
<mi>&tau;</mi>
<mo>,</mo>
<mi>h</mi>
<mo>)</mo>
</mrow>
<mo>/</mo>
<mi>D</mi>
<mrow>
<mo>(</mo>
<mi>&tau;</mi>
<mo>,</mo>
<mi>h</mi>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<mo>)</mo>
</mrow>
<mi>m</mi>
</msup>
</mrow>
</msqrt>
</mfrac>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>4</mn>
<mo>)</mo>
</mrow>
</mrow>
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710843785.2A CN107678062B (en) | 2017-09-15 | 2017-09-15 | The integrated forecasting deconvolution of the domain hyperbolic Radon and feedback loop methodology multiple suppression model building method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710843785.2A CN107678062B (en) | 2017-09-15 | 2017-09-15 | The integrated forecasting deconvolution of the domain hyperbolic Radon and feedback loop methodology multiple suppression model building method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107678062A true CN107678062A (en) | 2018-02-09 |
| CN107678062B CN107678062B (en) | 2019-05-07 |
Family
ID=61136547
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710843785.2A Expired - Fee Related CN107678062B (en) | 2017-09-15 | 2017-09-15 | The integrated forecasting deconvolution of the domain hyperbolic Radon and feedback loop methodology multiple suppression model building method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107678062B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109001813A (en) * | 2018-07-16 | 2018-12-14 | 中国石油天然气股份有限公司 | A kind of method, apparatus and system of multiple suppression |
| CN109471162A (en) * | 2018-10-08 | 2019-03-15 | 中国石油天然气集团有限公司 | Interbed multiple processing method, system, electronic equipment and readable medium |
| CN111239827A (en) * | 2020-03-09 | 2020-06-05 | 吉林大学 | Three-dimensional seismic data multiple suppression method based on local similarity coefficient |
| CN111239828A (en) * | 2020-03-09 | 2020-06-05 | 吉林大学 | Multiple suppression method based on optimal hyperbolic integral path superposition |
| CN111505719A (en) * | 2020-05-27 | 2020-08-07 | 中海石油(中国)有限公司湛江分公司 | Diffracted multiple suppression method based on wave field decomposition |
| CN112147693A (en) * | 2020-08-18 | 2020-12-29 | 中国石油天然气股份有限公司 | Method and device for suppressing residual multiples with near offset distance |
| CN113189646A (en) * | 2021-04-19 | 2021-07-30 | 滨州学院 | Method for removing dragging type shallow-section stratum abnormal fluctuation |
| CN117148443A (en) * | 2023-10-27 | 2023-12-01 | 胜利信科(山东)勘察测绘有限公司 | Shallow profile data signal-to-noise ratio enhancement method based on ghost wave extraction and conversion |
| CN117741750A (en) * | 2024-02-21 | 2024-03-22 | 东北石油大学三亚海洋油气研究院 | Multi-channel pre-stack deconvolution method and system based on Radon transformation |
| CN117741750B (en) * | 2024-02-21 | 2024-04-26 | 东北石油大学三亚海洋油气研究院 | Multi-channel pre-stack deconvolution method and system based on Radon transformation |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120213032A1 (en) * | 2011-02-22 | 2012-08-23 | Cggveritas Services Sa | Method for pz summation of 3-dimensional wide azimuth receiver gathers and device |
| CN103376466A (en) * | 2012-04-13 | 2013-10-30 | 中国石油化工股份有限公司 | Multiple suppression method |
| CN105510976A (en) * | 2016-01-11 | 2016-04-20 | 中国海洋大学 | Multiple-wave combined self-adaptive attenuation method |
-
2017
- 2017-09-15 CN CN201710843785.2A patent/CN107678062B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120213032A1 (en) * | 2011-02-22 | 2012-08-23 | Cggveritas Services Sa | Method for pz summation of 3-dimensional wide azimuth receiver gathers and device |
| CN103376466A (en) * | 2012-04-13 | 2013-10-30 | 中国石油化工股份有限公司 | Multiple suppression method |
| CN105510976A (en) * | 2016-01-11 | 2016-04-20 | 中国海洋大学 | Multiple-wave combined self-adaptive attenuation method |
Non-Patent Citations (1)
| Title |
|---|
| 巩向博 等: "各向异性Radon变换及其在多次波压制中的应用", 《地球物理学报》 * |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109001813A (en) * | 2018-07-16 | 2018-12-14 | 中国石油天然气股份有限公司 | A kind of method, apparatus and system of multiple suppression |
| CN109001813B (en) * | 2018-07-16 | 2020-01-07 | 中国石油天然气股份有限公司 | Method, device and system for suppressing multiple waves |
| CN109471162A (en) * | 2018-10-08 | 2019-03-15 | 中国石油天然气集团有限公司 | Interbed multiple processing method, system, electronic equipment and readable medium |
| CN111239828B (en) * | 2020-03-09 | 2021-07-30 | 吉林大学 | Multiple suppression method based on optimal hyperbolic integral path superposition |
| CN111239828A (en) * | 2020-03-09 | 2020-06-05 | 吉林大学 | Multiple suppression method based on optimal hyperbolic integral path superposition |
| CN111239827B (en) * | 2020-03-09 | 2021-07-30 | 吉林大学 | Three-dimensional seismic data multiple suppression method based on local similarity coefficient |
| CN111239827A (en) * | 2020-03-09 | 2020-06-05 | 吉林大学 | Three-dimensional seismic data multiple suppression method based on local similarity coefficient |
| CN111505719A (en) * | 2020-05-27 | 2020-08-07 | 中海石油(中国)有限公司湛江分公司 | Diffracted multiple suppression method based on wave field decomposition |
| CN111505719B (en) * | 2020-05-27 | 2022-05-27 | 中海石油(中国)有限公司湛江分公司 | Diffracted multiple suppression method based on wave field decomposition |
| CN112147693A (en) * | 2020-08-18 | 2020-12-29 | 中国石油天然气股份有限公司 | Method and device for suppressing residual multiples with near offset distance |
| CN113189646A (en) * | 2021-04-19 | 2021-07-30 | 滨州学院 | Method for removing dragging type shallow-section stratum abnormal fluctuation |
| CN113189646B (en) * | 2021-04-19 | 2022-08-30 | 滨州学院 | Method for removing dragging type shallow-section stratum abnormal fluctuation |
| CN117148443A (en) * | 2023-10-27 | 2023-12-01 | 胜利信科(山东)勘察测绘有限公司 | Shallow profile data signal-to-noise ratio enhancement method based on ghost wave extraction and conversion |
| CN117148443B (en) * | 2023-10-27 | 2024-03-19 | 胜利信科(山东)勘察测绘有限公司 | Shallow profile data signal-to-noise ratio enhancement method based on ghost wave extraction and conversion |
| CN117741750A (en) * | 2024-02-21 | 2024-03-22 | 东北石油大学三亚海洋油气研究院 | Multi-channel pre-stack deconvolution method and system based on Radon transformation |
| CN117741750B (en) * | 2024-02-21 | 2024-04-26 | 东北石油大学三亚海洋油气研究院 | Multi-channel pre-stack deconvolution method and system based on Radon transformation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107678062B (en) | 2019-05-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107678062B (en) | The integrated forecasting deconvolution of the domain hyperbolic Radon and feedback loop methodology multiple suppression model building method | |
| CN108181652B (en) | A kind of subsea node seismic data uplink and downlink wave field numerical method | |
| CN103713315B (en) | A kind of seismic anisotropy parameter full waveform inversion method and device | |
| CN103238158B (en) | Utilize the marine streamer data source inverting simultaneously that mutually related objects function is carried out | |
| CN108549100B (en) | The multiple dimensioned full waveform inversion method of time-domain for opening up frequency based on non-linear high order | |
| CN102590859B (en) | Anisotropic reverse time migration method for quasi-P wave equation in transverse isotropy with a vertical axis of symmetry (VTI) medium | |
| US20100054082A1 (en) | Reverse-time depth migration with reduced memory requirements | |
| CN102169189B (en) | Method for eliminating deep water interlamination multiple | |
| CN109669212B (en) | Seismic data processing method, stratum quality factor estimation method and device | |
| CN104122585A (en) | Seismic forward modeling method based on elastic wave field vector decomposition and low-rank decomposition | |
| CN107505654A (en) | Full waveform inversion method based on earthquake record integration | |
| CN103926622A (en) | Method for suppressing multiple waves based on L1 norm multichannel matched filtering | |
| CN103926619A (en) | Reverse time migration method of three-dimensional VSP data | |
| CN107884829A (en) | A kind of method for combining compacting shallow sea OBC Multiple Attenuation in Seismic Data | |
| CN104237945A (en) | Seismic data self-adaptive high-resolution processing method | |
| CN106646603B (en) | A kind of reliability judgment method of practical time-lapse seismic data processing difference | |
| Bashir et al. | Influence of seismic diffraction for high-resolution imaging: applications in offshore Malaysia | |
| CN111045077B (en) | Full waveform inversion method of land seismic data | |
| CN106199694A (en) | Synthetic record method based on deep varitron ripple | |
| CN104330826A (en) | A method for removing various noises under the condition of complex surface | |
| CN106324669A (en) | Method of separating surface-related multiples of different orders in seismic exploration data | |
| CN105182414B (en) | A kind of method that direct wave is removed based on Wave equation forward modeling | |
| CN102590858B (en) | Two-way wave imaging method based on broadband wavelet reconstruction | |
| CN113885079A (en) | Elastic wave field decoupling-based high-precision multi-azimuth reverse-time seismic source imaging method | |
| Bashir et al. | Preservation of seismic diffraction to enhance the resolution of seismic data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190507 Termination date: 20210915 |