CN102305940B - Method for extracting fluid factor - Google Patents

Abstract

The invention provides a method for extracting a fluid factor. The method comprises the following steps of: detecting at least one channel seismic data in a work area; setting an initial value of a fractional order number used for fractional order Fourier transform; calculating the fractional order Fourier transform of one channel seismic data in the at least one channel seismic data according to the fractional order number; calculating an analytic signal of the fractional order Fourier transform, and Wigner-Ville distribution with suppressed cross terms of the analytic signal; calculating Radon transform of a band window of the Wigner-Ville distribution with the suppressed cross terms; reconstructing a time frequency distribution map according to the Radon transform of the band window; and extracting the fluid factor from the time frequency distribution map. According to the method for extracting the fluid factor, the fluid factor can be accurately extracted from seismic waves even though the seismic waves have a dispersion effect.

Description

Method for extracting fluid factor

Technical field

The present invention relates to field of seismic exploration.More particularly, relate to a kind of method for extracting fluid factor.

Background technology

In seismic prospecting, the extraction of fluid factor is a very important technology.Accurate fluid factor can carry out detection and Identification to underground mineral reserve more accurately.Current, can come the earthquake data analysis by Time-Frequency Analysis Method, to obtain fluid factor.Time-Frequency Analysis Method can provide the joint distribution information of time domain and frequency field, can more clearly describe the time dependent relation of frequency of signal.

Below, a kind of method for extracting fluid factor of prior art is shown.

For certain given road earthquake data x (t), its Wigner (Eugene Wigner) distribution W _x(t, f) is defined as follows:

$W_x(t,f)={\∫}_{-\∞}^{+\∞}(t+\frac{\mathrm{\τ}}{2})x{(t-\frac{\mathrm{\τ}}{2})}^{*}{e}^{-j2\mathrm{\πf\τ}}\mathrm{d\τ}---\left(1\right)$

W _xThe time frequency plane of (t, f) has the highest resolution, and its shortcoming is that it is a quadratic forms, and the cross term of generation causes the appearance of a large amount of pseudo frequency components, has seriously restricted the extraction of the convection cell factor.Wigner distributes and has positive negative frequency, there is cross term between the positive negative frequency, for eliminating this positive and negative frequency translation item, can remove first negative frequency Wigner distributes, namely first signal x (t) is done Hilbert (Hilbert) conversion, obtain an analytic signal z (t), calculate again the Wigner distribution W of this analytic signal _z(t, f), such distribution is called Wigner-Ville (Eugene Wigner-Willie) and distributes.Here, geological data x (t) is the time series (for example, certain road earthquake data of concentrating of certain road, certain road collection carried out the post-stack seismic data that forms after overlap-add procedure etc.) of a seismologic record.

Do not have positive and negative frequency translation item although Wigner-Ville distributes, between the different frequency composition, even same frequency content is inner, still has cross term.The characteristics of cross term are: when interactional two frequency contents or same frequency content different between the frequency, by the amplitude concussion that is proportional to the distance between two frequency contents or the 2 o'clock frequencies, the perpendicular direction of concussion is in the line direction between the two.

In order to reduce the impact of the pseudo frequency component that these cross terms bring, usually adopt smoothing method to suppress cross term.This method realizes by carrying out low-pass filtering in the ambiguity function territory.Ambiguity function is defined as:

$A_z(w,\mathrm{\τ})={\∫}_{-\∞}^{+\∞}z(t+\frac{\mathrm{\τ}}{2})z{(t-\frac{\mathrm{\τ}}{2})}^{*}{e}^{-j2\mathrm{\πwt}}\mathrm{dt}---\left(2\right)$

Two-dimentional inverse Fourier transform relation between it and Wigner distribute, namely

$W_z(t,f)={\∫}_{-\∞}^{+\∞}{A}_z(w,\mathrm{\τ})e^{-j2\mathrm{\π}(\mathrm{wt}+\mathrm{\τf})}\mathrm{dwd\τ}---\left(3\right)$

With a low-pass filter g (w, τ) to W _z(t, f) carries out in (w, τ) territory smoothly, obtains the time-frequency distributions RID that suppresses with cross term _z(t, f), namely

${\mathrm{RID}}_z(t,f)={\∫}_{-\∞}^{+\∞}g(w,\mathrm{\τ})A_z(w,\mathrm{\τ})e^{-j2\mathrm{\π}(\mathrm{wt}+\mathrm{\τf})}\mathrm{dwd\τ}---\left(4\right)$

Wherein, g (w, τ) is the low-pass filter function of two dimension, and w is Doppler shift, and τ is time lag.On w=0 axle and τ=0 axle, g (w, τ) equals 1, and g (w, τ)=g (w ,-τ) ^*It can be the product of two one dimension lowpass functions.

Directly can do spectral factorization to geological data with this Wigner-Ville distribution with the cross term inhibition and obtain time frequency distribution map ρ _z(t, f) obtains fluid factor.Yet this method only is used for describing the seismic event of non-dispersive situation.If there is dispersion phenomenon to exist, then the seismic wave velocity of different frequency is different, for the stratum same point, although this speed difference is very little, but the linear frequency modulation component may occur.Therefore, the inapplicable situation that has dispersion phenomenon to exist of this method.

Therefore, even need a kind of dispersion situation that in seismic event, exists, also can from seismic event, extract the method for fluid factor.

Summary of the invention

The object of the present invention is to provide a kind of method for extracting fluid factor.

An aspect of of the present present invention provides a kind of method for extracting fluid factor, comprising: (a) detect at least one road earthquake data in the work area; (b) be provided for the initial value of the fractional order α of Fourier Transform of Fractional Order; (c) use fractional order α, calculate the Fourier Transform of Fractional Order of one geological data in described at least one road earthquake data; (d) calculate the analytic signal of Fourier Transform of Fractional Order, and calculate the Wigner-Ville distribution with the cross term inhibition of analytic signal; (e) calculating is with the Radon conversion with window of the Wigner-Ville distribution of cross term inhibition; (f) utilization is with the Radon conversion reconstruct time frequency distribution map of window; (g) extract fluid factor from time frequency distribution map.

On the other hand, come the reconstruct time frequency distribution map by the fractional order α that will turn clockwise with the territory at the Radon conversion place of window.

On the other hand, described method also comprises: (h) use the attribute factor that extracts to make up geologic structure; (i) geologic structure that makes up and the geologic structure of obtaining according to log data are compared, determine whether the geologic structure that makes up and the geologic structure of obtaining according to log data coincide in the predictive error scope; (j) if the geologic structure that makes up is coincide in the predictive error scope with the geologic structure of obtaining according to log data, then determine to obtain optimum fractional order α.

On the other hand, if the geologic structure that makes up is not coincide with the geologic structure of obtaining according to log data, then change the value of fractional order α in the predictive error scope, return step (c).

On the other hand, described method also can comprise: when described at least one road earthquake data during for twice geological data at least, use optimum fractional order α to other road earthquake data execution in step (c)-(g) in described at least one road earthquake data.

On the other hand, the step from time frequency distribution map extraction fluid factor can comprise from the time frequency distribution map extraction frequency attribute factor.

On the other hand, the frequency attribute factor can comprise at least one in peak value instantaneous value frequency, root mean square frequency, the frequency attenuation gradient.

On the other hand, can optimum fractional order α be set according to current known seismic data.

The initial value of fractional order α can be set according to current known seismic data on the other hand.

On the other hand, current known seismic data can be the geologic structure data in log data or contiguous work area.

On the other hand, can obtain described log data from well logging.

On the other hand, can before or after detecting described at least one road earthquake data, form described well logging.

An aspect of of the present present invention provides a kind of geologic structure detection method, comprising: use the fluid factor that extracts according to said method to detect geologic structure.

According to method for extracting fluid factor of the present invention, even in the situation that there is dispersion in seismic event, also can reduce the interference between the different linear frequency modulation components, thereby can realize the extraction to the fluid factor of the seismic event of the situation that has dispersion.Like this, by the fluid factor that extracts, more accurately detection and Identification geologic structure.

Will be in ensuing description part set forth the present invention other aspect and/or advantage, some will be clearly by describing, and perhaps can learn through enforcement of the present invention.

Description of drawings

By the detailed description of carrying out below in conjunction with accompanying drawing, above and other objects of the present invention, characteristics and advantage will become apparent, wherein:

Fig. 1 illustrates method for extracting fluid factor according to an embodiment of the invention;

Fig. 2 illustrate prior art with according to the contrast of method for extracting fluid factor of the present invention.

Embodiment

Now, describe more fully with reference to the accompanying drawings different example embodiment, wherein, some exemplary embodiments are shown in the drawings.

Fig. 1 illustrates method for extracting fluid factor according to an embodiment of the invention.

In step 101, detect at least one road earthquake data in the work area.

In step 102, be provided for carrying out the initial value of the fractional order α of Fourier Transform of Fractional Order, the initial value of fractional order α can be set arbitrarily here.

In step 103, use fractional order α, calculate the Fourier Transform of Fractional Order X of the one geological data x (t) in described at least one road earthquake data _α(u), that is,

$X_{\mathrm{\α}}\left(u\right)=\left\{\begin{array}{cc}\sqrt{\frac{1-j\cot \mathrm{\α}}{2\mathrm{\π}}}{e}^{{\mathrm{ju}}^2\cot \mathrm{\α}/2}{\∫}_{-\∞}^{\∞}x\left(t\right)e^{{\mathrm{jt}}^2\cot \mathrm{\α}/2-\mathrm{jut}\csc }& \mathrm{dt},\mathrm{\α}\≠\mathrm{n\π}\\ X\left(u\right),& \mathrm{\α}=2\mathrm{n\π}\\ X(-u),& \mathrm{\α}=(2n\±1)\mathrm{\π}\end{array}\right.---\left(5\right)$

Wherein, n is positive integer, the general Fourier transformation of X (u) expression x (t), and t represents the time, u represents the territory that obtains after the Fourier Transform of Fractional Order.

In step 104, calculate X _α(u) analytic signal Z _αAnd calculate analytic signal Z (u), _α(u) the Wigner-Ville distribution RID with the cross term inhibition _α(u, v).Should be appreciated that, v represents to calculate another dimension of tieing up outside the u in the two-dimensional field (u, v) that obtains after the Wigner-Ville that suppresses with cross term distributes.

In step 105, calculate the Wigner-Ville distribution RID that suppresses with cross term _αThe Radon with window of (u, v) (La Dun) conversion R _α(u, v), that is,

$R_{\mathrm{\α}}(u,v)={\∫}_{-\∞}^{\∞}w(u^{\′}-u){\mathrm{RID}}_{\mathrm{\α}}(u,v){\mathrm{du}}^{\′}---\left(6\right)$

Wherein, w (u) is window function.

In step 106, utilize the Radon conversion RID with window _α(u, v) comes reconstruct time frequency distribution map ρ _α(t, f).

Specifically, by will be with the Radon conversion R of window _αThe territory (u, v) of (u, v) the fractional order α that turns clockwise obtains time frequency distribution map ρ _α(t, f).Like this, there is following relation in territory (u, v) and territory (t, f):

$\left(\begin{array}{c}u\\ v\end{array}\right)=\left(\begin{array}{cc}\cos \mathrm{\α}& \sin \mathrm{\α}\\ -\sin & \cos \mathrm{\α}\end{array}\right)\left(\begin{array}{c}t\\ f\end{array}\right)---\left(7\right)$

In step 107, extract fluid factor from time frequency distribution map.

Can extract fluid factor from time frequency distribution map in known manner.For example, utilize time frequency distribution map in frequency domain, to extract the frequency attribute factor (for example, peak value instantaneous value frequency, root mean square frequency and/or frequency attenuation gradient etc.).

In step 108, use the fluid factor that extracts to make up geologic structure.

In step 109, the geologic structure that makes up and the geologic structure that obtains according to log data are compared, determine whether the geologic structure that makes up and the geologic structure that obtains according to log data coincide in the predictive error scope.

Can obtain log data by the well logging in the work area.It should be appreciated by those skilled in the art, can be before detecting geological data or form afterwards well logging.Because the geologic structure of obtaining and make up of log data is known, therefore no longer be described in detail.

If the geologic structure that makes up is not coincide in the predictive error scope with the geologic structure that obtains according to log data, then in step 110, change the value of fractional order α, return step 103.

If the geologic structure that makes up is coincide in the predictive error scope with the geologic structure that obtains according to log data, then in step 111, determine to obtain optimum fractional order α.At this moment, can be used as the fluid factor of that road earthquake data extraction of from determining optimum fractional order α process, selecting at the fluid factor that step 107 is extracted.

Subsequently, use optimum fractional order α to the per pass geological data execution in step 103-107 in other road earthquake data in described at least one road earthquake data (if existence), to extract fluid factor from other road earthquake data.In addition, all right further execution in step 108 is to make up geologic structure.

After the per pass geological data extracts fluid factor, can utilize the fluid factor of extraction to detect geologic structure.Owing to existing many known fluid factors that utilize to detect the geologic structure method, therefore no longer describing in detail.

In another embodiment, can the initial value of fractional order α be set or optimum fractional order α directly is set according to current known seismic data (for example, the geologic structure data in log data, contiguous work area etc.).For example, if the log data in the log data in current work area and other work areas close (perhaps the geologic structure in the represented geologic structure of the log data in current work area and other work areas is close) then can directly use the fractional order that obtains at other work areas execution in step 101-111 the initial value of fractional order α to be set or optimum fractional order α directly is set.Again for example, can use the fractional order that uses in contiguous work area the initial value of fractional order α to be set or optimum fractional order α directly is set.

Fig. 1 illustrates one embodiment of the present of invention, in this embodiment fractional order is optimized, thereby reduces substantially interference between the inner and different linear frequency modulation components of each linear frequency modulation component.

Because based on the effect that highlights to the linear frequency modulation component of the time-frequency distributions of fractional order, in another embodiment, can fractional order be optimized, directly use the initial value of fractional order α, come execution in step 101-107 to extract fluid factor at least one road earthquake data in described at least one road earthquake data.In other words, even do not carry out the optimization of fractional order, also can solve technical matters of the present invention.In addition, in this embodiment, when as previously described like that according to current known seismic data the initial value of fractional order α is set the time, can further reduce interference.

Fig. 2 illustrate prior art with according to the contrast of method for extracting fluid factor of the present invention.

Fig. 2 (a) illustrates the time-frequency distributions of the method for extracting fluid factor acquisition of adopting prior art.Can find out from Fig. 2 (a), although the cross term braking measure is arranged, all there is the intersection distracter between two linear frequency modulation components and each linear frequency modulation component inside.

Fig. 2 (b)-2 (f) illustrates and adopts the time-frequency distributions that obtains according to method for extracting fluid factor of the present invention.Can find out from Fig. 2 (b)-2 (f), based on the time-frequency distributions of fractional order, the linear frequency modulation component be highlighted effect, and cross term is had obvious inhibiting effect.When the fractional order at optimum, that is, during α=41 °, originally the time frequency component of two of the phase mutual interference linear frequency modulation components separates fully.

Therefore, according to method for extracting fluid factor of the present invention, the fluid factor that not only can realize the non-dispersive seismic event extracts, even and in the situation that there is dispersion in seismic event, also can reduce the interference between the different linear frequency modulation components, thereby can realize the extraction to the fluid factor of the seismic event of the situation that has dispersion.Like this, by the fluid factor that extracts, more accurately detection and Identification geologic structure.

Although specifically shown with reference to its exemplary embodiment and described the present invention, but it should be appreciated by those skilled in the art, in the situation that does not break away from the spirit and scope of the present invention that claim limits, can carry out various changes on form and the details to it.

Claims (12)

1. method for extracting fluid factor comprises:

(a) at least one road earthquake data in the detection work area;

(b) be provided for the initial value of the fractional order α of Fourier Transform of Fractional Order;

(c) use fractional order α, calculate the Fourier Transform of Fractional Order of one geological data in described at least one road earthquake data;

(d) calculate the analytic signal of Fourier Transform of Fractional Order, and calculate the Wigner-Ville distribution with the cross term inhibition of analytic signal;

(e) calculating is with the Radon conversion with window of the Wigner-Ville distribution of cross term inhibition;

(f) utilization is with the Radon conversion reconstruct time frequency distribution map of window;

(g) extract fluid factor from time frequency distribution map.

2. the method for claim 1, wherein: fractional order α comes the reconstruct time frequency distribution map by turning clockwise with the territory at the Radon conversion place of window.

3. the method for claim 1 also comprises:

(h) use the fluid factor that extracts to make up geologic structure;

(i) geologic structure that makes up and the geologic structure of obtaining according to log data are compared, determine whether the geologic structure that makes up and the geologic structure of obtaining according to log data coincide in the predictive error scope;

(j) if the geologic structure that makes up is coincide in the predictive error scope with the geologic structure of obtaining according to log data, then determine to obtain optimum fractional order α.

4. method as claimed in claim 3 wherein, if the geologic structure that makes up is not coincide with the geologic structure of obtaining according to log data, then changes the value of fractional order α in the predictive error scope, returns step (c).

5. method as claimed in claim 3, also comprise: when described at least one road earthquake data during for twice geological data at least, use optimum fractional order α to other road earthquake data execution in step (c)-(g) in described at least one road earthquake data.

6. the step of the method for claim 1, wherein extracting fluid factor from time frequency distribution map comprises from time frequency distribution map extracts the frequency attribute factor.

7. method as claimed in claim 6, wherein, the frequency attribute factor comprises at least one in peak value instantaneous value frequency, root mean square frequency, the frequency attenuation gradient.

8. the method for claim 1, wherein the initial value of fractional order α is set according to current known seismic data.

9. method as claimed in claim 8, wherein, current known seismic data is the geologic structure data in log data or contiguous work area.

10. method as claimed in claim 3, wherein, described log data is obtained in the well logging in described work area.

11. method as claimed in claim 10 wherein, formed described well logging before or after detecting described at least one road earthquake data.

12. a geologic structure detection method is characterized in that comprising:

According to claim 1, any one-11 extract fluid factor;

Utilize the fluid factor that extracts to detect geologic structure.

Images