CN107450102A - Multiple dimensioned full waveform inversion method based on the controllable envelope generating operator of resolution ratio - Google Patents
Multiple dimensioned full waveform inversion method based on the controllable envelope generating operator of resolution ratio Download PDFInfo
- Publication number
- CN107450102A CN107450102A CN201710633546.4A CN201710633546A CN107450102A CN 107450102 A CN107450102 A CN 107450102A CN 201710633546 A CN201710633546 A CN 201710633546A CN 107450102 A CN107450102 A CN 107450102A
- Authority
- CN
- China
- Prior art keywords
- model
- resolution ratio
- geological data
- data
- envelope generating
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 61
- 238000002945 steepest descent method Methods 0.000 claims abstract description 11
- 238000001615 p wave Methods 0.000 claims description 16
- 238000005070 sampling Methods 0.000 claims description 11
- 238000005516 engineering process Methods 0.000 claims description 8
- 238000004458 analytical method Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 239000004576 sand Substances 0.000 claims description 3
- 238000007711 solidification Methods 0.000 claims description 2
- 230000008023 solidification Effects 0.000 claims description 2
- 238000004422 calculation algorithm Methods 0.000 abstract description 5
- 230000000694 effects Effects 0.000 abstract description 3
- 238000004364 calculation method Methods 0.000 abstract description 2
- 238000002474 experimental method Methods 0.000 description 7
- 239000007789 gas Substances 0.000 description 5
- 238000001228 spectrum Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 238000012937 correction Methods 0.000 description 2
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 description 2
- 239000011435 rock Substances 0.000 description 2
- 239000004215 Carbon black (E152) Substances 0.000 description 1
- 241001269238 Data Species 0.000 description 1
- 241000083513 Punctum Species 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 229930195733 hydrocarbon Natural products 0.000 description 1
- 150000002430 hydrocarbons Chemical class 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 239000003345 natural gas Substances 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
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. for interpretation or for event detection
- G01V1/30—Analysis
- G01V1/306—Analysis for determining physical properties of the subsurface, e.g. impedance, porosity or attenuation profiles
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V2210/00—Details of seismic processing or analysis
- G01V2210/60—Analysis
- G01V2210/62—Physical property of subsurface
- G01V2210/622—Velocity, density or impedance
- G01V2210/6222—Velocity; travel time
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V2210/00—Details of seismic processing or analysis
- G01V2210/60—Analysis
- G01V2210/66—Subsurface modeling
- G01V2210/667—Determining confidence or uncertainty in parameters
Landscapes
- Engineering & Computer Science (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Acoustics & Sound (AREA)
- Environmental & Geological Engineering (AREA)
- Geology (AREA)
- General Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Geophysics (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention discloses a kind of multiple dimensioned full waveform inversion method based on the controllable envelope generating operator of resolution ratio：First, the controllable envelope generating operator of resolution ratio is established, frequency band range of the inverting using geological data is controlled by adjusting parameter；Secondly, based on the controllable envelope generating operator structure full waveform inversion method of resolution ratio, the gradient of derived object function pair model parameter and the calculation formula of iteration step length；Finally, renewal is iterated to model parameter using steepest descent method, untill meeting algorithm termination condition.The present invention has the stronger ability for overcoming cycle-skipping problem, is more suitable for the modeling of lithologic deposit rate pattern.Particularly when low-frequency information shortage in geological data or too high dominant frequency, advantage becomes apparent.No matter whether the present invention have sufficient low-frequency information in geological data, and the geological data after the controllable envelope generating operator effect of resolution ratio has sufficient low-frequency information；The Parameter adjustable of the present invention, it is possible to achieve Multi-scale inversion.
Description
Technical field
The invention belongs to technical field of geophysical exploration, is related to a kind of earth model p wave interval velocity parameter inversion method, special
It is not related to a kind of multiple dimensioned full waveform inversion method based on the controllable envelope generating operator of resolution ratio.
Background technology
Oil and natural gas is the strategic resource concerning national economic development and national security, is pushed further into China's oil gas
The exploration and development of resource belongs to national Major Strategic.Lithologic deposit is most important exploration in China's long duration from now on
Development object.Method of seismic prospecting is the main method of oil-gas exploration, and speed parameter is the important parameter of oil and gas detection.China's rock
Property oil-gas reservoir longitudinally there is strong anisotropism in Thin interbeds texture, in transverse direction in geology mostly, how to lithologic deposit
It is the key issue that China's oil-gas exploration faces that speed parameter, which carries out fine description,.Full waveform inversion be it is a kind of directly from prestack
Shake data are set out, and the method for accurately descending medium velocity model is obtained using prestack big gun collection geological data full detail.Cause
This, full waveform inversion is to portray the effective ways of lithologic deposit speed parameter.
Traditional full waveform inversion is a kind of to obtain Data Matching of the underground medium high-resolution parameter model as target
Method.Full waveform inversion weighs observation data with calculating the object function of Data Matching degree to obtain ground by minimizing one
Lower medium parameter.Earliest by full waveform inversion method push to practicality be Lailly and Tarantola work.They propose can
To carry out cross-correlation using the main story wave field by epicenter excitation and the anti-pass wave field excited by data residual error to build target letter
Several gradients to model parameter.This error-duration model technology, which overcomes explicit Frechet derivatives of asking for, needs a large amount of wave field forward modelings
This defect.Traditional full waveform inversion uses L2Norm observes the matching degree of data and calculating data to weigh, and it is applied to
Reflected seismic information faces many problems when building high-resolution rate pattern, such as：Full waveform inversion is required in observation data
Low-frequency information or initial model comprising abundance accurately ensure that algorithm can overcome cycle skipping problem and then avoid enough
It is absorbed in local extremum.Therefore, reduce full waveform inversion just turns into the one of full waveform inversion field to the degree of dependence of initial model
Individual study hotspot.
Bunks proposes multiple dimensioned full waveform inversion method in time-domain, and it is anti-to propose frequency domain Full wave shape by Pratt afterwards
Drill method.Both approaches carry out inverting in refutation process first by low-frequency data.The target formed due to low-frequency data
Function is more linear compared to high-frequency data, and the convergence domain of global minimum is bigger, so now algorithm is wanted to initial model
Asking will be lower.Inversion result based on low-frequency data can be as the initial model of the full waveform inversion based on high-frequency data.
By being gradually added into high-frequency data in inverting, high-precision rate pattern may finally be obtained.Although multi-scale method has one
A little successful application cases, but this kind of method can not fundamentally overcome cycle-skipping problem.When this kind of method is applied to reality
Following problems may be encountered during data：It is highly difficult to typically set up a sufficiently accurate initial model；The low frequency of real data
Information is often by noise effect；The offset distance of real data is often limited.
Many scholar's research, which are established for full waveform inversion, compares L2This problem of the more excellent measurement criterion of norm.Their mesh
Mark is to establish that convexity is more preferable, and robustness is stronger and can overcome the object function of cycle-skipping problem very well.Carried out with conventional method
Waveform Matching is different, the matching of information when this method is emphasized to travel to wave field.The representative work of this thinking is exactly
The wave equation Travel Time Inversion that Luo and Schuster is proposed.With traditional Travel Time Inversion method dependent on artificial pickup wave field
Information is different during travelling, and this method obtains forward modeling data using wave equation, can be considered as a kind of and use mutually related objects
The full waveform inversion method of function.However, because rate pattern caused by wave equation Travel Time Inversion method is excessively smooth, its
It cannot be guaranteed that can be that full waveform inversion establishes sufficiently accurate initial model.Chi etc. and Luo etc. proposes anti-in Full wave shape
The middle instantaneous amplitude (envelope) using wave field is drilled to build the object function of inverting.Wu etc. demonstrates the vibration of geological data envelope
The information of ultralow frequency is carried with decay, this information can be used to estimate the rate pattern of long wavelength.But this target
Function still can be perplexed by cycle skipping problem.
The content of the invention
The shortcomings that it is an object of the invention to overcome above-mentioned prior art, there is provided one kind is based on the controllable envelope generation of resolution ratio
The multiple dimensioned full waveform inversion method of operator.
The purpose of the present invention is achieved through the following technical solutions：
This multiple dimensioned full waveform inversion method based on the controllable envelope generating operator of resolution ratio, comprises the following steps：
1) original earthquake data is gathered, then the geological data collected is pre-processed, prestack big gun is obtained after processing
Collect geological data, be designated as dobs(t,xr；xs), wherein t represents time variable；xr,xsThe respectively position of wave detector and focus；
2) Rectangular grid geological model is built, sets spatial spreading interval, grid scale and the forward modeling of discrete grid block
The model time sampling interval；
3) geological data is analyzed, provides the geological data frequency band range that inverting uses；
4) geological model p wave interval velocity v (x) initial model is given, and specifies object function J (v (x)) to be optimized；
5) gradients of the object function J (v (x)) to model parameter v (x) is asked for, is efficiently asked using error-duration model technology
Solution；
6) iteration step length during steepest descent method iteration is asked for, is determined in frequency domain using linear search method；
7) global minimum of object function is obtained by steepest descent method iteration, so obtain the 3) step it is given
In frequency band range, the inversion result of p wave interval velocity；
8) frequency band range of geological data used when adjusting inverting, and the result of step 7) is used as initial model,
Repeat step 3) to all effective band scopes 7), have been circulated, export optimal p wave interval velocity model.
Further, above-mentioned steps 2) in, spatial spreading interval, grid scale and the forward model time of discrete grid block adopt
Sample interval was determined according to the geophone offset of actual seismic data, effective band scope and sampling time；Selected mesh parameter
When standard is so that based on grid progress finite-difference forward modeling, not only meets stability condition but also effectively suppress numerical value
Frequency dispersion.
Further, above-mentioned steps 3) in, analysis first receives the effective band scope of geological data, first makes initial stage in inverting
With the geological data of rather low-frequency, with the progress of inverting, gradually expand the frequency band range that inverting uses.
Further, above-mentioned steps 4) in, geological model p wave interval velocity v (x) initial model is to true p wave interval velocity model
A rough estimate, initial model is set as laterally homogeneous, the increased rate pattern of vertical linear；The object function is to carve
The standard of optimal solution is drawn, to calculate geological data and the fitting degree observed between geological data or error size as standard.
Further, observation big gun collection geological data and point calculated between big gun collection geological data based on model are weighed in setting
2 norms of the controllable envelope data residual error of resolution are object function, i.e. object function is：
In formula, T is the time span of forward modeling；G () is the controllable envelope generating operator of resolution ratio；Ss,rCome for a Restricted operator
Ensure only have focus and wave detector to be located at x respectivelysAnd xrWavefield data be used to build object function；ucal(t, x；xs) it is shake
Source is located at xsCalculating wave field；The smaller explanation rate pattern v (x) of target function value J (v (x)) are more accurate.
Further, above-mentioned steps 5) in, object function J (v (x)) is to use error-duration model to model parameter v (x) gradient
Technology carries out Efficient Solution.
Further, above-mentioned steps 6) in, iteration step length is to be determined in frequency domain using linear search method.
Further, above-mentioned steps 8) in, it is next after the geological data in using a certain frequency band range is iterated inverting
The information that the geological data frequency band range that wheel inverting uses than last round of iteration time width, will make more high frequencies be used to carry out instead
Drill.
The invention has the advantages that：
The step of method proposed by the invention, may be summarized to be following three step：First, the present invention establishes a kind of resolution ratio
Controllable envelope generating operator, the operator can control frequency band range of the inverting using geological data by adjusting parameter；Secondly, originally
Invention is based on the controllable envelope generating operator of resolution ratio, constructs full waveform inversion method, has derived its object function and model is joined
Several gradients and the calculation formula of iteration step length；Finally, renewal is iterated to model parameter using steepest descent method, until
Untill meeting algorithm termination condition.
Compared with conventional full waveform inversion method, this method has the stronger ability for overcoming cycle-skipping problem, is more suitable for
Lithologic deposit rate pattern models.Particularly when in geological data low-frequency information lack or dominant frequency it is too high when, this method it is excellent
Gesture becomes apparent.There is method proposed by the invention the reason for above-mentioned advantage to be：First, no matter whether have in geological data
Sufficient low-frequency information, the geological data after the controllable envelope generating operator effect of resolution ratio have sufficient low-frequency information；
Secondly, the Parameter adjustable of the controllable envelope generating operator of resolution ratio, it is possible to achieve Multi-scale inversion.
Brief description of the drawings
Fig. 1 is schematic flow sheet of the present invention；
Fig. 2 gives the amplitude spectrum of three groups of different parameters time shift linear operators；
Fig. 3 is the result of Marmousi model examples：(a) true velocity model；(b) initial velocity model；(c)0-8Hz
Data inversion rate pattern；(d) 0-12Hz data inversions rate pattern；(e) 0-16Hz data inversions rate pattern；(f)0-
20Hz data multiply rate pattern.
Embodiment
The present invention is described in further detail below in conjunction with the accompanying drawings：
High-precision rate pattern be it is a kind of can be applied to reservoir portray and hydrocarbon indication with rock property phase
The attribute of pass, full waveform inversion are a kind of to obtain the anti-of underground medium high precision velocity model based on prestack big gun collection geological data
Artistic skills art.The present invention proposes a kind of multiple dimensioned full waveform inversion method based on the controllable envelope generating operator of resolution ratio.The party
Method directly from geological data, constructs non-linear by using the information and nonlinear operator of geological data different frequency bands
Multiple dimensioned full waveform inversion method, substantially increase the ability for overcoming cycle skipping problem and fine rate pattern.
The material base of the present invention is a large amount of geological datas collected by field high-resolution seismic exploration collecting device.This
The multiple dimensioned full waveform inversion method framework based on the controllable envelope generating operator of resolution ratio proposed is invented as shown in figure 1, tool
Body step difference is as follows：
1) original earthquake data is gathered, conventional pretreatment then is carried out to the geological data collected, including：Static correction,
Obtain prestack big gun collection geological data after the processing such as dynamic correction, noise compacting, removal face ripple (also needs to remove when handling marine data
Terrible ripple operation), it is designated as dobs(t,xr；xs), wherein t represents time variable；xr,xsThe respectively position of wave detector and focus；
2) Rectangular grid geological model is built, sets spatial spreading interval, grid scale and the forward modeling of discrete grid block
The simulated time sampling interval；Spatial spreading interval, grid scale and the forward model time sampling interval of discrete grid block are roots
Factually the geophone offset of border geological data, effective band scope and sampling time determine；The standard of selected mesh parameter is so that
When carrying out finite-difference forward modeling based on the grid, not only meet stability condition but also effectively suppress numerical solidification.
By taking two dimensional inversion as an example, discrete grid block it needs to be determined that parameter have：Transverse grid interval dx, longitudinal grid interval
Dz, transverse grid number Nx and longitudinal grid number Nz, in addition also need to determine forward simulation time sampling interval dt.In practice
Can be according to geological data effective band scope, earthquake record time span, geological data observation system and Courant-
Friedrichs-Lewy (CFL) conditions determine relevant parameter；
3) analysis receives the effective band scope of geological data first, and the earthquake number of rather low-frequency is first used initial stage in inverting
According to, with the progress of inverting, the frequency band range that gradually expansion inverting uses.
4) geological model p wave interval velocity v (x) initial model is given, and specifies object function J (v (x)) to be optimized：Ground
Matter model p wave interval velocity v (x) initial model is a rough estimate to true p wave interval velocity model, when typically can be by walking
Inverting is obtained, or initial model is set as into laterally homogeneous, the increased rate pattern of vertical linear；The object function is to carve
The standard of optimal solution is drawn, to calculate geological data and the fitting degree observed between geological data or error size as standard.This
Observation big gun collection geological data and the controllable bag of resolution ratio calculated between big gun collection geological data based on model are weighed in setting in invention
2 norms of network residual error are object function, i.e. object function is：
In formula, T is the time span of forward modeling；Ss,rEnsure that only focus and wave detector are located at respectively for a Restricted operator
xsAnd xrWavefield data be used to build object function；ucal(t,x；xs) it is that focus is located at xsCalculating wave field；G () is point
The controllable envelope generating operator of resolution.The smaller explanation rate pattern v (x) of target function value J (v (x)) are more accurate.Calculated in above formula
Wave field ucal(t, x；xs) meet following Chang Midu ACOUSTIC WAVE EQUATIONs：
In formula, v (x) is the P-wave angle value at the x of locus；Δ is laplacian；F (t) is source function.Formula
(3) the controllable envelope generating operator G () of resolution ratio used in is defined as follows：
In formula, " * " represents time-domain convolution algorithm；H () represents Hilbert transform；D () represents time shift and linearly become
Change；D (t) is a certain linear function, an,Nξ,ξnFor d (t) three parameters.D (t) amplitude spectrum in formula (5)Join with it
Several relations is as follows：
In formula, ωsFor sample frequency.When being illustrated in figure 2 given three groups of different parameters, the form of d (t) amplitude spectrums.
The amplitude spectrum that Fig. 2 demonstrates d (t) can be adjusted by adjusting its parameter.
5) gradient of the object function to model parameter is asked for, error-duration model technology can be used to carry out Efficient Solution.Target
Function J (v (x)) is to carry out Efficient Solution using error-duration model technology to model parameter v (x) gradient；The pass of the solution procedure
Key is the focus item for determining error-duration model wave field.
Use us(t) u is representedcal(t,x；xs), ds,r(t) d is representedobs(t,xr；xs) and fs(t) f (t) δ (x- are represented
xs).In view of primary condition, due to focus fs(t) obtained pressure field u is exciteds(t) can be expressed as：
In formula, m is model parameter, i.e. v (x), using Adjoint State Method, can obtain following adjoint state equation：
In formula, y (t) is the focus item of error-duration model.Finally the expression formula of gradient is：
In formula, m is model parameter；qsMeet the wave equation as defined in formula (8).
6) step-length during steepest descent method iteration is asked for, is determined in frequency domain using linear search method.Iteration step length is
It is determined in frequency domain using linear search method；Only used in inverting the reason for frequency domain asks for iteration step length
The information of geological data partial-band, the step-length used during only with this partial information to determine iteration are only reasonably.In order to
Simplicity of exposition, following mark agreement is done,Expression model parameter is m (x), focus fs(t) frequency of the wave field excited
Rate domain expression formula；For dobs(t,xr；xs) frequency domain expression formula.So step-length α of the n-th step iterationnCan be according to such as
Lower formula is asked for：
In formula, mn(x) it is the model parameter after the n-th step iteration；dn(x) it is the negative gradient direction of the (n+1)th step；ε be one compared with
Small number, it is necessary to assure：
In formula, max () is to take max function.
7) global minimum of object function is obtained by steepest descent method iteration, and then obtains what is given in step 3) step
In frequency band range, the inversion result of p wave interval velocity.The iterative formula of wherein steepest descent method is as follows：
mn+1(x)=mn(x)+αndn(x) (12)
In formula, mn(x),mn+1(x) it is respectively that the model parameter of the n-th step and the (n+1)th step iteration (is herein rate pattern v
(x))；αnFor iteration step length；dn(x) it is iteration direction, in steepest descent method, iteration direction is negative gradient direction.
By above-mentioned steepest descent method iteration, optimal P wave velocity moulds in current data frequency band range can be obtained
Type.
8) frequency band range of geological data used when adjusting inverting, and the result of step 7) is used as initial model,
Repeat step 3) to all effective band scopes 7), have been circulated, export optimal p wave interval velocity model.When a certain frequency band model of use
After geological data in enclosing is iterated inverting, when the geological data frequency band range that next round inverting uses will be than last round of iteration
Width, the information of more high frequencies is set to be used to carry out inverting.
The mode for the geological data frequency band range that general adjustment inverting uses is gradually to expand inverting with the progress of inverting
The frequency ranges of data used, the geological data of high frequency is gradually added to inverting.
Model example
The specific implementation process of the present invention is applied to Marmousi rate patterns.The model laterally shares 5090 meters, indulges
To sharing 1330 meters, true velocity model such as Fig. 3 (a) of the model is shown.The given one-dimensional rate pattern as shown in Fig. 3 (b)
(increasing linearly to 3500m/s from 1500m/s from shallow to deep) is used as initial velocity model, and the model is true to underground medium
The rough estimate of rate pattern.
48 big guns are uniformly arranged in earth's surface altogether in experiment, big gun spacing is 100 meters, and each big gun is uniformly laid in ground by 510
The wave detector of table is received, and detector interval is 10 meters.The Ricker wavelets that dominant frequency is 20Hz are used in experiment as focus,
The sampling interval laterally and longitudinally is set in experiment as 10 meters, time sampling interval 0.001s, when forward modeling carries out 3001 altogether
Punctum, big gun integrate the record time as 3s.In experiment, it compared for method proposed by the present invention and overcome week more by force with a kind of generally acknowledged having
The full waveform inversion method of ripple skip capability：Envelope inverting.
First, from the initial model shown in Fig. 3 (b), using proposed by the invention non-linear multiple dimensioned in experiment
Full waveform inversion method is estimated underground medium velocity model.4 different frequencies of geological data have been used in experiment altogether
Band scope：0-8Hz, 0-12Hz, 0-16Hz and 0-20Hz, inverting is carried out, obtained inversion result is respectively such as Fig. 3 (c)-(f)
It is shown.Secondly, from the initial model shown in Fig. 3 (b), using envelope inversion method come to underground medium velocity in experiment
Model is estimated that Fig. 3 (g) show the rate pattern obtained using envelope inversion method inverting.
Experimental result shown in Fig. 3, which can be seen that envelope inversion method, can not effectively solve this inversion problem,
No matter its obtained inversion speed model and true velocity model in structure and numerically have obvious difference.With envelope inverting
Method forms a sharp contrast, it is proposed by the invention it is non-linear it is multiple dimensioned can obtain portraying the good of rate pattern, and
And it is portrayed precision and progressively increased with the extension using geological data frequency band.This experimental result confirms that the present invention is carried
Going out method has the ability that overcomes cycle skipping more stronger than envelope inversion method, further demonstrate side proposed by the invention
The superiority of method.
Claims (8)
- A kind of 1. multiple dimensioned full waveform inversion method based on the controllable envelope generating operator of resolution ratio, it is characterised in that including with Lower step：1) original earthquake data is gathered, then the geological data collected is pre-processed, prestack big gun collection is obtained after processing Data are shaken, are designated as dobs(t,xr；xs), wherein t represents time variable；xr,xsThe respectively position of wave detector and focus；2) Rectangular grid geological model is built, sets spatial spreading interval, grid scale and the forward model of discrete grid block Time sampling interval；3) geological data is analyzed, provides the geological data frequency band range that inverting uses；4) geological model p wave interval velocity v (x) initial model is given, and specifies object function J (v (x)) to be optimized；5) gradients of the object function J (v (x)) to model parameter v (x) is asked for, Efficient Solution is carried out using error-duration model technology；6) iteration step length during steepest descent method iteration is asked for, is determined in frequency domain using linear search method；7) global minimum of object function is obtained by steepest descent method iteration, and then obtained in the 3) given frequency band model of step In enclosing, the inversion result of p wave interval velocity；8) frequency band range of the geological data used when adjusting inverting, and use the result of step 7) to be repeated as initial model Step 3) exports optimal p wave interval velocity model to all effective band scopes 7), have been circulated.
- 2. the multiple dimensioned full waveform inversion method according to claim 2 based on the controllable envelope generating operator of resolution ratio, its It is characterised by, in the step 2), spatial spreading interval, grid scale and the forward model time sampling interval of discrete grid block It is to be determined according to the geophone offset of actual seismic data, effective band scope and sampling time；Selecting the standard of mesh parameter is During so that carrying out finite-difference forward modeling based on the grid, not only meet stability condition but also effectively suppress numerical solidification.
- 3. the multiple dimensioned full waveform inversion method according to claim 2 based on the controllable envelope generating operator of resolution ratio, its It is characterised by, in the step 3), analysis first receives the effective band scope of geological data, is first used relatively initial stage in inverting The geological data of low frequency, with the progress of inverting, gradually expand the frequency band range that inverting uses.
- 4. the multiple dimensioned full waveform inversion method according to claim 2 based on the controllable envelope generating operator of resolution ratio, its It is characterised by, in the step 4), geological model p wave interval velocity v (x) initial model is one to true p wave interval velocity model Rough estimate, initial model are set as laterally homogeneous, the increased rate pattern of vertical linear；The object function be portray it is optimal The standard of solution, to calculate geological data and the fitting degree observed between geological data or error size as standard.
- 5. the multiple dimensioned full waveform inversion method according to claim 5 based on the controllable envelope generating operator of resolution ratio, its Be characterised by, setting weigh observation big gun collection geological data and based on model calculate big gun collection geological data between resolution ratio it is controllable 2 norms of envelope data residual error are object function, i.e. object function is：In formula, T is the time span of forward modeling；G () is the controllable envelope generating operator of resolution ratio；Ss,rEnsure for a Restricted operator Only focus and wave detector are located at x respectivelysAnd xrWavefield data be used to build object function；ucal(t,x；xs) it is focus position In xsCalculating wave field；The smaller explanation rate pattern v (x) of target function value J (v (x)) are more accurate.
- 6. the multiple dimensioned full waveform inversion method according to claim 2 based on the controllable envelope generating operator of resolution ratio, its It is characterised by, in the step 5), object function J (v (x)) to model parameter v (x) gradient is entered using error-duration model technology Row Efficient Solution.
- 7. the multiple dimensioned full waveform inversion method according to claim 2 based on the controllable envelope generating operator of resolution ratio, its It is characterised by, in the step 6), iteration step length is to be determined in frequency domain using linear search method.
- 8. the multiple dimensioned full waveform inversion method according to claim 2 based on the controllable envelope generating operator of resolution ratio, its It is characterised by, in the step 8), after the geological data in using a certain frequency band range is iterated inverting, next round inverting The information that the geological data frequency band range used than last round of iteration time width, will make more high frequencies be used to carry out inverting.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710633546.4A CN107450102A (en) | 2017-07-28 | 2017-07-28 | Multiple dimensioned full waveform inversion method based on the controllable envelope generating operator of resolution ratio |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710633546.4A CN107450102A (en) | 2017-07-28 | 2017-07-28 | Multiple dimensioned full waveform inversion method based on the controllable envelope generating operator of resolution ratio |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107450102A true CN107450102A (en) | 2017-12-08 |
Family
ID=60489742
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710633546.4A Pending CN107450102A (en) | 2017-07-28 | 2017-07-28 | Multiple dimensioned full waveform inversion method based on the controllable envelope generating operator of resolution ratio |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107450102A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109459789A (en) * | 2018-12-18 | 2019-03-12 | 吉林大学 | Time-domain full waveform inversion method based on amplitude decaying and linear interpolation |
| CN110007340A (en) * | 2019-02-01 | 2019-07-12 | 西安理工大学 | Salt dome speed density estimation method based on the direct envelope inverting of angle domain |
| CN110095773A (en) * | 2019-06-03 | 2019-08-06 | 中南大学 | The two-parameter inversion method of the multiple dimensioned Full wave shape of Ground Penetrating Radar |
| CN111175822A (en) * | 2020-01-19 | 2020-05-19 | 吉林大学 | Strong scattering medium inversion method for improving direct envelope inversion and disturbance decomposition |
| CN111239819A (en) * | 2020-02-12 | 2020-06-05 | 西安理工大学 | Direct envelope inversion method with polarity based on seismic channel attribute analysis |
| CN111580163A (en) * | 2020-05-28 | 2020-08-25 | 中国科学院地质与地球物理研究所 | Full waveform inversion method and system based on non-monotonic search technology |
| CN113640878A (en) * | 2021-08-12 | 2021-11-12 | 西南石油大学 | Method for constructing azimuth-apparent velocity radar map by using virtual seismic source scanning |
| CN113848253A (en) * | 2021-08-20 | 2021-12-28 | 国网江苏省电力有限公司技能培训中心 | Acoustic emission monitoring method and device for water seepage state of main transformer substrate of simulation transformer substation |
| CN115166822A (en) * | 2022-08-02 | 2022-10-11 | 中国矿业大学(北京) | Reservoir elastic parameter prediction method and device and electronic equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103499835A (en) * | 2013-10-13 | 2014-01-08 | 中国石油集团西北地质研究所 | Method for inverting near-surface velocity model by utilizing preliminary waveforms |
| CN105891888A (en) * | 2016-03-28 | 2016-08-24 | 吉林大学 | Multi-domain frequency-division parallel multi-scale full-waveform inversion method |
| CN106054244A (en) * | 2016-06-16 | 2016-10-26 | 吉林大学 | Low-pass filter multi-scale full waveform inversion method of cut-off time window |
| CN106501852A (en) * | 2016-10-21 | 2017-03-15 | 中国科学院地质与地球物理研究所 | A kind of multiple dimensioned full waveform inversion method of three-dimensional acoustic wave equation arbitrarily-shaped domain and device |
| CN106908835A (en) * | 2017-03-01 | 2017-06-30 | 吉林大学 | Band limit Green's function filters multiple dimensioned full waveform inversion method |
-
2017
- 2017-07-28 CN CN201710633546.4A patent/CN107450102A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103499835A (en) * | 2013-10-13 | 2014-01-08 | 中国石油集团西北地质研究所 | Method for inverting near-surface velocity model by utilizing preliminary waveforms |
| CN105891888A (en) * | 2016-03-28 | 2016-08-24 | 吉林大学 | Multi-domain frequency-division parallel multi-scale full-waveform inversion method |
| CN106054244A (en) * | 2016-06-16 | 2016-10-26 | 吉林大学 | Low-pass filter multi-scale full waveform inversion method of cut-off time window |
| CN106501852A (en) * | 2016-10-21 | 2017-03-15 | 中国科学院地质与地球物理研究所 | A kind of multiple dimensioned full waveform inversion method of three-dimensional acoustic wave equation arbitrarily-shaped domain and device |
| CN106908835A (en) * | 2017-03-01 | 2017-06-30 | 吉林大学 | Band limit Green's function filters multiple dimensioned full waveform inversion method |
Non-Patent Citations (5)
| Title |
|---|
| HUANG CHAO 等: "elastic envelope inversion using multicomponent seismic data with filtered-out low frequencies", 《APPLIED GEOPHYSICS》 * |
| ZHAOQI GAO 等: "A novel full waveform inversion method based on a time-shift nonlinear operator", 《GEOPHYSICAL JOURNAL INTERNATIONAL》 * |
| 王官超 等: "基于包络目标函数的弹性波波形反演", 《石油物探》 * |
| 罗静蕊 等: "地震包络反演对局部极小值的抑制特性", 《地球物理学报》 * |
| 陈永芮: "时间域全波形反演方法研究", 《中国优秀硕士学位论文全文数据库 基础科学辑》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109459789A (en) * | 2018-12-18 | 2019-03-12 | 吉林大学 | Time-domain full waveform inversion method based on amplitude decaying and linear interpolation |
| CN110007340A (en) * | 2019-02-01 | 2019-07-12 | 西安理工大学 | Salt dome speed density estimation method based on the direct envelope inverting of angle domain |
| CN110095773A (en) * | 2019-06-03 | 2019-08-06 | 中南大学 | The two-parameter inversion method of the multiple dimensioned Full wave shape of Ground Penetrating Radar |
| CN110095773B (en) * | 2019-06-03 | 2022-11-01 | 中南大学 | Multi-scale full-waveform two-parameter inversion method for ground penetrating radar |
| CN111175822A (en) * | 2020-01-19 | 2020-05-19 | 吉林大学 | Strong scattering medium inversion method for improving direct envelope inversion and disturbance decomposition |
| CN111239819B (en) * | 2020-02-12 | 2022-05-03 | 西安理工大学 | Direct envelope inversion method with polarity based on seismic channel attribute analysis |
| CN111239819A (en) * | 2020-02-12 | 2020-06-05 | 西安理工大学 | Direct envelope inversion method with polarity based on seismic channel attribute analysis |
| CN111580163B (en) * | 2020-05-28 | 2021-08-10 | 中国科学院地质与地球物理研究所 | Full waveform inversion method and system based on non-monotonic search technology |
| CN111580163A (en) * | 2020-05-28 | 2020-08-25 | 中国科学院地质与地球物理研究所 | Full waveform inversion method and system based on non-monotonic search technology |
| CN113640878A (en) * | 2021-08-12 | 2021-11-12 | 西南石油大学 | Method for constructing azimuth-apparent velocity radar map by using virtual seismic source scanning |
| CN113640878B (en) * | 2021-08-12 | 2024-03-29 | 西南石油大学 | Method for constructing azimuth-apparent velocity radar chart by utilizing virtual seismic source scanning |
| CN113848253A (en) * | 2021-08-20 | 2021-12-28 | 国网江苏省电力有限公司技能培训中心 | Acoustic emission monitoring method and device for water seepage state of main transformer substrate of simulation transformer substation |
| CN113848253B (en) * | 2021-08-20 | 2024-02-23 | 国网江苏省电力有限公司技能培训中心 | Acoustic emission monitoring method and device for base water seepage state of main transformer of simulation substation |
| CN115166822A (en) * | 2022-08-02 | 2022-10-11 | 中国矿业大学(北京) | Reservoir elastic parameter prediction method and device and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107450102A (en) | Multiple dimensioned full waveform inversion method based on the controllable envelope generating operator of resolution ratio | |
| CN108345031B (en) | Full waveform inversion method for elastic medium active source and passive source mixed mining seismic data | |
| CN107462924B (en) | A kind of absolute wave impedance inversion method independent of well-log information | |
| CN101334483B (en) | Method for attenuating rayleigh wave scattered noise in earthquake data-handling | |
| CN107505654B (en) | Full waveform inversion method based on earthquake record integral | |
| Tarabusi et al. | The use of HVSR measurements for investigating buried tectonic structures: the Mirandola anticline, Northern Italy, as a case study | |
| CN104570082B (en) | Extraction method for full waveform inversion gradient operator based on green function characterization | |
| CN113740901B (en) | Land seismic data full-waveform inversion method and device based on complex undulating surface | |
| CN106526674A (en) | Three-dimensional full waveform inversion energy weighted gradient preprocessing method | |
| CN106896409B (en) | A kind of varying depth cable ghost reflection drawing method based on wave equation boundary values inverting | |
| CN107894618B (en) | A kind of full waveform inversion gradient preprocess method based on model smoothing algorithm | |
| CN106842295A (en) | The waveform inversion method of logging information constrained | |
| CN109669212A (en) | Seismic data processing technique, interval quality factors evaluation method and device | |
| CN111722284B (en) | Method for establishing speed depth model based on gather data | |
| CN112327358B (en) | Forward modeling method for acoustic seismic data in viscous medium | |
| CN104898161A (en) | Effective sandstone predicting method based on logging response simulator | |
| CN104237937A (en) | Pre-stack seismic inversion method and system thereof | |
| US10132945B2 (en) | Method for obtaining estimates of a model parameter so as to characterise the evolution of a subsurface volume | |
| JP7254310B2 (en) | Seismic frequency resonance survey method | |
| CN102053269A (en) | Analysis method of speed in seismic data | |
| CN109239776B (en) | Seismic wave propagation forward modeling method and device | |
| CN113945982A (en) | Method and system for removing low frequency and low wave number noise to generate enhanced image | |
| CN105259577B (en) | A kind of method and device for the angle information for determining bed boundary | |
| CN103424774B (en) | A kind of earthquake hydrocarbon-bearing pool formation method and device | |
| Bhaumik et al. | Higher-order thin layer method (HTLM) based wavefield modeling approach |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171208 |
|
| RJ01 | Rejection of invention patent application after publication |