CN106324675A - Broad earthquake wave impedance low-frequency information prediction method and system - Google Patents

Abstract

The invention discloses a broad earthquake wave impedance low-frequency information prediction method, comprising the steps of utilizing broadband overlapped earthquake information and earthquake wavelet to establish complex frequency domain earthquake impedance calculation; utilizing Bayesian inversion theory to calculate initial target function of earthquake wave impedance according to positive calculation of complex frequency domain earthquake impedance; adding initial model constraint to initial target function of earthquake wave impedance to get low-frequency prediction target function and work out the low-frequency prediction target function to get earthquake wave impedance low-frequency information. The method is only utilizes broadband earthquake materials to realize low-frequency prediction of earthquake wave impedance, that is to say, shaft information constraint is not required to measure, and low-frequency information prediction result can still be obtained if complicated geological conditions such as lenticle and strong change of physical property exist. The invention also discloses a broad earthquake wave impedance low-frequency information prediction system, having the above beneficial effect.

Description

A kind of broad-band teleseismic natural impedance low-frequency information Forecasting Methodology and system

Technical field

The present invention relates to oil-gas seismic exploration field, particularly to a kind of broad-band teleseismic natural impedance low-frequency information Forecasting Methodology And system.

Background technology

Question of seismic wave impedance inversion is the main path obtaining underground medium natural impedance.The low cut-off frequency scope of common seismic data Typically at 8-10Hz, lack low frequency component, it is generally required to provide by well logging when utilizing seismic inversion to obtain underground medium natural impedance Material carries out the foundation of elastic parameter initial low frequency model.Initial low frequency model directly affects the reasonability of pre-stack seismic inversion result And reliability.

Present stage, the method for building up of elastic parameter initial low frequency model is mainly with seismic interpretation layer position or sedimentary facies as horizontal stroke To constraint, obtained by low frequency filtering again as after carrying out elastic parameter modeling under longitudinally controlled at well logging and other geologic information Arrive.But, between well, there is lenticular body or physical property changes the complex geological conditions such as violent, particularly exploration phase well information is more In the case of sparse, it be difficult with the method and set up effective model.Therefore, how to overcome conventional low-frequency information Forecasting Methodology at well Between exist under lenticular body or the physical property change complex geological condition such as acutely cannot the deficiency of Accurate Prediction, be those skilled in the art The technical issues that need to address.

Summary of the invention

It is an object of the invention to provide a kind of broad-band teleseismic natural impedance low-frequency information Forecasting Methodology and system, merely with wideband Seismic data i.e. achieves the prediction of seismic impedance low frequency without logging information constrained, there is lenticular body or physical property change between well Acutely wait complex geological condition still to obtain rational low-frequency information to predict the outcome.

For solving above-mentioned technical problem, the present invention provides a kind of broad-band teleseismic natural impedance low-frequency information Forecasting Methodology, including:

Utilize wideband poststack seismic data and seismic wavelet to build complex frequency domain seismic impedance and just calculate son；

Just calculate son according to described complex frequency domain seismic impedance, utilize Bayes's inversion theory to be calculated seismic impedance Initial target function；

In described seismic impedance initial target function, add initial model constraint obtain low-frequency information prediction target letter Number, and solve described low-frequency information prediction object function obtain seismic impedance low-frequency information.

Optionally, wideband poststack seismic data and the seismic wavelet data construct complex frequency domain seismic sequence of input are utilized Resist and just calculating son, including:

Complex frequency domain seismic data in described wideband poststack seismic data is utilized to determine complex frequency domain seismic convolution model；

Time domain seismic data in described seismic wavelet and described wideband poststack seismic data is utilized to determine noiselessness situation Lower time domain earthquake record；

It is calculated complex frequency domain seismic sequence according to described complex frequency domain seismic convolution model and described time domain earthquake record Resist and just calculating son.

Optionally, described complex frequency domain seismic convolution model is

Wherein, ω is frequency, and σ is damping factor, and t is the time, and exp is exponential function, and S (ω-i σ) is complex frequency domain earthquake Data, s (t) is time domain seismic data.

Optionally, described time domain earthquake is recorded as

Wherein, w (t-τ) is seismic wavelet, and r (τ) is reflection coefficient, and τ is integral parameter, and s (t) is time domain earthquake money Material.

Optionally, it is calculated complex frequency domain ground according to described complex frequency domain seismic convolution model and described time domain earthquake record Son is just being calculated in seismic wave impedance, including:

Described time domain earthquake record is brought in described complex frequency domain seismic convolution model and obtainsWherein, ξ=ω-i σ is complex frequency domain；

AbbreviationObtain complex frequency domain geological data matrix S'=F* C'_σ*R；

Consider S'=F*C'_σ* obtain complex frequency domain seismic impedance after the real part of R and imaginary part and just calculate sub-S_ri=F_ri*C'_σ* R；

Wherein, S' is complex frequency domain geological data matrix, and F is complex frequency domain Wavelet Martrix, C'_σDamping factor matrix, R is reflection Coefficient matrix；S_riFor complex frequency domain geological data matrix real part and imaginary part, F_riFor complex frequency domain Wavelet Martrix real part and imaginary part.

Optionally, just calculate son according to described complex frequency domain seismic impedance, utilize Bayes's inversion theory to be calculated ground Seismic wave impedance initial target function, including:

Just calculate son according to described complex frequency domain seismic impedance, utilize Bayes's inversion theory to be calculated seismic impedance Posterior probability density function p (R, σ_n|S_ri) it is:

$p(R,{\mathrm{\σ}}_n|S_{ri})\∝\exp \[-\underset{1}{\overset{n}{\Σ}}ln(1+R^2/{\mathrm{\σ}}_m^2)\]\exp \[\frac{-{(S_{ri}-F_{ri}{C}_{\mathrm{\σ}}^{\′}R)}^T(S_{ri}-F_{ri}{C}_{\mathrm{\σ}}^{\′}R)}{2{{\mathrm{\σ}}_n}^2}\]$

By p (R, σ_n|S_ri) carry out maximization and process and obtain seismic impedance initial target function F and be:

$F={(S_{ri}-F_{ri}{C}_{\mathrm{\σ}}^{\′}R)}^T(S_{ri}-F_{ri}{C}_{\mathrm{\σ}}^{\′}R)+2{\mathrm{\σ}}_n^2\underset{i=1}{\overset{n}{\Σ}}ln(1+R^2/{\mathrm{\σ}}_m^2)$

Wherein,And σ_n ²It is respectively geological data and the variance of model data.

Optionally, described low-frequency information prediction object function is:

$F={(S_{ri}-F_{ri}{C}_{\mathrm{\σ}}^{\′}R)}^T(S_{ri}-F_{ri}{C}_{\mathrm{\σ}}^{\′}R)+2{\mathrm{\σ}}_n^2\underset{i=1}{\overset{n}{\Σ}}ln(1+R^2/{\mathrm{\σ}}_m^2)+\mathrm{\λ}{(\mathrm{\α}-\mathrm{\β}R)}^T(\mathrm{\α}-\mathrm{\β}R)$

Wherein, λ is constraint factor,

The present invention also provides for a kind of broad-band teleseismic natural impedance low-frequency information prognoses system, including:

Complex frequency domain seismic impedance is just calculating sub-structure module, is used for utilizing wideband poststack seismic data and seismic wavelet structure Build complex frequency domain seismic impedance and just calculate son；

Seismic impedance initial target function computation module, for just calculating son according to described complex frequency domain seismic impedance, Bayes's inversion theory is utilized to be calculated seismic impedance initial target function；

Seismic impedance low-frequency information prediction module, initial for adding in described seismic impedance initial target function Model constraint obtain low-frequency information prediction object function, and solve described low-frequency information prediction object function obtain seismic impedance Low-frequency information.

Optionally, described complex frequency domain seismic impedance is just calculating sub-structure module and is including:

Complex frequency domain seismic convolution model unit, is used for utilizing complex frequency domain seismic data in described wideband poststack seismic data true Determine complex frequency domain seismic convolution model；

Time domain seismic recording unit, is used for utilizing time domain in described seismic wavelet and described wideband poststack seismic data Seismic data determines time domain earthquake record in the case of noiselessness；

Complex frequency domain seismic impedance is just calculating subelement, for according to described complex frequency domain seismic convolution model and described time Territory earthquake record is calculated complex frequency domain seismic impedance and is just calculating son.

Optionally, described seismic impedance initial target function computation module includes:

Probability density function unit, for just calculating son according to described complex frequency domain seismic impedance, utilizes Bayes's inverting Theoretical Calculation obtains seismic impedance posterior probability density function p (R, σ_n|S_ri) it is:

Seismic impedance initial target function unit, for by p (R, σ_n|S_ri) carry out maximization process obtain seismic sequence Anti-initial target function F is:

Wherein,And σ_n ²It is respectively geological data and the variance of model data.

A kind of broad-band teleseismic natural impedance low-frequency information Forecasting Methodology provided by the present invention, including: utilize wideband poststack ground Shake data and seismic wavelet build complex frequency domain seismic impedance and are just calculating son；Son is just being calculated, profit according to complex frequency domain seismic impedance It is calculated seismic impedance initial target function with Bayes's inversion theory；Seismic impedance initial target function adds Initial model constraint obtain low-frequency information prediction object function, and solve low-frequency information prediction object function obtain seismic impedance Low-frequency information；

Visible, the method is merely with broad-band teleseismic data, it is achieved that seismic impedance low frequency is predicted, i.e. without well logging information , there is lenticular body in constraint between well or physical property change acutely waits complex geological condition still to obtain the prediction of rational low-frequency information to tie Really；Present invention also offers a kind of broad-band teleseismic natural impedance low-frequency information prognoses system, there is above-mentioned beneficial effect, at this no longer Repeat.

Accompanying drawing explanation

In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing In having technology to describe, the required accompanying drawing used is briefly described, it should be apparent that, the accompanying drawing in describing below is only this Inventive embodiment, for those of ordinary skill in the art, on the premise of not paying creative work, it is also possible to according to The accompanying drawing provided obtains other accompanying drawing.

The flow chart of the broad-band teleseismic natural impedance low-frequency information Forecasting Methodology that Fig. 1 is provided by the embodiment of the present invention；

The underground seismic wave impedance model schematic diagram truly that Fig. 2 is provided by the embodiment of the present invention；

Fig. 3 is inverting initial model schematic diagram in prior art；

The seismic impedance low-frequency information schematic diagram that the method that Fig. 4 is provided by utilizing the embodiment of the present invention obtains；

Fig. 5 is the comparing result schematic diagram of three kinds of situations in the 50th road in extraction Fig. 4；

The structured flowchart of the broad-band teleseismic natural impedance low-frequency information prognoses system that Fig. 6 is provided by the embodiment of the present invention.

Detailed description of the invention

The core of the present invention is to provide a kind of broad-band teleseismic natural impedance low-frequency information Forecasting Methodology and system, merely with wideband Seismic data i.e. achieves the prediction of seismic impedance low frequency without logging information constrained, there is lenticular body or physical property change between well Acutely wait complex geological condition still to obtain rational low-frequency information to predict the outcome.

For making the purpose of the embodiment of the present invention, technical scheme and advantage clearer, below in conjunction with the embodiment of the present invention In accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is The a part of embodiment of the present invention rather than whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art The every other embodiment obtained under not making creative work premise, broadly falls into the scope of protection of the invention.

In embodiment, natural impedance (wave impedance) is seismic wave when propagating in media as well, acts on certain area Pressure and unit interval in perpendicular through the ratio of particle flow (i.e. area takes advantage of Particle Vibration Velocity) of this area, there is resistance Implication, its numerical value product equal to Media density p and velocity of wave V.

The present embodiment selects wideband poststack seismic data to be owing to the available effective low-frequency information of broad-band teleseismic can be extremely 1.5Hz.But the existing method to passing through conventional Inverse modeling after broad-band teleseismic data low-pass filtering more also lacks wideband The consideration of the low cut-off frequency of earthquake following lower frequency component, it is still necessary to needed for relying on other information compensation inverting Whole frequency band elastic parameters Lower frequency information.The most still can not overcome between well, there is lenticular body or physical property change acutely wait under complex geological condition reasonable Prediction low-frequency information.

The methods such as complex frequency domain or Laplace territory full waveform inversion, it is contemplated that the attenuation of seismic wave, can produce frequency spectrum and change Making, the frequency content amplitude making original energy the least is strengthened, so, and ultra-low frequency near time domain zero-frequency originally Amplitude can be had obvious amplitude from almost nil becoming, thus can obtain the lower speed that seismic data time domain is difficult to obtain Frequency component.

Therefore, the present embodiment can utilize complex frequency domain wave impedance inversion to obtain the stable low frequency letter contained in broad-band teleseismic Breath i.e. can directly get parms low frequency component, it is not necessary to relies on needed for other information compensation inverting Whole frequency band elastic parameters more Low frequency information.Realize existing lenticular body or the physical property change complex geological condition such as acutely between well and still obtain rational low frequency Information prediction result.Concrete processing procedure refer to Fig. 1.

Refer to the stream of the broad-band teleseismic natural impedance low-frequency information Forecasting Methodology that Fig. 1, Fig. 1 are provided by the embodiment of the present invention Cheng Tu；The method mainly includes that two steps i.e. S100 complex frequency domain seismic impedance is just calculating the complex frequency domain of sub-structure and S110 and S120 Question of seismic wave impedance inversion.

S100, utilize wideband poststack seismic data and seismic wavelet to build complex frequency domain seismic impedance just to calculate son；

Concrete complex frequency domain seismic impedance is just calculating sub-building process and may include that and utilize in wideband poststack seismic data Complex frequency domain seismic data determines complex frequency domain seismic convolution model；Utilize time domain ground in seismic wavelet and wideband poststack seismic data Shake data determines time domain earthquake record in the case of noiselessness；According to complex frequency domain seismic convolution model and time domain earthquake recording gauge Calculation obtains complex frequency domain seismic impedance and is just calculating son.

Above-mentioned complex frequency domain seismic convolution model is

Wherein, ω is frequency, and σ is damping factor, and t is the time, and exp is exponential function, and S (ω-i σ) is complex frequency domain earthquake Data, s (t) is time domain seismic data.

Above-mentioned time domain earthquake is recorded as

Wherein, w (t-τ) is seismic wavelet, and r (τ) is reflection coefficient, and τ is integral parameter, and s (t) is time domain earthquake money Material.

It is calculated complex frequency domain seismic impedance according to complex frequency domain seismic convolution model and time domain earthquake record just to calculate Son may include that

Formula (2) is brought in formula (1) and obtains Wherein, ξ=ω-i σ is complex frequency domain；

Can obtain after exchange integral orderOrder Obtain η (ξ, τ)=W (ξ) exp (-i ξ τ) (5)；

Equation (5) is brought into (4) obtain

Wherein, order

Wherein, j is jth damping factor, and m, n are respectively Wavelet Martrix and damping matrix line number.Reduced equation (6) is S' =F*C'_σ*R；

After considering real part and imaginary part, just drilling equation and becoming S_ri=F_ri*C'_σ*R；

Wherein, S' is complex frequency domain geological data matrix, and F is complex frequency domain Wavelet Martrix, C'_σDamping factor matrix, R is reflection Coefficient matrix；S_riFor complex frequency domain geological data matrix real part and imaginary part, F_riFor complex frequency domain Wavelet Martrix real part and imaginary part.

S110, just calculate son according to described complex frequency domain seismic impedance, utilize Bayes's inversion theory to be calculated earthquake Natural impedance initial target function；

S120, in described seismic impedance initial target function add initial model constraint obtain low-frequency information prediction mesh Scalar functions, and solve described low-frequency information prediction object function obtain seismic impedance low-frequency information.

In the present embodiment, the detailed process of complex frequency domain question of seismic wave impedance inversion may is that

Just calculate son according to described complex frequency domain seismic impedance, utilize Bayes's inversion theory to be calculated seismic impedance Posterior probability density function p (R, σ_n|S_ri) it is:

$p(R,{\mathrm{\σ}}_n|S_{ri})\∝\exp \[-\underset{1}{\overset{n}{\Σ}}\ln (1+R^2/{\mathrm{\σ}}_m^2)\]\exp \[\frac{-{(S_{ri}-F_{ri}{C}_{\mathrm{\σ}}^{\′}R)}^T(S_{ri}-F_{ri}{C}_{\mathrm{\σ}}^{\′}R)}{2{{\mathrm{\σ}}_n}^2}\]---\left(7\right)$

(7) carry out maximization process and obtain seismic impedance initial target function F and be:

$F={(S_{ri}-F_{ri}{C}_{\mathrm{\σ}}^{\′}R)}^T(S_{ri}-F_{ri}{C}_{\mathrm{\σ}}^{\′}R)+2{\mathrm{\σ}}_n^2\underset{i=1}{\overset{n}{\Σ}}ln(1+R^2/{\mathrm{\σ}}_m^2)---\left(8\right)$

Wherein,And σ_n ²It is respectively geological data and the variance of model data.

Obtaining final low-frequency information prediction object function in (8) middle addition initial model constraint is:

$F={(S_{ri}-F_{ri}{C}_{\mathrm{\σ}}^{\′}R)}^T(S_{ri}-F_{ri}{C}_{\mathrm{\σ}}^{\′}R)+2{\mathrm{\σ}}_n^2\underset{i=1}{\overset{n}{\Σ}}ln(1+R^2/{\mathrm{\σ}}_m^2)+\mathrm{\λ}{(\mathrm{\α}-\mathrm{\β}R)}^T(\mathrm{\α}-\mathrm{\β}R)---\left(9\right)$

Wherein, λ is constraint factor,

It is optimized to solve to equation (9) and i.e. can obtain seismic impedance low-frequency information.

The feasibility of said method is described below by comparison diagram.Fig. 2 is underground seismic wave impedance model truly, and Fig. 3 is Initial model, the natural impedance low-frequency information that the method that Fig. 4 provides for utilizing the present embodiment obtains.Fig. 5 is the 50th road in extraction Fig. 4 Three kinds in the case of comparing result, wherein, 1 curve is true seismic impedance, and 2 curves are inverting primary wave in prior art Impedance i.e. utilizes the result that the initial model in Fig. 3 obtains, and 3 for utilizing predicting the outcome of method that the present embodiment provides.From knot Fruit is it can be seen that the present embodiment can effectively realize the prediction of seismic impedance low-frequency information.

Based on technique scheme, the broad-band teleseismic natural impedance low-frequency information Forecasting Methodology that the embodiment of the present invention carries, only profit Use broad-band teleseismic data, it is achieved that seismic impedance low frequency predict, i.e. without logging information constrained, exist between well lenticular body or Physical property change acutely waits complex geological condition still to obtain rational low-frequency information and predicts the outcome.

The broad-band teleseismic natural impedance low-frequency information prognoses system provided the embodiment of the present invention below is introduced, and hereafter retouches The broad-band teleseismic natural impedance low-frequency information prognoses system stated and above-described broad-band teleseismic natural impedance low-frequency information Forecasting Methodology Can be mutually to should refer to.

Refer to the knot of the broad-band teleseismic natural impedance low-frequency information prognoses system that Fig. 6, Fig. 6 are provided by the embodiment of the present invention Structure block diagram；This system may include that

Complex frequency domain seismic impedance is just calculating sub-structure module 100, is used for utilizing wideband poststack seismic data and earthquake Ripple builds complex frequency domain seismic impedance and is just calculating son；

Seismic impedance initial target function computation module 200, for just calculating according to described complex frequency domain seismic impedance Son, utilizes Bayes's inversion theory to be calculated seismic impedance initial target function；

Seismic impedance low-frequency information prediction module 300, for adding in described seismic impedance initial target function Initial model constraint obtain low-frequency information prediction object function, and solve described low-frequency information prediction object function obtain seismic wave Impedance low-frequency information.

Based on above-described embodiment, described complex frequency domain seismic impedance is just calculating sub-structure module 100 and may include that

Complex frequency domain seismic convolution model unit, is used for utilizing complex frequency domain seismic data in described wideband poststack seismic data true Determine complex frequency domain seismic convolution model；

Time domain seismic recording unit, is used for utilizing time domain in described seismic wavelet and described wideband poststack seismic data Seismic data determines time domain earthquake record in the case of noiselessness；

Complex frequency domain seismic impedance is just calculating subelement, for according to described complex frequency domain seismic convolution model and described time Territory earthquake record is calculated complex frequency domain seismic impedance and is just calculating son.

Based on above-described embodiment, described seismic impedance initial target function computation module 200 may include that

Probability density function unit, for just calculating son according to described complex frequency domain seismic impedance, utilizes Bayes's inverting Theoretical Calculation obtains seismic impedance posterior probability density function p (R, σ_n|S_ri) it is:

Seismic impedance initial target function unit, for by p (R, σ_n|S_ri) carry out maximization process obtain seismic sequence Anti-initial target function F is:

Wherein,And σ_n ²It is respectively geological data and the variance of model data.

In description, each embodiment uses the mode gone forward one by one to describe, and what each embodiment stressed is real with other Executing the difference of example, between each embodiment, identical similar portion sees mutually.For system disclosed in embodiment Speech, owing to it corresponds to the method disclosed in Example, so describe is fairly simple, relevant part sees method part explanation ?.

Above broad-band teleseismic natural impedance low-frequency information Forecasting Methodology provided by the present invention and system are carried out detailed Jie Continue.Principle and the embodiment of the present invention are set forth by specific case used herein, and the explanation of above example is only It is the method and core concept thereof being adapted to assist in and understanding the present invention.It should be pointed out that, for those skilled in the art For, under the premise without departing from the principles of the invention, it is also possible to the present invention is carried out some improvement and modification, these improve and repair Decorations also fall in the protection domain of the claims in the present invention.

Claims (10)

1. a broad-band teleseismic natural impedance low-frequency information Forecasting Methodology, it is characterised in that including:

Utilize wideband poststack seismic data and seismic wavelet to build complex frequency domain seismic impedance and just calculate son；

Just calculate son according to described complex frequency domain seismic impedance, utilize Bayes's inversion theory to be calculated seismic impedance initial Object function；

In described seismic impedance initial target function, add initial model constraint obtain low-frequency information prediction object function, and Solve described low-frequency information prediction object function and obtain seismic impedance low-frequency information.

Broad-band teleseismic natural impedance low-frequency information Forecasting Methodology the most according to claim 1, it is characterised in that utilize input Wideband poststack seismic data and seismic wavelet data construct complex frequency domain seismic impedance are just calculating son, including:

Complex frequency domain seismic data in described wideband poststack seismic data is utilized to determine complex frequency domain seismic convolution model；

When utilizing in the case of in described seismic wavelet and described wideband poststack seismic data, time domain seismic data determines noiselessness Between territory earthquake record；

Just it is being calculated complex frequency domain seismic impedance according to described complex frequency domain seismic convolution model and described time domain earthquake record Calculation.

Broad-band teleseismic natural impedance low-frequency information Forecasting Methodology the most according to claim 2, it is characterised in that described complex frequency domain Seismic convolution model is

Wherein, ω is frequency, and σ is damping factor, and t is the time, and exp is exponential function, and S (ω-i σ) is complex frequency domain seismic data, S (t) is time domain seismic data.

Broad-band teleseismic natural impedance low-frequency information Forecasting Methodology the most according to claim 3, it is characterised in that described time domain Earthquake is recorded as

Wherein, w (t-τ) is seismic wavelet, and r (τ) is reflection coefficient, and τ is integral parameter, and s (t) is time domain seismic data.

Broad-band teleseismic natural impedance low-frequency information Forecasting Methodology the most according to claim 4, it is characterised in that according to described multiple Frequency domain seismic convolution model and described time domain earthquake record are calculated complex frequency domain seismic impedance and are just calculating son, including:

Described time domain earthquake record is brought in described complex frequency domain seismic convolution model and obtainsWherein, ξ=ω-i σ is complex frequency domain；

AbbreviationObtain complex frequency domain geological data matrix S'=F*C'_σ*R；

Consider S'=F*C'_σ* obtain complex frequency domain seismic impedance after the real part of R and imaginary part and just calculate sub-S_ri=F_ri*C'_σ*R；

Wherein, S' is complex frequency domain geological data matrix, and F is complex frequency domain Wavelet Martrix, C'_σDamping factor matrix, R is reflection coefficient Matrix；S_riFor complex frequency domain geological data matrix real part and imaginary part, F_riFor complex frequency domain Wavelet Martrix real part and imaginary part.

Broad-band teleseismic natural impedance low-frequency information Forecasting Methodology the most according to claim 5, it is characterised in that according to described multiple Frequency domain seismic impedance is just calculating son, utilizes Bayes's inversion theory to be calculated seismic impedance initial target function, including:

Just calculate son according to described complex frequency domain seismic impedance, utilize Bayes's inversion theory to be calculated seismic impedance posteriority Probability density function p (R, σ_n|S_ri) it is:

$p(R,{\mathrm{\σ}}_n|S_{ri})\∝\exp \[-\underset{1}{\overset{n}{\Σ}}ln(1+R^2/{\mathrm{\σ}}_m^2)\]\exp \[\frac{-{(S_{ri}-F_{ri}{C}_{\mathrm{\σ}}^{\′}R)}^T(S_{ri}-F_{ri}{C}_{\mathrm{\σ}}^{\′}R)}{2{{\mathrm{\σ}}_n}^2}\]$

By p (R, σ_n|S_ri) carry out maximization and process and obtain seismic impedance initial target function F and be:

$F={(S_{ri}-F_{ri}{C}_{\mathrm{\σ}}^{\′}R)}^T(S_{ri}-F_{ri}{C}_{\mathrm{\σ}}^{\′}R)+2{\mathrm{\σ}}_n^2\underset{i=1}{\overset{n}{\Σ}}ln(1+R^2/{\mathrm{\σ}}_m^2)$

Wherein,WithIt is respectively geological data and the variance of model data.

Broad-band teleseismic natural impedance low-frequency information Forecasting Methodology the most according to claim 6, it is characterised in that described low frequency is believed Breath prediction object function is:

$F={(S_{ri}-F_{ri}{C}_{\mathrm{\σ}}^{\′}R)}^T(S_{ri}-F_{ri}{C}_{\mathrm{\σ}}^{\′}R)+2{\mathrm{\σ}}_n^2\underset{i=1}{\overset{n}{\Σ}}ln(1+R^2/{\mathrm{\σ}}_m^2)+\mathrm{\λ}{(\mathrm{\α}-\mathrm{\β}R)}^T(\mathrm{\α}-\mathrm{\β}R)$

Wherein, λ is constraint factor,

8. a broad-band teleseismic natural impedance low-frequency information prognoses system, it is characterised in that including:

Complex frequency domain seismic impedance is just calculating sub-structure module, is used for utilizing wideband poststack seismic data and seismic wavelet to build multiple Frequency domain seismic impedance is just calculating son；

Seismic impedance initial target function computation module, for just calculating son according to described complex frequency domain seismic impedance, utilizes Bayes's inversion theory is calculated seismic impedance initial target function；

Seismic impedance low-frequency information prediction module, for adding initial model in described seismic impedance initial target function Constraint obtain low-frequency information prediction object function, and solve described low-frequency information prediction object function obtain seismic impedance low frequency Information.

Broad-band teleseismic natural impedance low-frequency information prognoses system the most according to claim 8, it is characterised in that described complex frequency domain Seismic impedance is just calculating sub-structure module and is including:

Complex frequency domain seismic convolution model unit, is used for utilizing complex frequency domain seismic data in described wideband poststack seismic data to determine multiple Frequency domain seismic convolution model；

Time domain seismic recording unit, is used for utilizing time domain earthquake in described seismic wavelet and described wideband poststack seismic data Data determines time domain earthquake record in the case of noiselessness；

Complex frequency domain seismic impedance is just calculating subelement, for according to described complex frequency domain seismic convolution model and described time domain ground Shake record is calculated complex frequency domain seismic impedance and is just calculating son.

Broad-band teleseismic natural impedance low-frequency information prognoses system the most according to claim 9, it is characterised in that described earthquake Natural impedance initial target function computation module includes:

Probability density function unit, for just calculating son according to described complex frequency domain seismic impedance, utilizes Bayes's inversion theory It is calculated seismic impedance posterior probability density function p (R, σ_n|S_ri) it is:

Seismic impedance initial target function unit, for by p (R, σ_n|S_ri) carry out maximization process obtain at the beginning of seismic impedance Beginning, object function F was:

Wherein,WithIt is respectively geological data and the variance of model data.